The New Snapshot Narrators: Changing Your Visions and Perspectives!

The New Snapshot Narrators: Changing Your Visions and Perspectives! Abstract The emergence of high-performance mobile devices and communication networks has provided people with new ways of producing video content. In the past, people recorded videos using professional devices such as camcorders, but now they use their smartphones, wearable video recorders and drones. Because of this change, there is strong interest among users in Internet of Things environments in the context of how video is captured, edited, shared and interacted with. In particular, trends show a shift from professional-edited video content to user-generated video content, with the increasing popularity of video-sharing social media and the emergence of new video recording devices. However, there have been very few studies that have explored how a snapshot video can be edited. Only a few researchers have studied the effects of video on social interaction among users, and they have failed to consider how video content creation can facilitate social interaction. Therefore, we conducted experimental research to discover the impact of scene format (narrative focus and perspective) of everyday videos on narrative engagement and social interaction. We conducted two studies: (i) single condition effect of narrative focus and narrative perspective, and (ii) mixed condition effect of narrative focus and narrative perspective. The results indicated that the single narrative focus and narrative perspective affects narrative engagement and its four sub-constructs. In addition, the mixed narrative focus and narrative perspective affected narrative engagement, and the effects of interaction between them were determined. According to narrative engagement patterns, the tendencies of social interactions, including various system features, were different. The implications and limitations of the study’s results are discussed in the final section of the article. RESEARCH HIGHLIGHTS The effects of four sub-constructs of narrative engagement on social interactivity in video-sharing social media are presented. Scene format of a video strongly influences the narrative conveyed to others. Effects of narrative focus and narrative perspective on shared video are stronger when the scene format condition of the video is mixed. The effect of narrative engagement on viewers’ social interaction behaviors is explained. The main and interaction effects of scene formats on social interactivity in terms of the four sub-constructs of narrative engagement are verified. 1. INTRODUCTION Advances in technology over the past few years, including high-performance mobile devices and high-bandwidth communication networks, have provided new ways of producing video content (Kirk et al., 2007). In the past, people recorded videos using professional devices such as camcorders, but now they use their smartphones. This means that it is easier than ever before to produce a video. As a result, mobile video traffic on the Internet increased by 50% from 2012 to 2014 (Cisco, 2013). In particular, the amount of mobile video data has increased 16-fold. This massive growth in video traffic accounts for nearly two-thirds of all Internet traffic. Anderson (2010) said that the phenomena surrounding video usage are becoming integrated with social media and transcending traditional media. Because of this change, users in mobile environments are highly interested in how video is captured, edited, shared and engaged with. In particular, recent trends indicate a shift from professional-edited video content to user-generated video content, with the increasing popularity of online video-sharing applications such as Vine, and the emergence of new video recording devices, such as smart glasses and drones. User-generated video content has various properties. These are captured spontaneously, edited, and shared instantly, and are often meaningful in the context of the sharer’s experience. Owing to these properties, user-generated video provides a detailed record of the contextual and situational factors that become the background of the sharer’s experience (Steeples, 2002). Thus, editing after shooting is necessary to communicate the sharer’s experience, even if this is regarded as a cumbersome and inefficient task. In particular, most users have been to special events, such as a concert or sports event, where numerous people in the crowd hold up their mobile phones to record the event (Ojala et al., 2014); they often edit and share videos taken this way. For example, they use the video recording and editing functions of Instagram: users record videos while touching the record button. If they want to pause, they simply take their finger off the screen; if they want to continue recording, they just put their finger back on the screen. All the scenes thus recorded are automatically connected as if they were recorded continuously in one take. In this way, users can record only the scene at a desired moment on the fly. They know that the edited video can help them share their experiences more vividly with viewers than a non-edited video. In addition, video has an innate quality and richness that can never be fully captured via text (Kellogg et al., 1997). This means that the sharer can deliver their personal perspective of the experience to a viewer by video. Thus, unlike cinema, TV or animation, user-generated video content can lead to more impressive narratives in a short running time. Owing to these characteristics of videos, many studies on video usage have recently been conducted in the mobile environment context. O’Hara et al. (2006) verified a range of underlying motivations and values in various contexts in terms of the social practices surrounding video consumption on mobile devices. Puikkonen et al. (2008, 2009) investigated mobile video work in real life and identified video usage patterns. In particular, a video taken of everyday life is referred to as a snapshot video. Lehmuskallio and Sarvas (2008) described snapshot videos as lacking concern for techniques that characterize amateur or professional videography; instead they are rather quick, aesthetically poor exposures of real life. He said that a snapshot video is usually captured within and outside familiar contexts, disregarding professional video recording techniques. While a snapshot video is quite popular among users today, there are very few studies that have explored how a snapshot video is edited. A few researchers have conducted studies on the effect of video on social interaction among users, but they have neglected the methods of producing video content for facilitating social interaction. Some studies have described video as an effective medium for delivering impressions to others. For example, video has been described as a medium for giving one’s impressions about the presence of others (Chung et al., 2015; Daft and Lengel, 1986). Some researchers have argued that the effects of mediums such as a video can vary depending on the relationships between the communications channel and the type of medium (Dennis and Valacich, 1999; Hiltz et al., 2000). Communicating with others on low synchronicity (e.g. e-mail or bulletin board) may be appropriate for conveyance of information, whereas communicating with others on high synchronicity (e.g. face-to-face or video interaction) may be more desirable for convergence on shared meaning (Dennis et al., 2008). However, the way video sharers make videos to encourage further social interaction with others is yet to be investigated in the field of human–computer interaction (Bornoe and Barkhuus, 2010; Peng et al., 2011; Zsombori et al., 2011). In addition, it is difficult to deliver the video sharer’s experience without appropriate editing, because video can provide a richer narrative compared with other media, such as text and photo (Girgensohn et al., 2000; Hua et al., 2004; Lienhart, 1999). Accordingly, a well-edited video in terms of narrative is important to deliver the sharer’s experience and facilitate social interaction among users through the snapshot video. This study focuses on narrative structure, because a snapshot video has a meaningful story that is strongly linked to the sharer’s personal experience. The experience can be structuralized by a narrative in which people present their story (Chatman, 1975; Somers, 1994). Narratives include not only the sharer’s actions and feelings but also reflections about his/her actions and feelings (Bruner and Lucariello, 1989). The main purpose of making the narrative is to convey what has happened or what is happening to them through structuralization to show their experience to others. Therefore, it is essential to edit the sharer’s video based on the principle of narrative structure. There are two important types of video narrative structures in this context: narrative focus and narrative perspective. Narrative focus is what the video sharer, as a narrator, focuses on regarding his/her experience. The ‘focus’ on narrative can be controlled by a highlight effect. A video scene can be focused on the overall scenery, but it can also be focused on highlighted scenery as the central event of experience. Narrative perspective refers to the video sharer’s perspective in each video scene. The ‘perspective’ of the narrative can be controlled by point of view, i.e. videos can have a first-person or third-person point of view. Each video scene can have the same point of view continuously, but can also shift between different points of view. Our aim in this study is to investigate the narrative structure of video for facilitating social interaction among users. In particular, we aim to verify the structural effects of video, including narrative focus and narrative perspective, when delivering the sharer’s experience to a viewer. The effect of video narratives can be explained by the concept of narrative engagement. In our previous study, we verified that the duration of a video can affect the viewer’s narrative engagement. Narrative engagement is defined as the outcome of a convergent process where all mental systems and capacities of the viewer are focused on the sharer’s narrative being played out in the video (Jang et al., 2016). In our previous study, we found that the optimum running time and number of cut scenes to facilitate viewer’s narrative engagement with a snapshot video is 24 s and 6 scenes, respectively, in a mobile environment. However, we could not understand why the viewer tends to engage with the video narrative based on the structural effect of the narrative, nor the tendency viewers to use social features. Therefore, we modified the duration of video stimuli and manipulated the scene format in order to verify the effect of narrative structure in this study. In addition, we focused not only on the detailed differentiation of narrative engagement by the narrative structure of videos but also on the tendency toward social interactive behavior through the socially functional system of our study. For the experiments, we improved our experimental system to provide participants with a more advanced video recording system than in the previous study. In our previous experimental system, we created the video stimuli manually, and the participants used the low-fidelity version of the socially interactive system in the limited mobile environment. In the current experimental system, we developed an automatic synchronous video recording system connected with Google Glass and a drone. Before the main experiment, participants were given access to our video recording system, which they used to share their videos by posting these on our experimental social media. The videos were used to determine what people are usually interested in. From this, we finally decided on the event, story and narrative of video stimuli, and manipulating these to control external factors, such as contents, characters and events. The next section describes our theoretical background and hypotheses in terms of the narrative structure of snapshot videos, including narrative engagement, narrative focus and narrative perspective related to point of view. Section 3 explains the method used in our study. Section 4 shows the results based on the statistical analysis. Section 5 presents a general discussion, and Section 6 explains the limitations and implications of this study. 2. THEORETICAL BACKGROUND 2.1. Narrative structure and engagement According to the narrative paradigm, a narrative is a way of constructing stories (Fisher, 1985). Schank and Berman (2002) defined narrative as the detailed reconstruction of the human experience. In this perspective, McCarthy and Wright (2004) said that it is important to decide what the beginning and end of the story are, and how the experience is structuralized by narrative. In other words, while a story is the event that occurred itself, a narrative is the reconstructed story. These reconstructions provide us with the meaning behind the event (Bruner and Lucariello, 1989; Green and Brock, 2000). The primary goal of structuring a narrative is to communicate a specific message to the audience and to impress it upon them (Neitzel, 2005; Qin et al., 2009). Audiences understand the narrative flow of causal relations among events having a linear time sequence (Abolafia, 2010; Bordwell et al., 1997). Through this understanding, the narrative affects the audience’s beliefs, attitudes and behavioral intentions (Appel and Richter, 2007; Brock et al., 2002; Slater et al., 2006). A well-structured narrative draws the attention of viewers, who feel and imagine that they are in narrative themselves (Green and Brock, 2000). The status of a viewer’s immersion in the narrative is explained by the concept of narrative engagement (Busselle and Bilandzic, 2009). According to Busselle and Bilandzic (2009), narrative engagement can be explained by four sub-constructs: attentional focus, narrative understanding, narrative presence and emotional engagement. Attentional focus is defined as the degree of concentration on a specific target without distraction. Narrative understanding is defined as the degree of understanding of the narrative. Narrative presence is defined as the extent of feeling present in the narrated world. Emotional engagement is defined as the level of sympathy for characters in the narrative. These sub-constructs are based on a mental model created for understanding narratives. People understand a narrative by interpreting its background, characters and situations through their knowledge of the real world (Graesser et al., 2002; Roskos-Ewoldsen et al., 2004). In this process, people understand the situation in the narrative or empathize with the emotions of characters, given their absorption with the characters (Cohen, 2001; Zillmann, 1995). In addition, people feel immersed in the narrative, which is referred to as telepresence (Green and Brock, 2000; Green, 2004). Consequently, the viewer has fun and understands what the narrative is saying through immersion in the narrative (Prensky, 2001). This study clarifies the effects of narrative structure on each sub-construct of narrative engagement. We already know that people in the mobile environment receive the sharer’s experience from a snapshot video through narrative, and then conduct socially interactive behavior depending on their level of narrative engagement (Jang et al., 2016). However, we could not verify how the viewer of the video engages with the sharer’s thoughts, feelings and situations through video narrative. For the purpose of the present study, we assumed that the change in narrative structure caused by the format of scenes in videos affects viewers’ narrative engagement. 2.2. Narrative focus Narrative focus is defined as the structural range of focus on the narrative. Sharers of the narrative can focus on the totality of actions, events, characters and settings, but they can also focus on just one or a combination of these (Kraus, 2006). Stage lighting for drama is a good example of the narrative focus effect. The lighting sometimes lights up the whole stage using floodlights, but can also light a specific area of the stage using a spotlight. These lighting methods direct viewers’ attention to the central event occurring on the whole stage to emphasize important points in the narrative (Wilson, 1994). The narrative focus effect can also be explained by selective focusing. The effect of selective focusing is that the background or foreground of a photo is used to highlight the subject of the photo (Nalder, 2013; Wignall, 2012). As people pay attention to narrative in a snapshot video, their eyes respond to scene changes, including physiological changes in the retina. The retina is made up of rod and cone cells. Cones detect the detailed information of objects. Because they are at the center of the retina, the angle of sight in order to clearly see an object is only 5° (Gould et al., 2007). The areas of high and low acuity in the central vision are called foveal vision and peripheral vision, respectively (Rayner et al., 1981; Webb and Griffin, 2003). In this study, we adopted methods of controlling narrative focus in each video scene through two types of vision: foveal vision and peripheral vision. The video scene for foveal vision has a focusing effect such that it emphasizes a specific area of the sharer’s experience by calculating the degree of central acuity in their eyes. In contrast, the video scene for peripheral vision has high clarity of visibility by high resolution without any other blurred areas in the whole scene. 2.3. Narrative perspective Narrative perspective is explained as the structural sight of delivering a situation to viewers. Brooks and Warren (1959) divided the narrative perspective into four types/parts depending on who the narrator is. The most commonly used criteria is point of view. According to this method of presenting a situation, the narrative perspective can take the form of a first-person point of view, first-person narrator’s point of view, third-person narrator’s point of view, and omniscient point of view (Klatzky, 1998; Stern, 1991; Vogeley and Fink, 2003). A first-person point of view is a narration in which the central character of the story describes the events of a narrative directly. A first-person narrator’s point of view is a narration in which the narrator is present in the story itself and describes the events. A third-person narrator’s point of view is one in which the narrator is outside of the story and describes the events objectively as an observer. Finally, the omniscient point of view offers narration in which the narrator, who is outside of the story, describes not only the events but also the inner thoughts of characters in the story (Galyean, 1995). We assumed that the snapshot videos were taken using wearable glasses or drones. A video scene recorded using wearable glasses is equivalent to a first-person point of view, with a narrator describing his or her story directly. When this point of view is recorded using wearable glasses, we can know that the sharer’s eyes move toward their central experience, which is subjective and emotional. A video scene taken by a drone is equivalent to a third-person narrator’s point of view. When this point of view is recorded by the drone, we can know the sharer’s surrounding environment. In this study, we controlled the narrative perspective in each video scene through two types of perspectives: first-person perspective and third-person perspective. The video scene with the first-person perspective contains images recorded at/from the sharer’s own eye level. On the other hand, the video scene with the third-person perspective has a bird’s eye-view recorded by means of a helicam. 2.4. Active social interaction Social interaction has been described by the concept of interactivity. Many scholars have studied interactivity as a technical property belonging to computer-mediated environments (Heeter, 2000; Sundar, 2004). In general, interaction features have been considered in terms of functionalities, such as frequency (Liu and Shrum, 2002), controllability (Betrancourt, 2005; Coyle and Thorson, 2001), complexity (Heeter, 2000) and so on. However, these have some limitations. First, the functional features of mobile environments may reveal the potential interactivity but not the actual social interaction. The potential interaction between users is facilitated when they engage perceptually or physically with the virtual environment (Bucy, 2004). Second, in interactions between people in face-to-face communications, it is possible for users to interact with others more deeply when a particular feature is encountered in a mobile environment. This means that existing functions of systems, such as ‘Like’ buttons, need to be considered from a cognitive perspective; we also need to interpret social features according to differences in social interaction behavior. Some scholars have studied this as a user’s perceptual experience (Bucy, 2004; McMillan and Hwang, 2002). This perspective describes social interaction as a kind of expression of psychological state by users during the interaction with a medium. In particular, because the nature of an interaction depends on the process, it is important to understand social interaction as a cognitive level of interaction. This is revealed by the effort level of the action (Heeter, 1989) as an active social interaction in which people use social features to interact with others more deeply. While many studies have applied the concept of active social interaction, few have investigated the features of socially active behavior. Moreover, there has been little research on video and system features that affect active social interaction via narrative engagement. Therefore, we defined active social interaction as a socially engaging behavior, and we applied four system features: click a ‘Like’ button, write a ‘Comment’, ‘Recall’ his or her friend, and ‘Share’ the video to others with comments. These were applied by the concept of effort level of action, with ‘Like’ relatively lighter and ‘Give’ relatively deeper in terms of level of social interaction. We applied these features according to relational effort using system features for measuring the amount of effort people actually exert in socially active behavior. 3. RESEARCH HYPOTHESIS Based on the theoretical background and previous studies, we set hypotheses on how narrative focus and narrative perspective affect attentional focus, narrative understanding, narrative presence and emotional engagement, as the detailed concepts of narrative engagement. We also set hypotheses on how the differences between single and mixed conditions affect narrative engagement. 3.1. The effect of narrative focus on narrative engagement and four sub-constructs A video having a low vision effect makes viewers feel discomfort and cognitive dissonance. In particular, the viewer feels the artificial situations owing to the forced attention, and experiences physical discomfort (Riedl and Young, 2010). These feelings and physical discomfort induce cognitive side-effects among viewers, including dizziness, unsteadiness and disorientation, when they are restricted to using foveal vision (Alfano and Michel, 1990). Likewise, in a mobile environment, videos with foveal vision compel viewers to focus on the central event of the sharer’s experience. This can trigger viewers’ lack of attentional focus to the snapshot video because of their lack of freedom to watch what they want. This lack of freedom leads to low interest in the snapshot video; consequently, viewers may give up or shift their attention to other videos on social media. Therefore, Hypothesis 1. All video scenes having peripheral vision have more positive effects on attentional focus than those having only foveal vision. When viewers are compelled to watch a video with their foveal vision, their task performance in relation to the video is low. For instance, viewers’ game performance decreased when foveal vision was 10°, and their shopping performance decreased when foveal vision was 4° (Pelli, 1987). Dolezal’s (1982) study required participants to wear glasses that induced restricted foveal vision in real life environments. The participants were surprised when they suddenly caught sight of the object because of the lack of recognition of the surrounding environment. These results indicate that people can experience lack of recognition and understanding of surrounding environments when experiencing these through foveal vision. Likewise, we assumed that videos having scenes of foveal vision trigger a lack of narrative understanding among viewers. Videos having scenes of peripheral vision can increase viewers’ willingness to focus on what they want to watch (Lamme, 2003). Conversely, videos with scenes of foveal vision can lead to viewers’ lack of narrative understanding of snapshot videos because of the limited information to understand narrative. This means that the limitation of vision leads to a low understanding of snapshot videos; consequently, the viewer may miss the video’s narrative or ask the sharer for clarification on social media. Therefore, Hypothesis 2. All video scenes having peripheral vision have more positive effects on narrative understanding than those having only foveal vision. Vividness is important to provide a sense of presence, which refers to the ability to perceive (Kim, 2015). This is explained as an abundance of stimulation that is provided by a medium to a person (Steuer et al., 1995). Multimedia content, such as photos and videos, provide more vividness than text because of the different amounts of abundant information (Taylor and Thompson, 1982). This can be manipulated by two technical factors: the breadth and depth of stimulation (Biocca, 1992). In particular, the depth of stimulation is the most effective means of increasing stimulation. Depth is the quality of the perception of the stimulation. For example, videos with high resolution can provide greater depth of perception. Likewise, videos with scenes of peripheral vision can provide a greater sense of presence than videos with foveal vision only. Foveal vision induces people to focus on the central event of the sharer’s experience, but it also triggers the poor depth of perception about the surrounding environment of the event. On the other hand, peripheral vision provides visual clarity of the whole scene and enables viewers to perceive surroundings in with great depth of perception. This means that videos having peripheral vision can provide a strong sense of presence; consequently, the viewer feels as though they are involved in the sharer’s experience. Hypothesis 3. All video scenes having peripheral vision have more positive effects on narrative presence than those having only foveal vision. Foveal vision disrupts the processing of emotional information (Calvo and Lang, 2005; De Cesarei et al., 2009; Rigoulot et al., 2012). According to Rigoulot et al. (2012), emotional information is processed automatically, even when it is displayed in low visual conditions. This result indicates that peripheral vision affects human cognition of emotional information in the surrounding environment. The reaction time for the emotional information of peripheral vision was faster than when the information was presented in foveal vision (Rigoulot et al., 2008). This means that people are sensitive to their peripheral vision in order to feel a situation comprehensively. Likewise, videos having scenes of foveal vision can trigger a lack of emotional engagement in the viewer. In particular, the processing of emotional scenes could be influenced by the characteristics of scenes, such as greater luminance, contrast or amount of color saturation (Calvo and Lang, 2005). This means that foveal vision perceives less vivid emotional information than peripheral vision; consequently, the viewer may find it difficult to empathize with the narrative’s emotional flow. Therefore, Hypothesis 4. All video scenes having peripheral vision have more positive effects on emotional engagement than those having only foveal vision. Overall, we assumed that video scenes having peripheral vision can strongly facilitate viewers’ narrative engagement. Hypothesis 5. All video scenes having peripheral vision have more positive effects on narrative engagement than those having only foveal vision. 3.2. The effect of narrative perspective on narrative engagement and four sub-constructs First-person narration has the advantage of expression, whereby it can communicate faithfully represented thoughts to others (Lim and Reeves, 2009). In the game-playing environment, the first-person perspective appears to enhance the immersion effect, and people feel the dynamics and focus on the game’s narrative (Farrar et al., 2006; Tamborini et al., 2001). These feelings of trust and immersion facilitate attention, which is being drawn to the main event in the narrative. Likewise, we assumed that snapshot videos having a first-person perspective provide viewers with more dynamic feelings. First-person video narratives cannot include the sharer as a character in the events in the video; therefore, the viewer directs his/her attention to the sharer’s experience by putting his/her ego at the center of the video narrative. This means that a first-person video attracts viewer’s attention more than that of a third-person video; consequently, the viewer tends to exhibit more attentional focus. Therefore, Hypothesis 6. All video scenes with a first-person perspective have more positive effects on attentional focus than those using a third-person perspective. The degree of consumers’ comprehension of the message of advertisements is influenced by the difference in perspectives on message delivery (Stern, 1991). This results from the amount of information on video. A snapshot video using the first-person perspective gives limited information for viewers to understand the narrative of the video, which leads to a lack of understanding of the video. However, a snapshot video using the third-person gives viewers gives viewers a sufficient amount of information for them to understand the narrative. Likewise, snapshot videos on social media often have various narrative elements, including surrounding events with no artificial editing (Bornoe and Barkhuus, 2010). This is the reason why the perspective of the video affects the amount of information delivered in the narrative. The overall situation of the sharer’s experience is shown by the third-person perspective, which has a richer narrative than the first-person perspective. Consequently, the viewer can obtain increased understanding of the narrative. Therefore, Hypothesis 7. All video scenes having a third-person perspective have more positive effects on narrative understanding than those with a first-person perspective. The first-person element in games, such as shooting games, is more dynamic in engendering feelings of presence (Farrar et al., 2006). This results from the loss of awareness of self and surroundings (Busselle and Bilandzic, 2009). However, the first-person view makes people focus on the game situation. This effect can disturb them by requiring them to construct an alternate world mentally in a short span of time. Thus, the construction of an alternate world is gradually achieved by gamers. Likewise, it can be speculated that first-person snapshots can disturb the feeling of presence because of the impossibility of constructing an alternate world in a short time. In contrast, third-person videos can make people construct their mental model by imagining the sharer’s experience. It has vividness because people receive situational and contextual information to feel presence (Kim, 2015). This means that the feeling of narrative presence requires an understanding of the whole situation (Schuurink and Toet, 2010; Tench, 1989); consequently, the third-person view of videos can provide more narrative presence than the first-person view. Therefore, Hypothesis 8. All videos having the third-person perspective have more positive effects on narrative presence than having the first-person perspective. People can feel empathy toward others’ emotions by means of various factors, including voice, facial expressions, gestures and movement (Castellano et al., 2008). They construct the mental model of another person’s world by grasping whole situations, at which point they become assimilated emotionally (Fischer et al., 2004; Janssen et al., 2014). In particular, vivid information on emotion helps them engage with the emotional narrative and achieve empathy. Likewise, third-person videos can induce more emotional engagement than first-person videos. The third-person provides varied information, including the whole atmosphere of the event, surrounding situation, and the sharer’s behaviors and facial expressions. This means that the third-person in a snapshot video creates an omniscient viewpoint; consequently, viewers can engage with the sharer’s emotions naturally. Therefore, Hypothesis 9. All video scenes having the third-person perspective have more positive effects on emotional engagement than those having the first-person perspective. Despite the effects of the first-person perspective on attentional focus, we assumed that the overall effects of videos having the third-person perspective can more strongly facilitate viewers’ narrative engagement than videos having the first-person perspective. Hypothesis 10. All video scenes having the third-person perspective have more positive effects on narrative engagement than those having the first-person perspective. 3.3. The effect of mixed narrative focus on narrative engagement People usually communicate their willingness to share experiences with others using highlighting (Baudisch et al., 2003; Luo and Tang, 2008). If the sharer imposes his or her ideas continuously onto others, however, they feel restricted because they have no freedom to imagine for understanding narrative. On the other hand, if the sharer gradually highlights his/her ideas to others, they make a cognitive effort to understand the narrative (Kolfschoten, 2011). In this respect, mixed vision of videos has the advantage of engaging viewers more than single vision. Foveal vision has a lack of surrounding information, while peripheral vision has relatively abundant information. Because of these differences in the video narrative, scenes that change between foveal and peripheral visions ‘increase viewers’ cognitive load for engaging with the video narrative. In particular, viewers can grasp the core event of the narrative in foveal vision, and infer the whole narrative of the sharer’s experience in peripheral vision. This natural process has a more positive effect on engaging with the narrative. Therefore, Hypothesis 11. Videos with mixed vision have more positive effects on narrative engagement than having only a single vision. 3.4. The effect of mixed narrative perspective on narrative engagement The point-of-view determines the range of content and the expressions for narrating events and characters (Black et al., 1979). The first-person point of view is commonly used to express personal attitudes and values. This point of view reveals intimate feelings, moods, state of mind, and is instantly interesting (Eliot, 1957; Stern, 1991). The third-person point of view provides information to an audience about the whole situation. This point of view reveals everything, including surrounding information, character’s behavior, moods with facial expressions and various factors of the situation (Kenney, 1988). However, using a single point-of-view has some disadvantages when describing the narrative. The first-person point of view lacks information variety in which the narrator plausibly talks about others as well as he/she does about him/herself; eventually, audiences may become bored by the limited and self-centered point of view. On the other hand, the third-person point of view lacks naturality when delivering the narrator’s real experience. The third-person point of view is unnatural because it frequently shows obvious situations. In this respect, the mixed point-of-view video is more engaging than a single point-of-view video. The viewer can feel the dynamics and perceive the sharer’s personal experiences more vividly through the first-person. In addition, the viewer can feel the whole atmosphere of the video narrative and receive the sharer’s situational experience more obviously through the third-person (Lim and Reeves, 2009). Mixing these in a video enables us to understand the narrative and empathize more with the character’s situation in the video. Therefore, Hypothesis 12. A mixed point-of-view video has more positive effects on narrative engagement than a single point-of-view video. 3.5. The interaction effect of scene format on narrative engagement We can assume that the mixed point-of-view affects narrative engagement more positively when the vision of the video is mixed. As referred to earlier, mixed vision can facilitate viewers’ engagement more than single vision. This effect is caused by the cognitive effort of the viewer in response to the lack of information from the foveal and peripheral vision. Then, if the mixed point-of-view is provided, viewers can better understand the sharer’s situation by bridging the gap between each scene. Thus, we hypothesize that the positive impacts of mixed point-of-view on viewers’ narrative engagement will be stronger when the vision of the video is also mixed. Therefore, Hypothesis 13. The effect of the mixed point-of-view on narrative engagement will strengthen if the vision of the video is mixed. 4. STUDY This study sought to verify how narrative focus and narrative perspective affect the four sub-constructs of narrative engagement, and the tendency toward social interaction in a mobile environment. 4.1. Study design The purpose of this study is to verify the effect of narrative focus as a structural range, and narrative perspective as a structural sight, on narrative engagement. In addition, we wanted to know how narrative engagement affects viewers’ tendency towards social interaction. We believe that these are important in the design of wearable devices and new video-sharing social media. The study is divided into two parts. To examine hypotheses 1–8, we set a 2 × 2, i.e. (narrative focus: foveal vision/peripheral vision) × (narrative perspective: first person/third person), within-subject experimental design as study 1. To examine hypotheses 9 and 10, we set a 2 × 2, i.e. (narrative focus: single vision/mixed vision) × (narrative perspective: single person/mixed person), within-subject experiment design as study 2 (Table 1). TABLE 1. Study design. Study 2 Single vision (F and P) Mixed vision (F–P or P–F) Study 1 Foveal vision (F) Peripheral vision (P) Single person (1 and 3) The first-person (1) 1F-1F-1F-1F-1F-1F 1P-1P-1P-1P-1P-1P 1F-1P-1F-1P-1F-1P The third-person (3) 3F-3F-3F-3F-3F-3F 3P-3P-3P-3P-3P-3P 3P-3F-3P-3F-3P-3F Mixed-person (1–3 or 3-1) 1F-3F-1F-3F-1F-3F 3P-1P-3P-1P-3P-1P 1F-3P-1F-3P-1F-3P 3P-1F-3P-1F-3P-1F Study 2 Single vision (F and P) Mixed vision (F–P or P–F) Study 1 Foveal vision (F) Peripheral vision (P) Single person (1 and 3) The first-person (1) 1F-1F-1F-1F-1F-1F 1P-1P-1P-1P-1P-1P 1F-1P-1F-1P-1F-1P The third-person (3) 3F-3F-3F-3F-3F-3F 3P-3P-3P-3P-3P-3P 3P-3F-3P-3F-3P-3F Mixed-person (1–3 or 3-1) 1F-3F-1F-3F-1F-3F 3P-1P-3P-1P-3P-1P 1F-3P-1F-3P-1F-3P 3P-1F-3P-1F-3P-1F TABLE 1. Study design. Study 2 Single vision (F and P) Mixed vision (F–P or P–F) Study 1 Foveal vision (F) Peripheral vision (P) Single person (1 and 3) The first-person (1) 1F-1F-1F-1F-1F-1F 1P-1P-1P-1P-1P-1P 1F-1P-1F-1P-1F-1P The third-person (3) 3F-3F-3F-3F-3F-3F 3P-3P-3P-3P-3P-3P 3P-3F-3P-3F-3P-3F Mixed-person (1–3 or 3-1) 1F-3F-1F-3F-1F-3F 3P-1P-3P-1P-3P-1P 1F-3P-1F-3P-1F-3P 3P-1F-3P-1F-3P-1F Study 2 Single vision (F and P) Mixed vision (F–P or P–F) Study 1 Foveal vision (F) Peripheral vision (P) Single person (1 and 3) The first-person (1) 1F-1F-1F-1F-1F-1F 1P-1P-1P-1P-1P-1P 1F-1P-1F-1P-1F-1P The third-person (3) 3F-3F-3F-3F-3F-3F 3P-3P-3P-3P-3P-3P 3P-3F-3P-3F-3P-3F Mixed-person (1–3 or 3-1) 1F-3F-1F-3F-1F-3F 3P-1P-3P-1P-3P-1P 1F-3P-1F-3P-1F-3P 3P-1F-3P-1F-3P-1F 4.2. Participants We recruited participants who were active users of video-sharing social media, such as Instagram (usually its video-sharing feature), Vine, Roadmovie and so on (Constine, 2017; Langer, 2014). In particular, those with experience in using video editing features, like scene filters, could apply. We chose our participants based on the criterion of level of experience in using video-sharing social media, such as Instagram, Vine and RoadMovie, which were the most popular video-sharing social media among users at the time of the experiment. We also considered the active use of video-sharing social media as a level of experience. The level of experience with video-sharing social media was set according to the ratio of shares or likes of other videos by the user. This criterion is important when choosing participants for our experiment because this indicates their adaptability to the experimental system. If a participant is not experienced in using video-sharing social media, he/she may report strange feelings when faced with our experimental system. Through these criteria and considerations, we recruited participants from university online communities and Facebook. All potential participants self-reported their usage of video-sharing social media. In general, participants usually shared at least one daily personal video a day and liked at least five videos made by others per day. To confirm their usage, we collected data on their ‘sharing’ and ‘liking’ behaviors from their personal social media accounts for 3 months. Finally, 52 undergraduate students were chosen for our study (i.e. studies 1 and 2). The subjects were between the ages of 20 and 28, with average age of 24.2. There were 27 (43.5%) males and 35 (56.5%) females. They voluntarily participated in the experiment and received monetary compensation equivalent to USD 15. 4.3. Stimuli manipulation The narrative focus was manipulated by ‘bokeh’, which is the aesthetic quality of the blur produced in the out-of-focus parts of an image caused by a lens (Buhler and Wexler, 2002). This effect occurs in parts of the scene that lie outside the depth of field. When we made the video stimuli, we applied the sharer’s eye-tracking data to adjust the effect to their foveal vision. The application of foveal vision can provide the central event of the sharer’s experience. We adopted two scene manipulations as the structural range of narrative focus: foveal vision and peripheral vision. Foveal vision has a low-level depth of field, whereas peripheral vision has a high-level depth of field. The narrative perspective was manipulated by points of view. We adopted two scene manipulations as the structural sight of the narrative perspective: the first-person point of view and third-person point of view. The first-person point of view is a visual representation method with an egocentric reference frame which shows that one’s own multimodal experiential space is centered on one’s own body (Vogeley and Fink, 2003). On the other hand, the third-person point of view is a visual representation method with an allocentric reference frame that has a perspective other than that of the narrator (Aguirre and D’Esposito, 1999). 4.4. Narrative of video stimuli Video stimuli consisted of fourteen separate video narratives. As a side effect, the narrative effect of video stimuli may vary depending on the interest of participants in their content. To avoid side effects due to contents, raw video contents were recorded by following the same storyline. One of our researchers played the role of the heroine, and she rode various amusement park rides over the course of one day. Video narrative subjects include ‘Air Soccer’, ‘Drum Basket’, ‘Flume Ride’, ‘Roller-Coaster’, ‘Gyro Drop’, ‘Swing Tree’, ‘Bungee Drop’, ‘Suspended Swinging Coaster’, ‘Merry-Go-Round’, ‘Dance Parade’, ‘Cups’, ‘Ferris Wheel’, ‘Mini Basket Ball’ and ‘Dodgems’. Among them, ‘Flume Ride’, ‘Gyro Drop’, ‘Suspended Swinging Coaster’ and ‘Ferris Wheel’ were used for the preliminary study. Ten video narratives were selected and used according to our study process, including study 1 and study 2. All videos had a running time of 24 s and 6 cut scenes. All of the video contents contained a narrative structure, including a beginning, middle and end of scene. In total, there were 100 (10 video narratives × 10 video structures) video stimuli selected randomly for each condition per participant (see Appendix) (Table 2). TABLE 2. Narrative contents. TABLE 2. Narrative contents. 4.5. Measurement We already verified the measurement of narrative engagement and its four sub-constructs (i.e. attentional focus, narrative understanding, narrative presence and emotional engagement) in our previous study (Jang et al., 2016). We applied these to the present experiment, and the phrasing of the questions was modified according to the wordings from the preliminary study. In addition, all the questions were checked against the original versions of measurement, including Appel and Richter (2007), Kim and Biocca (1997), and Cohen (2001). To get more validation of measurement, we adopted an additional item in each questionnaire for the four sub-constructs. Finally, we set questionnaires for the four constructs, including three-, four- or five-question items, and measured the responses with a seven-point Likert scale (see Table 3). TABLE 3. Measurement items. Variables Item Measurement References from Original Source Attentional focus AF1 While watching, I was fully concentrated on the narrative. Jang et al. (2016) Busselle and Bilandzic (2009) AF2 While watching, I was fully focused in the narrative. AF3 While watching, I had forgotten about other thoughts. New item for this study Narrative understanding NU1 This video’ narrative was naturally understandable. Jang et al. (2016) Busselle and Bilandzic (2009) NU2 It was difficult to understand the narrative of this video. NU3 I could easily follow the flow of the narrative in this video Appel and Richter (2007) NU4 I had a hard time recognizing the narrative of this video. NU5 While watching, I understood why the narrative of video was proceed. New item for this study Narrative presence NP1 While watching the video, I lost track of time. Jang et al. (2016) Kim and Biocca (1997) NP2 While watching the video, I completely forgot that I was in the middle of the experiment. Busselle and Bilandzic (2009) NP3 While watching the video, I forgot my everyday concerns. New item for this study Emotional engagement EE1 While watching the video, I felt I knew exactly what the sharer was going through emotionally. Jang et al. (2016) Cohen (2001) EE2 I never really shared the emotions of the sharer. EE3 While watching the video, I could feel the emotions the sharer portrayed. Busselle and Bilandzic (2009) EE4 This video affected the emotional changes to me. New item for this study Variables Item Measurement References from Original Source Attentional focus AF1 While watching, I was fully concentrated on the narrative. Jang et al. (2016) Busselle and Bilandzic (2009) AF2 While watching, I was fully focused in the narrative. AF3 While watching, I had forgotten about other thoughts. New item for this study Narrative understanding NU1 This video’ narrative was naturally understandable. Jang et al. (2016) Busselle and Bilandzic (2009) NU2 It was difficult to understand the narrative of this video. NU3 I could easily follow the flow of the narrative in this video Appel and Richter (2007) NU4 I had a hard time recognizing the narrative of this video. NU5 While watching, I understood why the narrative of video was proceed. New item for this study Narrative presence NP1 While watching the video, I lost track of time. Jang et al. (2016) Kim and Biocca (1997) NP2 While watching the video, I completely forgot that I was in the middle of the experiment. Busselle and Bilandzic (2009) NP3 While watching the video, I forgot my everyday concerns. New item for this study Emotional engagement EE1 While watching the video, I felt I knew exactly what the sharer was going through emotionally. Jang et al. (2016) Cohen (2001) EE2 I never really shared the emotions of the sharer. EE3 While watching the video, I could feel the emotions the sharer portrayed. Busselle and Bilandzic (2009) EE4 This video affected the emotional changes to me. New item for this study AF, attentional focus; NU, narrative understanding; NP, narrative presence; EE, emotional engagement. TABLE 3. Measurement items. Variables Item Measurement References from Original Source Attentional focus AF1 While watching, I was fully concentrated on the narrative. Jang et al. (2016) Busselle and Bilandzic (2009) AF2 While watching, I was fully focused in the narrative. AF3 While watching, I had forgotten about other thoughts. New item for this study Narrative understanding NU1 This video’ narrative was naturally understandable. Jang et al. (2016) Busselle and Bilandzic (2009) NU2 It was difficult to understand the narrative of this video. NU3 I could easily follow the flow of the narrative in this video Appel and Richter (2007) NU4 I had a hard time recognizing the narrative of this video. NU5 While watching, I understood why the narrative of video was proceed. New item for this study Narrative presence NP1 While watching the video, I lost track of time. Jang et al. (2016) Kim and Biocca (1997) NP2 While watching the video, I completely forgot that I was in the middle of the experiment. Busselle and Bilandzic (2009) NP3 While watching the video, I forgot my everyday concerns. New item for this study Emotional engagement EE1 While watching the video, I felt I knew exactly what the sharer was going through emotionally. Jang et al. (2016) Cohen (2001) EE2 I never really shared the emotions of the sharer. EE3 While watching the video, I could feel the emotions the sharer portrayed. Busselle and Bilandzic (2009) EE4 This video affected the emotional changes to me. New item for this study Variables Item Measurement References from Original Source Attentional focus AF1 While watching, I was fully concentrated on the narrative. Jang et al. (2016) Busselle and Bilandzic (2009) AF2 While watching, I was fully focused in the narrative. AF3 While watching, I had forgotten about other thoughts. New item for this study Narrative understanding NU1 This video’ narrative was naturally understandable. Jang et al. (2016) Busselle and Bilandzic (2009) NU2 It was difficult to understand the narrative of this video. NU3 I could easily follow the flow of the narrative in this video Appel and Richter (2007) NU4 I had a hard time recognizing the narrative of this video. NU5 While watching, I understood why the narrative of video was proceed. New item for this study Narrative presence NP1 While watching the video, I lost track of time. Jang et al. (2016) Kim and Biocca (1997) NP2 While watching the video, I completely forgot that I was in the middle of the experiment. Busselle and Bilandzic (2009) NP3 While watching the video, I forgot my everyday concerns. New item for this study Emotional engagement EE1 While watching the video, I felt I knew exactly what the sharer was going through emotionally. Jang et al. (2016) Cohen (2001) EE2 I never really shared the emotions of the sharer. EE3 While watching the video, I could feel the emotions the sharer portrayed. Busselle and Bilandzic (2009) EE4 This video affected the emotional changes to me. New item for this study AF, attentional focus; NU, narrative understanding; NP, narrative presence; EE, emotional engagement. We measured the active social interaction by counting actual usage behavior. In order to apply the effort level of action, we multiplied the usage count and the weighted value, in which ‘Like’ is 1.3, ‘Comment’ is 1.6, ‘Recall’ is 1.9, and ‘Give’ is 2.2. The minimum value for active social interaction is 0 if the participant does not use the social features of a video stimulus, whereas the maximum value for active social interaction is 7 if the participant uses all social features. If the participant uses ‘Like’ and ‘Give’, the count is the same, but the value (1, same) of active social interaction is different (1.3 and 2.2). 4.6. Experimental system and procedure Before the experiment, we explained to participants the concept of video recording through Google Glass and a drone, and provided them with samples of Google Glass and drone devices (all devices were developer versions) to aid in with their understanding of our experiment stimuli. In addition, participants were asked to view the heroine in the video stimuli as the sharer, and that she is an acquaintance. To provide an environment similar to an actual situation, we set the experimental system as multi-platform, including mobile phones and Google Glass. Participants could use their own mobile phones and installed the experimental application as a prototype video-sharing social media. They tried out this application to understand how our prototype video-sharing social media operated. In the prototype, we provided various video contents recorded by Google Glass and a drone. After letting participants try out our prototype, we conducted the main experiment in a controlled laboratory. The experimental system was automated and did not require researchers’ intervention. All descriptions of the experimental process were provided by a text-to-speech system. The researchers observed participant behaviors through two-way mirrors (Fig. 1). FIGURE 1. View largeDownload slide Experimental system. FIGURE 1. View largeDownload slide Experimental system. To conduct the survey relevant to the experiment, we applied Typeform (typeform.com) in our experimental system. Typeform is an online survey tool that provides an optimal mobile survey environment, as it is implemented in a mobile environment. Each participant was asked to access the experiment’s URL through our prototype application. We assigned ID numbers to each participant in order to give each one a new set of random combinations of nine video stimuli in the controlled laboratory environment. The environment was set via PC connected to Google Glass for watching video stimuli. All sets of video stimuli were assigned randomly. Participants were informed about the research purpose, description and experimental procedures through the automated text-to-speech system. The full-scale experimental procedure is as follows. In the beginning, participants watched the video stimuli using Google Glass; then, they were allowed to use four social features of the mobile experimental system. Participants had to write down why they did or did not use system features in response to a certain stimulus in our controlled experiment. 4.7. Measurement model We used PASW Statistics 18 to conduct analysis of variance and regression analysis to test the hypotheses. We also used the partial least squares method of structural equation modeling (Smart PLS 2.0) to test the measurement model. Reliability was measured with Cronbach’s alpha and composite reliability, both of which must exceed 0.70 (Fornell and Larcker, 1981). Table 5 indicates that all values exceeded the required minimum of 0.70. Convergent validity was verified when the standardized factor loading of each construct exceeds 0.70, with a t-value >1.96 as well as when the average variance extracted is >0.50 (Arnold and Reynolds, 2003). Table 5 shows that all standardized factor loadings exceed the required minimum of 0.70. In addition, all average variance extracted values exceed the required minimum of 0.50 (Table 4). TABLE 4. Convergent validity and reliability. Construct Items Factor t-value Composition Reliability AVE Cronbach’s α Attentional focus AF1 0.943 51.948 0.928 0.764 0.894 AF2 0.952 51.575 AF3 0.909 44.997 Narrative understanding NU1 0.841 27.385 0.930 0.726 0.906 NU2 0.802 13.535 NU3 0.920 57.546 NU4 0.872 27.948 NU5 0.822 15.215 Narrative presence NP1 0.920 43.334 0.913 0.725 0.871 NP2 0.896 25.023 NP3 0.920 32.171 Emotional engagement EE1 0.917 52.168 0.931 0.771 0.900 EE2 0.860 19.177 EE3 0.919 12.685 EE4 0.812 18.694 Like Lik 0.719 1 1 1 Comment Com 0.761 1 1 1 Recall Rec 0.798 1 1 1 Give Giv 0.780 1 1 1 Construct Items Factor t-value Composition Reliability AVE Cronbach’s α Attentional focus AF1 0.943 51.948 0.928 0.764 0.894 AF2 0.952 51.575 AF3 0.909 44.997 Narrative understanding NU1 0.841 27.385 0.930 0.726 0.906 NU2 0.802 13.535 NU3 0.920 57.546 NU4 0.872 27.948 NU5 0.822 15.215 Narrative presence NP1 0.920 43.334 0.913 0.725 0.871 NP2 0.896 25.023 NP3 0.920 32.171 Emotional engagement EE1 0.917 52.168 0.931 0.771 0.900 EE2 0.860 19.177 EE3 0.919 12.685 EE4 0.812 18.694 Like Lik 0.719 1 1 1 Comment Com 0.761 1 1 1 Recall Rec 0.798 1 1 1 Give Giv 0.780 1 1 1 TABLE 4. Convergent validity and reliability. Construct Items Factor t-value Composition Reliability AVE Cronbach’s α Attentional focus AF1 0.943 51.948 0.928 0.764 0.894 AF2 0.952 51.575 AF3 0.909 44.997 Narrative understanding NU1 0.841 27.385 0.930 0.726 0.906 NU2 0.802 13.535 NU3 0.920 57.546 NU4 0.872 27.948 NU5 0.822 15.215 Narrative presence NP1 0.920 43.334 0.913 0.725 0.871 NP2 0.896 25.023 NP3 0.920 32.171 Emotional engagement EE1 0.917 52.168 0.931 0.771 0.900 EE2 0.860 19.177 EE3 0.919 12.685 EE4 0.812 18.694 Like Lik 0.719 1 1 1 Comment Com 0.761 1 1 1 Recall Rec 0.798 1 1 1 Give Giv 0.780 1 1 1 Construct Items Factor t-value Composition Reliability AVE Cronbach’s α Attentional focus AF1 0.943 51.948 0.928 0.764 0.894 AF2 0.952 51.575 AF3 0.909 44.997 Narrative understanding NU1 0.841 27.385 0.930 0.726 0.906 NU2 0.802 13.535 NU3 0.920 57.546 NU4 0.872 27.948 NU5 0.822 15.215 Narrative presence NP1 0.920 43.334 0.913 0.725 0.871 NP2 0.896 25.023 NP3 0.920 32.171 Emotional engagement EE1 0.917 52.168 0.931 0.771 0.900 EE2 0.860 19.177 EE3 0.919 12.685 EE4 0.812 18.694 Like Lik 0.719 1 1 1 Comment Com 0.761 1 1 1 Recall Rec 0.798 1 1 1 Give Giv 0.780 1 1 1 Discriminant validity was measured with the criterion that the square root of the average variance extracted for each construct should be greater than the corresponding correlation coefficients (Fornell and Larcker, 1981). All square roots of each corresponding average variance extracted exceeded the corresponding correlation coefficients, as shown in Table 5. The results indicate that the factor loadings of the items of each construct show correlation, meaning that the items of each construct explain their corresponding constructs in a statistically significant manner. In addition, the factor loading of each item on its construct was higher than those on the other constructs. This fact indicates that the items being part of their corresponding constructs is statistically supported. TABLE 5. Discriminant validity. Attentional focus Narrative understanding Narrative presence Emotional engagement Like Comment Recall Give Attentional focus 0.874 Narrative understanding 0.522 0.852 Narrative presence 0.578 0.454 0.851 Emotional engagement 0.586 0.513 0.605 0.878 Like 0.412 0.371 0.418 0.479 1 Comment 0.332 0.224 0.324 0.376 0.460 1 Recall 0.351 0.275 0.364 0.423 0.444 0.318 1 Give 0.276 0.279 0.305 0.383 0.453 0.260 0.545 1 Attentional focus Narrative understanding Narrative presence Emotional engagement Like Comment Recall Give Attentional focus 0.874 Narrative understanding 0.522 0.852 Narrative presence 0.578 0.454 0.851 Emotional engagement 0.586 0.513 0.605 0.878 Like 0.412 0.371 0.418 0.479 1 Comment 0.332 0.224 0.324 0.376 0.460 1 Recall 0.351 0.275 0.364 0.423 0.444 0.318 1 Give 0.276 0.279 0.305 0.383 0.453 0.260 0.545 1 The boldface figures on the diagonal are the square root of AVE. TABLE 5. Discriminant validity. Attentional focus Narrative understanding Narrative presence Emotional engagement Like Comment Recall Give Attentional focus 0.874 Narrative understanding 0.522 0.852 Narrative presence 0.578 0.454 0.851 Emotional engagement 0.586 0.513 0.605 0.878 Like 0.412 0.371 0.418 0.479 1 Comment 0.332 0.224 0.324 0.376 0.460 1 Recall 0.351 0.275 0.364 0.423 0.444 0.318 1 Give 0.276 0.279 0.305 0.383 0.453 0.260 0.545 1 Attentional focus Narrative understanding Narrative presence Emotional engagement Like Comment Recall Give Attentional focus 0.874 Narrative understanding 0.522 0.852 Narrative presence 0.578 0.454 0.851 Emotional engagement 0.586 0.513 0.605 0.878 Like 0.412 0.371 0.418 0.479 1 Comment 0.332 0.224 0.324 0.376 0.460 1 Recall 0.351 0.275 0.364 0.423 0.444 0.318 1 Give 0.276 0.279 0.305 0.383 0.453 0.260 0.545 1 The boldface figures on the diagonal are the square root of AVE. Cross-loadings between constructs are shown in Table 6. Discriminant validity was measured with the criterion that the square root of the average variance extracted should be greater than the corresponding correlation coefficients. All square roots of each corresponding average variance extracted exceeded the corresponding correlation coefficients. TABLE 6. Cross-loadings. AF NU NP EE LIK COM REC GIV Attentional focus AF1 0.915 0.545 0.674 0.555 0.413 0.306 0.349 0.287 AF2 0.922 0.517 0.664 0.543 0.411 0.274 0.342 0.267 AF3 0.917 0.438 0.761 0.541 0.352 0.289 0.318 0.23 Narrative understanding NU1 0.515 0.843 0.485 0.494 0.379 0.253 0.304 0.263 NU2 0.352 0.800 0.265 0.354 0.244 0.150 0.143 0.173 NU3 0.528 0.920 0.449 0.476 0.365 0.192 0.269 0.288 NU4 0.413 0.870 0.308 0.380 0.244 0.149 0.237 0.213 NU5 0.386 0.823 0.389 0.457 0.321 0.197 0.193 0.235 Narrative presence NP1 0.768 0.445 0.908 0.596 0.393 0.300 0.371 0.288 NP2 0.652 0.375 0.888 0.497 0.381 0.319 0.308 0.23 NP3 0.726 0.374 0.885 0.509 0.281 0.233 0.285 0.225 Emotional engagement EE1 0.533 0.475 0.526 0.916 0.391 0.339 0.344 0.309 EE2 0.428 0.418 0.424 0.859 0.360 0.287 0.339 0.316 EE3 0.505 0.466 0.498 0.918 0.440 0.342 0.346 0.327 EE4 0.578 0.436 0.657 0.813 0.483 0.345 0.448 0.389 Like LIK 0.412 0.371 0.418 0.479 1 0.460 0.444 0.453 Comment COM 0.332 0.224 0.324 0.376 0.460 1 0.318 0.260 Recall REC 0.351 0.275 0.364 0.423 0.444 0.318 1 0.545 Give GIV 0.276 0.279 0.305 0.383 0.453 0.260 0.545 1 AF NU NP EE LIK COM REC GIV Attentional focus AF1 0.915 0.545 0.674 0.555 0.413 0.306 0.349 0.287 AF2 0.922 0.517 0.664 0.543 0.411 0.274 0.342 0.267 AF3 0.917 0.438 0.761 0.541 0.352 0.289 0.318 0.23 Narrative understanding NU1 0.515 0.843 0.485 0.494 0.379 0.253 0.304 0.263 NU2 0.352 0.800 0.265 0.354 0.244 0.150 0.143 0.173 NU3 0.528 0.920 0.449 0.476 0.365 0.192 0.269 0.288 NU4 0.413 0.870 0.308 0.380 0.244 0.149 0.237 0.213 NU5 0.386 0.823 0.389 0.457 0.321 0.197 0.193 0.235 Narrative presence NP1 0.768 0.445 0.908 0.596 0.393 0.300 0.371 0.288 NP2 0.652 0.375 0.888 0.497 0.381 0.319 0.308 0.23 NP3 0.726 0.374 0.885 0.509 0.281 0.233 0.285 0.225 Emotional engagement EE1 0.533 0.475 0.526 0.916 0.391 0.339 0.344 0.309 EE2 0.428 0.418 0.424 0.859 0.360 0.287 0.339 0.316 EE3 0.505 0.466 0.498 0.918 0.440 0.342 0.346 0.327 EE4 0.578 0.436 0.657 0.813 0.483 0.345 0.448 0.389 Like LIK 0.412 0.371 0.418 0.479 1 0.460 0.444 0.453 Comment COM 0.332 0.224 0.324 0.376 0.460 1 0.318 0.260 Recall REC 0.351 0.275 0.364 0.423 0.444 0.318 1 0.545 Give GIV 0.276 0.279 0.305 0.383 0.453 0.260 0.545 1 AF, attentional focus; NU, narrative understanding; NP, narrative presence; EE, emotional engagement; LIK, like; COM, comment; REC, recall; GIV, give. Bold italic figures indicate statistically significant values. TABLE 6. Cross-loadings. AF NU NP EE LIK COM REC GIV Attentional focus AF1 0.915 0.545 0.674 0.555 0.413 0.306 0.349 0.287 AF2 0.922 0.517 0.664 0.543 0.411 0.274 0.342 0.267 AF3 0.917 0.438 0.761 0.541 0.352 0.289 0.318 0.23 Narrative understanding NU1 0.515 0.843 0.485 0.494 0.379 0.253 0.304 0.263 NU2 0.352 0.800 0.265 0.354 0.244 0.150 0.143 0.173 NU3 0.528 0.920 0.449 0.476 0.365 0.192 0.269 0.288 NU4 0.413 0.870 0.308 0.380 0.244 0.149 0.237 0.213 NU5 0.386 0.823 0.389 0.457 0.321 0.197 0.193 0.235 Narrative presence NP1 0.768 0.445 0.908 0.596 0.393 0.300 0.371 0.288 NP2 0.652 0.375 0.888 0.497 0.381 0.319 0.308 0.23 NP3 0.726 0.374 0.885 0.509 0.281 0.233 0.285 0.225 Emotional engagement EE1 0.533 0.475 0.526 0.916 0.391 0.339 0.344 0.309 EE2 0.428 0.418 0.424 0.859 0.360 0.287 0.339 0.316 EE3 0.505 0.466 0.498 0.918 0.440 0.342 0.346 0.327 EE4 0.578 0.436 0.657 0.813 0.483 0.345 0.448 0.389 Like LIK 0.412 0.371 0.418 0.479 1 0.460 0.444 0.453 Comment COM 0.332 0.224 0.324 0.376 0.460 1 0.318 0.260 Recall REC 0.351 0.275 0.364 0.423 0.444 0.318 1 0.545 Give GIV 0.276 0.279 0.305 0.383 0.453 0.260 0.545 1 AF NU NP EE LIK COM REC GIV Attentional focus AF1 0.915 0.545 0.674 0.555 0.413 0.306 0.349 0.287 AF2 0.922 0.517 0.664 0.543 0.411 0.274 0.342 0.267 AF3 0.917 0.438 0.761 0.541 0.352 0.289 0.318 0.23 Narrative understanding NU1 0.515 0.843 0.485 0.494 0.379 0.253 0.304 0.263 NU2 0.352 0.800 0.265 0.354 0.244 0.150 0.143 0.173 NU3 0.528 0.920 0.449 0.476 0.365 0.192 0.269 0.288 NU4 0.413 0.870 0.308 0.380 0.244 0.149 0.237 0.213 NU5 0.386 0.823 0.389 0.457 0.321 0.197 0.193 0.235 Narrative presence NP1 0.768 0.445 0.908 0.596 0.393 0.300 0.371 0.288 NP2 0.652 0.375 0.888 0.497 0.381 0.319 0.308 0.23 NP3 0.726 0.374 0.885 0.509 0.281 0.233 0.285 0.225 Emotional engagement EE1 0.533 0.475 0.526 0.916 0.391 0.339 0.344 0.309 EE2 0.428 0.418 0.424 0.859 0.360 0.287 0.339 0.316 EE3 0.505 0.466 0.498 0.918 0.440 0.342 0.346 0.327 EE4 0.578 0.436 0.657 0.813 0.483 0.345 0.448 0.389 Like LIK 0.412 0.371 0.418 0.479 1 0.460 0.444 0.453 Comment COM 0.332 0.224 0.324 0.376 0.460 1 0.318 0.260 Recall REC 0.351 0.275 0.364 0.423 0.444 0.318 1 0.545 Give GIV 0.276 0.279 0.305 0.383 0.453 0.260 0.545 1 AF, attentional focus; NU, narrative understanding; NP, narrative presence; EE, emotional engagement; LIK, like; COM, comment; REC, recall; GIV, give. Bold italic figures indicate statistically significant values. In this research, narrative engagement is the second-order construct. The method of analyzing a second-order construct involves including at least three first-order factors that are measured by more than two items for each construct. In addition, all statistically supported values have to exceed each criterion. Narrative engagement has four first-order constructs (attentional focus, narrative understanding, narrative presence and emotional engagement), and each of them has more than two items. We verified the statistical values for measurement items which fulfilled the necessary conditions for analyzing second-order constructs. All items used for generating the second-order construct had to be converted into a single value. For this, we used latent variable scores obtained from partial least squares analysis (Wilson and Henseler, 2007). Additional data on statistical considerations for the second-order construct are shown in Table 7. TABLE 7. Additional data for analyzing NE as the second-order construct. Second-order construct Items First-order construct First-order items t-value Narrative engagement NE1 Attentional focus AF1 24.768 NE2 AF2 22.124 NE3 AF3 20.062 NE4 Narrative understanding NU1 12.396 NE5 NU2 6.361 NE6 NU3 12.096 NE7 NU4 8.030 NE8 NU5 7.902 NE9 Narrative presence NP1 19.480 NE10 NP2 11.789 NE11 NP3 14.089 NE12 Emotional engagement EE1 14.008 NE13 EE2 8.921 NE14 EE3 12.685 NE15 EE4 16.400 Second-order construct Items First-order construct First-order items t-value Narrative engagement NE1 Attentional focus AF1 24.768 NE2 AF2 22.124 NE3 AF3 20.062 NE4 Narrative understanding NU1 12.396 NE5 NU2 6.361 NE6 NU3 12.096 NE7 NU4 8.030 NE8 NU5 7.902 NE9 Narrative presence NP1 19.480 NE10 NP2 11.789 NE11 NP3 14.089 NE12 Emotional engagement EE1 14.008 NE13 EE2 8.921 NE14 EE3 12.685 NE15 EE4 16.400 NE, narrative engagement; AF, attentional focus; NU, narrative understanding; NP, narrative presence; EE, emotional engagement. TABLE 7. Additional data for analyzing NE as the second-order construct. Second-order construct Items First-order construct First-order items t-value Narrative engagement NE1 Attentional focus AF1 24.768 NE2 AF2 22.124 NE3 AF3 20.062 NE4 Narrative understanding NU1 12.396 NE5 NU2 6.361 NE6 NU3 12.096 NE7 NU4 8.030 NE8 NU5 7.902 NE9 Narrative presence NP1 19.480 NE10 NP2 11.789 NE11 NP3 14.089 NE12 Emotional engagement EE1 14.008 NE13 EE2 8.921 NE14 EE3 12.685 NE15 EE4 16.400 Second-order construct Items First-order construct First-order items t-value Narrative engagement NE1 Attentional focus AF1 24.768 NE2 AF2 22.124 NE3 AF3 20.062 NE4 Narrative understanding NU1 12.396 NE5 NU2 6.361 NE6 NU3 12.096 NE7 NU4 8.030 NE8 NU5 7.902 NE9 Narrative presence NP1 19.480 NE10 NP2 11.789 NE11 NP3 14.089 NE12 Emotional engagement EE1 14.008 NE13 EE2 8.921 NE14 EE3 12.685 NE15 EE4 16.400 NE, narrative engagement; AF, attentional focus; NU, narrative understanding; NP, narrative presence; EE, emotional engagement. 4.8. Study 1 First, we conducted a 2 × 2, i.e. (narrative focus: foveal vision (F)/peripheral vision (P)) × (narrative perspective: first person (1)/third person (3)), within-subject experiment (Table 1). The making of mixed conditions, including vision and perspective, were decided by the results of participants’ opinions. We did not know what order of mixed vision would encourage participants to engage with a video narrative without a sense of difference (FP or PF). Mixed-person stimuli faced the same problem as well (13 or 31). Furthermore, the mixed condition between vision and perspective can be made up of four stimuli cases (1F-3P/3F-1P/1P-3F/3P-1F, three times this pattern per video). Therefore, we first verified the single condition effect of narrative focus and perspective on the detailed narrative engagement, including attentional focus, narrative understanding, narrative presence and emotional engagement. 4.8.1. Hypothesis testing To verify the Hypotheses 1–10, we conducted multivariate analysis of variance with attentional focus, narrative understanding, narrative presence and emotional engagement. We also converted four sub-constructs into narrative engagement. Then, we analyzed Hypotheses 5 and 10. Effects of single narrative focus Hypothesis 1 assumed that all video scenes having peripheral vision had more positive effects on attentional focus than those having only foveal vision. The results showed a significant main effect of narrative focus on attentional focus (F(1, 206) = 8.391, P < 0.05). Hypothesis 2 assumed that all video scenes having a peripheral vision had more positive effects on narrative understanding than those having only foveal vision. However, the results did not show a main effect of narrative focus on narrative understanding (F(1, 206) = 1.979, P > 0.05). Hypothesis 3 assumed that all video scenes having a peripheral vision had more positive effects on narrative presence than those having only foveal vision. The results showed a significant main effect of narrative focus on narrative presence (F(1, 206) = 12.659, P < 0.001). Hypothesis 4 assumed that all video scenes having a peripheral vision had more positive effects on emotional engagement than those having only foveal vision. The results showed a significant main effect of narrative focus on emotional engagement (F(1, 206) = 5.462, P < 0.05). Hypothesis 5 assumed that all video scenes having a peripheral vision had more positive effects on narrative engagement than those having only foveal vision. The results showed a significant main effect of narrative focus on narrative engagement (F (1, 206) = 9.783, P < 0.05). Overall, Hypotheses 1, 3 and 4 were supported, but Hypothesis 2 was not supported. Hypothesis 5, as the sum total of the effects of four sub-constructs was supported (Fig. 2). FIGURE 2. View largeDownload slide The effect of single narrative focus on four sub-constructs of narrative engagement. *P < 0.05, **P < 0.001. FIGURE 2. View largeDownload slide The effect of single narrative focus on four sub-constructs of narrative engagement. *P < 0.05, **P < 0.001. Post-hoc for analyzing the tendency of active social interaction by the effect of narrative focus All social features, including ‘Like’, ‘Comment’, ‘Recall’ and ‘Give’, showed statistically significant difference between foveal and peripheral vision: ‘Like’, F(1, 206) = 14.187, P < 0.001; ‘Comment’, F(1, 206) = 5.063, P < 0.05; ‘Recall’, F(1, 206) = 21.508, P < 0.001; and ‘Give’, F(1, 206) = 9.524, P < 0.05. The results showed that the value of active social interaction is almost twice more on the condition of peripheral vision without ‘Comment’. With this result, the cumulative count of social interaction showed statistically significant difference. Peripheral vision facilitated more social behavior (count = 361.7) than foveal vision (count = 201.3) (F(1, 206) = 23.325) (Figs 3 and 4). FIGURE 3. View largeDownload slide The effect of single narrative focus on four features of social interaction in video. *P< 0.05, **P < 0.001. FIGURE 3. View largeDownload slide The effect of single narrative focus on four features of social interaction in video. *P< 0.05, **P < 0.001. FIGURE 4. View largeDownload slide The effect of single narrative focus on the amount of social interaction in video. **P < 0.001. FIGURE 4. View largeDownload slide The effect of single narrative focus on the amount of social interaction in video. **P < 0.001. Effects of single narrative perspective Hypothesis 6 assumed that all video scenes having the first-person perspective had more positive effects on attentional focus than those having the third-person perspective. The results showed a significant main effect of narrative perspective on attentional focus (F(1, 206) = 8.581, P < 0.05) but the hypothesis was not supported. Hypothesis 7 assumed that all video scenes having the third-person perspective had more positive effects on narrative understanding than those having the first-person perspective. The results showed a significant main effect of narrative perspective on narrative understanding (F(1, 206) = 14.751, P < 0.001). Hypothesis 8 assumed that all video scenes having the third-person perspective had more positive effects on narrative presence than those having the first-person perspective. However, the results did not show a main effect of narrative perspective on narrative presence (F(1, 206) = 3.795, P > 0.05). Hypothesis 9 assumed that all video scenes having the third-person perspective had more positive effects on emotional engagement than those having the first-person perspective. However, the results did not show a main effect of narrative perspective on emotional engagement (F(1, 206) = 2.087, P > 0.05). Hypothesis 10 assumed that all video scenes having the third-person perspective had more positive effects on narrative engagement than those having the first-person perspective. The results showed a significant main effect of narrative perspective on narrative engagement (F(1, 206) = 5.502, P < 0.05). Overall, only Hypothesis 7 was supported, while Hypotheses 6, 8, and 9 were not supported. Hypothesis 10, as the sum total of the effects of four sub-constructs, was supported (Fig. 5). FIGURE 5. View largeDownload slide The effect of single narrative perspective on four sub-constructs of narrative engagement. *P < 0.05, **P < 0.001. FIGURE 5. View largeDownload slide The effect of single narrative perspective on four sub-constructs of narrative engagement. *P < 0.05, **P < 0.001. Post-hoc for analyzing the tendency of active social interaction by the effect of narrative perspective All social features, including ‘Like’, ‘Comment’, ‘Recall’ and ‘Give’, did not show statistically significant difference: ‘Like’, F(1, 206) = 0.709, P > 0.05; ‘Comment’, F (1, 206) = 2.818, P > 0.05; ‘Recall’, F (1, 206) = 0.347, P > 0.05; and ‘Give’, F (1, 206) = 0.823, P > 0.05. The results showed that the value of active social interaction is less changed by the different perspective conditions of video stimuli. With this result, the cumulative count of social interaction showed no statistically significant difference. The social behavior counts for the third-person (count = 286.2) was almost the same as that for the first-person (count = 272.8) (F(1, 206) = 0.072) (Figs 6 and 7). FIGURE 6. View largeDownload slide The effect of single narrative perspective on four features of social interaction in video. FIGURE 6. View largeDownload slide The effect of single narrative perspective on four features of social interaction in video. FIGURE 7. View largeDownload slide The effect of single narrative perspective on the amount of social interaction in video. FIGURE 7. View largeDownload slide The effect of single narrative perspective on the amount of social interaction in video. The results of Hypotheses 1–10 testing are shown in Table 8 as the detailed data. TABLE 8. Results of hypothesis testing. IV DV SS dF MS F η2 Sig. Result NF AF 9.233 1 9.233 8.391 .039 .004 H1 Supported NU 2.089 1 2.089 1.97 .010 .161 H2 Not supported NP 11.614 1 11.614 12.659 .058 .000 H3 Supported EE 4.970 1 4.970 5.462 .026 .020 H4 Supported NE 9.398 1 9.398 9.783 .047 .002 H5 Supported NPR AF 9.433 1 9.433 8.581 .040 .004 H6 Not supported NU 14.667 1 14.667 14.751 .067 .000 H7 Supported NP 3.629 1 3.629 3.795 .018 .053 H8 Not supported EE 1.930 1 1.930 2.087 .010 .150 H9 Not supported NE 5.201 1 5.201 5.502 .026 .020 H10 Supported IV DV SS dF MS F η2 Sig. Result NF AF 9.233 1 9.233 8.391 .039 .004 H1 Supported NU 2.089 1 2.089 1.97 .010 .161 H2 Not supported NP 11.614 1 11.614 12.659 .058 .000 H3 Supported EE 4.970 1 4.970 5.462 .026 .020 H4 Supported NE 9.398 1 9.398 9.783 .047 .002 H5 Supported NPR AF 9.433 1 9.433 8.581 .040 .004 H6 Not supported NU 14.667 1 14.667 14.751 .067 .000 H7 Supported NP 3.629 1 3.629 3.795 .018 .053 H8 Not supported EE 1.930 1 1.930 2.087 .010 .150 H9 Not supported NE 5.201 1 5.201 5.502 .026 .020 H10 Supported IV, independent variable; DV, dependent variable; AF, attentional focus; NU, narrative understanding; NP, narrative presence; EE, emotional engagement; NE, narrative engagement; SS, sum of squares; dF, degree of freedom; MS, mean square; F, F-ratio; Sig., significance; η2, eta squared; B, unstandardized coefficients; β, beta coefficient; VIF, variance inflation factor. Bold figures indicate statistically significant values. TABLE 8. Results of hypothesis testing. IV DV SS dF MS F η2 Sig. Result NF AF 9.233 1 9.233 8.391 .039 .004 H1 Supported NU 2.089 1 2.089 1.97 .010 .161 H2 Not supported NP 11.614 1 11.614 12.659 .058 .000 H3 Supported EE 4.970 1 4.970 5.462 .026 .020 H4 Supported NE 9.398 1 9.398 9.783 .047 .002 H5 Supported NPR AF 9.433 1 9.433 8.581 .040 .004 H6 Not supported NU 14.667 1 14.667 14.751 .067 .000 H7 Supported NP 3.629 1 3.629 3.795 .018 .053 H8 Not supported EE 1.930 1 1.930 2.087 .010 .150 H9 Not supported NE 5.201 1 5.201 5.502 .026 .020 H10 Supported IV DV SS dF MS F η2 Sig. Result NF AF 9.233 1 9.233 8.391 .039 .004 H1 Supported NU 2.089 1 2.089 1.97 .010 .161 H2 Not supported NP 11.614 1 11.614 12.659 .058 .000 H3 Supported EE 4.970 1 4.970 5.462 .026 .020 H4 Supported NE 9.398 1 9.398 9.783 .047 .002 H5 Supported NPR AF 9.433 1 9.433 8.581 .040 .004 H6 Not supported NU 14.667 1 14.667 14.751 .067 .000 H7 Supported NP 3.629 1 3.629 3.795 .018 .053 H8 Not supported EE 1.930 1 1.930 2.087 .010 .150 H9 Not supported NE 5.201 1 5.201 5.502 .026 .020 H10 Supported IV, independent variable; DV, dependent variable; AF, attentional focus; NU, narrative understanding; NP, narrative presence; EE, emotional engagement; NE, narrative engagement; SS, sum of squares; dF, degree of freedom; MS, mean square; F, F-ratio; Sig., significance; η2, eta squared; B, unstandardized coefficients; β, beta coefficient; VIF, variance inflation factor. Bold figures indicate statistically significant values. 4.9. Study 2 In study 2, we conducted a 2 × 2, i.e. (narrative focus: single vision (F or P)/mixed vision (F–P or P–F)) × (narrative perspective: single person (1 or 3)/mixed-person (1–3 or 3-1)), within-subject experiment (Table 1). To verify Hypotheses 11 and 12, we provided mixed conditions of narrative focus and perspective to our participants. The method for making the mixed condition, including vision and perspective, was decided by the results of study 1 and participants’ opinions. Participants reported that the scene with 1F condition felt more natural than 1P, and the scene with 3P felt more natural than 3F. Therefore, we set the criteria for making the mixed condition video stimuli: the first-person combined with foveal vision, the third-person combined with peripheral vision, and the first video scene started through the natural combination (1F or 3P). According to these criteria, we set the mixed condition video stimuli as 1F-3F-1F-3F-1F-3F, 3P-1P-3P-1P-3P-1P, 1F-1P-1F-1P-1F-1P and 3P-3F-3P-3F-3P-3F. In addition, we made two double mixed condition (mixed vision × mixed-person) video stimuli, including 1F-3P (repeated three times) and 3P-1F (repeated three times). Therefore, we finally set the six mixed condition videos, including two double mixed conditions. 4.9.1. Single vs. Mixed vision and Single vs. Mixed point-of-view effect on narrative engagement To verify Hypotheses 11 and 12, we conducted a planned contrast analysis with narrative engagement. Hypothesis 11 assumed that mixed vision videos having foveal and peripheral conditions had more positive effects on narrative engagement than those having only single vision. The results showed a statistically significant effect of narrative focus on narrative engagement (t = −4.091, df = 516, P = 0.000). Hypothesis 12 assumed that mixed point-of-view videos (having the first and third-person perspectives) had more positive effects on narrative engagement than those having only a single perspective. The results showed a statistically significant effect of narrative perspective on narrative engagement (t = −3.701, df = 516, P = 0.000). To test Hypothesis 13 as an interaction effect, we conducted two-way analysis of variance. The interaction effect between two mixed conditions on narrative engagement was significant (F(1, 206) = 8.137, P < 0.05). This means that the effect of the mixed point-of-view on narrative engagement will strengthen as the vision of the video is mixed (Figure 8). FIGURE 8. View largeDownload slide The interaction effect between two mixed conditions (mixed-point of view × mixed vision) on narrative engagement. FIGURE 8. View largeDownload slide The interaction effect between two mixed conditions (mixed-point of view × mixed vision) on narrative engagement. 4.9.2. Post-hoc for analyzing the tendency of active social interaction between single and mixed vision or single and mixed point-of-view All social features, including ‘Like’, ‘Comment’, ‘Recall’ and ‘Give’ showed statistically significant difference between single and mixed vision: ‘Like’, F(1, 206) = 0.709, P > 0.05, ‘Comment’, F(1, 206) = 2.818, P > 0.05, ‘Recall’, F(1, 206) = 0.347, P > 0.05, and ‘Give’, F(1, 206) = 0.823, P > 0.05. The results showed that the active social interaction values changed less on the different perspective conditions of video stimuli. With this result, the cumulative count of social interaction showed no statistically significant difference. The social behavior count in the third-person (count = 286.2) was almost the same as that in the first-person (count = 272.8) (F(1, 206) = 0.072). 5. CONCLUSION AND DISCUSSION The primary goal of this study is to verify how the scene formats in snapshot videos affect the four detailed narrative engagement factors: attentional focus, narrative understanding, narrative presence and emotional engagement. In addition, we assumed that the method of creating narrative structure through scene formats is important to facilitate viewers’ active social interaction. To verify this theoretically and practically, scene formats related to narrative structure were considered not only by narrative theory, including the concepts of narrative focus and narrative perspective, but also through the practical possibility of technical realization, including ‘bokeh’ effects and perspective taking. Using Google Glass and a drone, we took snapshot videos inspired by our scenario of a person’s daily experiences. We edited fourteen narratives for our experiment using snapshot videos. In terms of narrative focus, the results indicated that peripheral vision leads to more attentional focus, narrative presence and emotional engagement than foveal vision. Narrative understanding was not differentiated between peripheral and foveal vision. In addition, we found that peripheral vision effects on attentional focus, narrative presence and emotional engagement facilitated more active social interaction among viewers than foveal vision. Furthermore, we found that mixed vision videos lead to more narrative engagement than single vision videos. In terms of narrative perspective, the results indicated that the third-person perspective leads to more attentional focus and narrative understanding than the first-person perspective. Narrative presence and emotional engagement were not differentiated between the first- and third-person perspectives. In addition, we found that differences in perspective between the first and third person did not occur during active social interaction. Meanwhile, we found that a mixed-person videos lead to more narrative engagement than a single-person video. Based on the study results, three interesting issues are worth discussing in more detail. Firstly, as a narrative structure, the effect of narrative focus fulfills the three conditions of narrative engagement. First, peripheral vision facilitates more attentional engagement (i.e. attentional focus) than foveal vision. It is important to note that videos having vivid scenes can create the environment for attentional focus. Second, viewers feel more narrative presence with peripheral vision than foveal vision. This means that delivering the sharer’s experience by video requires the vivid expression of scenes to provide dynamics to viewers. Finally, emotional engagement is triggered more by peripheral vision than foveal vision. Contrary to foveal vision, which concentrates on the central information of the event, peripheral vision has emotional information for triggering emotional engagement. It is important to note that these structural effects of detailed narrative engagement affect active social interaction. Secondly, as a narrative structure, the effect of narrative perspective fulfills the two conditions of narrative engagement. First, the third-person perspective facilitates more attentional engagement (i.e. attentional focus) than the first-person perspective. It is important to note that videos showing the whole situation encourage viewers to engage with the sharer’s narrative more effectively. Second, viewers understand the narrative better in the third person rather than the first person. This means that viewers can understand the sharer’s experience better through the third person than the first person. In addition, it is important to note that these structural effects on detailed narrative engagement affect active social interaction. Thirdly, viewers try to interact with others while engaging with the sharers’ narrative through the narrative structure of snapshot videos. In particular, the leverage of narrative presence and emotional engagement makes a difference in active social interaction. This result implies that viewers communicate with others, using snapshot videos as the medium of active social interaction, beyond the simplified interaction of clicking ‘Like’ or writing a ‘Comment’. 6. LIMITATIONS AND IMPLICATIONS 6.1. Limitations and further study In our study, part of the experimental system was showing video stimuli with Google Glass. This can provide participants with real experience in using new video-sharing social media with wearable devices. However, our experimental system, including the effects of narrative focus and narrative perspective, felt unnatural to participants. In order to aid in participants’ adaptation, we gave them sufficient time to become familiar with Google Glass. However, they felt somewhat dizzy when watching our video stimuli with Google Glass. Furthermore, aside from wearing Google Glass to watch our video stimuli, they also participated in our survey and used social features step by step. These processes may increase the cognitive load of participants and cause confusion. Therefore, future works should adopt a more natural experimental system. Nevertheless, this study has various theoretical and practical implications. 6.2. Theoretical implications Theoretically, this study verified the effects of four detailed concepts of narrative engagement (attentional focus, narrative understanding, narrative presence and emotional engagement) through empirical data. In particular, we examined how social interaction actively depended on the tendency of detailed engaging effects. Our prior study provided a theoretical framework on narrative engagement in terms of the temporal format of snapshot videos. In addition, we already verified the effects of narrative engagement on social interaction. However, our prior study had limitations in explaining how people try to interact actively with others in terms of the quality of socially interactive behavior. Accordingly, we extended our previous study by using system features, applying the concept of effort level of action in which ‘Like’ is relatively lighter and ‘Give’ is relatively heavier in terms of level of social interaction. In particular, we set concepts of narrative structure by summarizing past studies on narratives and related video effects. From the perspective of human–computer interaction research, two different effects related to narrative structure can be described, as mentioned earlier. From the perspective of narrative structure, we logically understood and applied 2D constructs to explain the structural narrative effects of the sharer’s experience: structural range and structural sight of the narrative. This result indicates theoretically that people are affected by changes in narrative structure during active social interaction. 6.3. Practical implications Practically, this research points to the effects of two structural changes, including narrative focus and narrative perspective, on active social interaction. First, in order to increase narrative engagement, we believe that the peripheral focusing effect should be considered. We know that creating this scene structure has a high impact on attentional focus, narrative presence and emotional engagement. This means that designers and developers who want to create an immersive video-sharing social media application or platform can focus on making an automated scene filter to facilitate users’ attention, realistic feelings and emotional assimilation. For example, designers can provide a sharpened filter for making snapshot videos more vivid to viewers. Instagram has provided similar features to make scenes clearer using the ‘sharpen’ filter. In addition, in order to increase narrative engagement, we believe that the third-person point of view should be considered. We know that creating this scene structure has a high impact on attentional focus and narrative understanding. This means that designers and developers who want to make an immersive video-sharing social medium can focus on the use of recording devices to facilitate users’ attention to, and understanding of, narratives. For example, developers can consider a wearable drone to help sharers record and share selfie snapshot videos from the third-person point of view. Nixie, a wearable drone device, can take photos or videos, and then flies back to users automatically. It provides a new perspective in which viewers can pay attention to what the sharer does. For this reason, we can assume that social media like Facebook should consider a third-person point of view. Because the dynamics of timelines are too fast to focus on and understand a specific video, it is necessary to attract viewers’ interest without the content of the video narrative. In addition, this research points to the effects of mixed structural changes on narrative engagement. In order to increase narrative engagement, we know that the mixed condition of narrative focus should be considered. Contrary to the single condition of narrative focus, the mixed condition provides viewers with room to imagine the sharer’s situation for themselves. Therefore, developers who want to make a video editing application can consider automatic mixed scene filters that provide scene effects automatically in order to edit snapshot videos more attractively. For example, we can create the effect of foveal vision by using eye-tracking data that can provide the sharer’s focusing spot. If the sharer’s smartglass is capable of performing eye-tracking, the mixed scene vision of a snapshot video can be edited automatically using the sharer’s eye-tracking data to apply the scene filter. For this reason, we suggest that social media like Snapchat consider the vision effects when shared video is recorded in a short time. If Spectacles, the smartglasses launched by Snapchat, is capable of collecting eye-tracking data, it will be an effective way of editing video scenes automatically by adopting the concept of narrative focus. Furthermore, we know that the mixed condition of narrative perspective should be considered. Contrary to the single condition, the mixed condition provides viewers with room to interpret the sharer’s situation for themselves. Therefore, developers who want to make video recording devices for the Internet of Things environment can consider the automatic mixed scene perspective to provide the scene perspective automatically by collaborative video-taking environment. For example, we can take the third-person point of view by using others’ video data. A location-based cloud video-sharing platform such as Vyclone can be provided to make the sharer’s own narrative by using one’s own and others’ video data simultaneously, and the mixed scene perspective of the snapshot video can be edited automatically. Finally, we know that new social features are needed for interacting with others more deeply. People are already used to clicking the ‘Like’ button to express their empathy. However, viewers may want to interact more deeply and directly with others because of their desire to share their own dramatic experiences as a result of the sharer’s snapshot video. Therefore, designers should consider deeper expressive social features for video-sharing social media. For example, we can reply to the timeline of shared snapshot videos directly and then share it with others. In addition, the viewer can immediately talk about the sharer’s snapshot video with others without making any posts. ACKNOWLEDGEMENTS The authors thank Minji Kim who kindly proofread the entire draft and all of our research team members. Be a good storyteller with wearable devices and drones! FUNDING This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (NRF-2016R1D1A1B02015987). REFERENCES Abolafia , M.Y. ( 2010 ) Narrative construction as sensemaking: how a central bank thinks . Organ. Stud. , 31 , 349 – 367 . Google Scholar CrossRef Search ADS Aguirre , , G.K. and D'Esposito , M. ( 1999 ) Topographical disorientation: a synthesis and taxonomy . Brain , 122 , 1613 – 1628 . Google Scholar CrossRef Search ADS PubMed Alfano , P.L. and Michel , G.F. ( 1990 ) Restricting the field of view: perceptual and performance effects . Percept. Motor Skills , 70 , 35 – 45 . Google Scholar CrossRef Search ADS Anderson , J. ( 2010 ) Understanding the Changing Needs of the US Online Consumer, 2010. Forrester Research. Appel , M. and Richter , T. ( 2007 ) Persuasive effects of fictional narratives increase over time . Media Psychol. , 10 , 113 – 134 . Arnold , M.J. and Reynolds , K.E. ( 2003 ) Hedonic shopping motivations . J. Retail. , 79 , 77 – 95 . Google Scholar CrossRef Search ADS Baudisch , P. , DeCarlo , D. , Duchowski , A.T. and Geisler , W.S. ( 2003 ) Focusing on the essential: considering attention in display design . Commun. ACM , 46 , 60 – 66 . Google Scholar CrossRef Search ADS Betrancourt , M. ( 2005 ) The animation and interactivity principles in multimedia learning. In Mayer , R. (ed.) , The Cambridge Handbook of Multimedia Learning . pp. 287 – 296 . Cambridge University Press , Cambridge . Google Scholar CrossRef Search ADS Biocca , F. ( 1992 ) Communication within virtual reality: creating a space for research . J. Commun. , 42 , 5 – 22 . Google Scholar CrossRef Search ADS Black , J.B. , Turner , T.J. and Bower , G.H. ( 1979 ) Point of view in narrative comprehension, memory, and production . J. Verb. Learn. Verb. Behav. , 18 , 187 – 198 . Google Scholar CrossRef Search ADS Bordwell , D. , Thompson , K. and Ashton , J. ( 1997 ) Film Art: An Introduction . McGraw-Hill , New York . Bornoe , N. and Barkhuus , L. ( 2010 , April) Video microblogging: Your 12 seconds of fame. In CHI'10 Extended Abstracts on Human Factors in Computing Systems. ACM, pp. 3325–3330. Brock , T.C. , Strange , J.J. and Green , M.C. ( 2002 ) Power beyond reckoning. In Green , M. C. , Strange , J. J. and Brock , T. C. (eds) , Narrative impact: Social and Cognitive Foundations . pp. 1 – 16 . Lawrence Erlbaum , Mahwah, NJ . Brooks , C. and Warren , R.P. ( 1959 ) Understanding Fiction , Vol. 5 , Appleton-Century-Crofts , New York . Bruner , J. and Lucariello , J. ( 1989 ) Monologue as narrative recreation of the world, in Narratives from the Crib. pp. 73–97. Bucy , E.P. ( 2004 ) Second generation net news: interactivity and information accessibility in the online environment . Int. J. Media Manag. , 6 , 102 – 113 . Google Scholar CrossRef Search ADS Buhler , J. and Wexler , D. , 2002 . A phenomenological model for bokeh rendering. In ACM SIGGRAPH 2002 Conference Abstracts and Applications. ACM, pp. 142 – 142 . Busselle , R. and Bilandzic , H. ( 2009 ) Measuring narrative engagement . Media Psychol. , 12 , 321 – 347 . Google Scholar CrossRef Search ADS Calvo , M.G. and Lang , P.J. ( 2005 ) Parafoveal semantic processing of emotional visual scenes . J. Exp. Psychol. , 31 , 502 . Castellano , G. , Kessous , L. and Caridakis , G. ( 2008 ) Emotion recognition through multiple modalities: face, body gesture, speech. In Peter , C. and Beale , R. (eds) , Affect and Emotion in Human-Computer Interaction . pp. 92 – 103 . Springer , Berlin, Heidelberg . Google Scholar CrossRef Search ADS Chatman , S. ( 1975 ) Towards a theory of narrative . New Literary Hist. , 6 , 295 – 318 . Google Scholar CrossRef Search ADS Chung , N. , Han , H. and Koo , C. ( 2015 ) Adoption of travel information in user-generated content on social media: the moderating effect of social presence . Behav. Inform. Technol. , 34 , 902 – 919 . Google Scholar CrossRef Search ADS Cohen , J. ( 2001 ) Defining identification: a theoretical look at the identification of audiences with media characters . Mass Commun. Soc. , 4 , 245 – 264 . Google Scholar CrossRef Search ADS Constine , J. ( 2017 , June 20) Instagram Stories hits 250M daily users, adds Live video replays. Retrieved March 27, 2018, from https://techcrunch.com/2017/06/20/instagram-live-video-replays/ Coyle , J.R. and Thorson , E. ( 2001 ) The effects of progressive levels of interactivity and vividness in web marketing sites . J. Advert. , 30 , 65 – 77 . Google Scholar CrossRef Search ADS Daft , R.L. and Lengel , R.H. ( 1986 ) Organizational information requirements, media richness and structural design . Manag. Sci. , 32 , 554 – 571 . Google Scholar CrossRef Search ADS De Cesarei , A. , Codispoti , M. and Schupp , H.T. ( 2009 ) Peripheral vision and preferential emotion processing . NeuroReport , 20 , 1439 – 1443 . Google Scholar CrossRef Search ADS PubMed Dennis , A.R. , Fuller , R.M. and Valacich , J.S. ( 2008 ) Media, tasks, and communication processes: a theory of media synchronicity . MIS Quarterly , 32 , 575 – 600 . Google Scholar CrossRef Search ADS Dennis , A.R. and Valacich , J.S. ( 1999 , January) Rethinking media richness: towards a theory of media synchronicity. In Systems Sciences, 1999. HICSS-32. Proc. 32nd Annual Hawaii International Conference on (pp. 10-pp). IEEE. Dolezal , H. ( 1982 ) Living in a World Transformed: Perceptual and Performatory Adaptation to Visual Distortion. Academic Press. Eliot , T.S. ( 1957 ) The social function of poetry . On Poetry Poets , 19 , 33 . Farrar , K.M. , Krcmar , M. and Nowak , K.L. ( 2006 ) Contextual features of violent video games, mental models, and aggression . J. Commun. , 56 , 387 – 405 . Google Scholar CrossRef Search ADS Fischer , A.H. , Rotteveel , M. , Evers , C. and Manstead , A.S. ( 2004 ) Emotional assimilation: how we are influenced by others’ emotions . Curr. Psychol. Cogn. , 22 , 223 – 246 . Fisher , W.R. ( 1985 ) The narrative paradigm: an elaboration . Commun. Monogr. , 52 , 347 – 367 . Google Scholar CrossRef Search ADS Fornell , C. and Larcker , D.F. ( 1981 ) Evaluating structural equation models with unobservable variables and measurement error . J. Market. Res. , 18 , 39 – 50 . Google Scholar CrossRef Search ADS Galyean , T.A. , III . ( 1995 ) Narrative guidance of interactivity. Doctoral dissertation, Massachusetts Institute of Technology. Girgensohn , A. , Boreczky , J. , Chiu , P. , Doherty , J. , Foote , J. , Golovchinsky , G. , Uchihashi , S. and Wilcox , L. 2000 . A semi-automatic approach to home video editing. Proc. 13th Annual ACM Symposium on User Interface Software and Technology. ACM, pp. 81–89. Gould , S. , Arfvidsson , J. , Kaehler , A. , Sapp , B. , Messner , M. , Bradski , G.R. , Baumstarck , P. , Chung , S. and Ng , A.Y. ( 2007 ) Peripheral-foveal vision for real-time object recognition and tracking in video . IJCAI , 7 , 2115 – 2121 . Graesser , A.C. , Olde , B. and Klettke , B. ( 2002 ) How does the mind construct and represent stories . Narrative impact: Social and cognitive foundations , 229 – 262 . Green , M.C. ( 2004 ) Transportation into narrative worlds: the role of prior knowledge and perceived realism . Discourse Processes , 38 , 247 – 266 . Google Scholar CrossRef Search ADS Green , M.C. and Brock , T.C. ( 2000 ) The role of transportation in the persuasiveness of public narratives . J. Personal. Soc. Psychol. , 79 , 701 . Google Scholar CrossRef Search ADS Heeter , C. ( 1989 ) Classifying mediated communication systems . Commun. Yearbook , 12 , 477 – 489 . Heeter , C. ( 2000 ) Interactivity in the context of designed experiences . J. Interact. Advert. , 1 , 3 – 14 . Google Scholar CrossRef Search ADS Hiltz , S.R. , Coppola , N. , Rotter , N. , Toroff , M. and Benbunan-Fich , R. ( 2000 ) Measuring the importance of collaborative learning for the effectiveness of ALN: a multi-measure . Online Educ. , 1 , 101 – 119 . Cisco . Visual Networking Index: Global Mobile Data Traffic Forecast Update 2014–2019 White Paper. (n.d.). Retrieved June 24, 2015, from http://www.cisco.com/c/en/us/solutions/collateral/service-provider/visual-networking-index-vni/white_paper_c11-520862.html Hua , X.S. , Lu , L. and Zhang , H.J. ( 2004 ) Optimization-based automated home video editing system . IEEE Trans. Circuits Syst. Video Technol. , 14 , 572 – 583 . Google Scholar CrossRef Search ADS Jang , J. , Kim , J. , Shin , H. , Aum , H. and Kim , J. ( 2016 ) Effects of temporal format of everyday video on narrative engagement and social interactivity . Interact. Comput. , 28 , 718 – 736 . iwv043. Google Scholar CrossRef Search ADS Janssen , J.H. , IJsselsteijn , W.A. and Westerink , J.H. ( 2014 ) How affective technologies can influence intimate interactions and improve social connectedness . Int. J. Hum. Comput. Stud. , 72 , 33 – 43 . Google Scholar CrossRef Search ADS Kellogg , W.A. , Bellamy , R.K.E. and Van Deusen , M. ( 1997 ) A practical guide to working with edited video. In Handbook of Human-Computer Interaction ( 2 edn ). pp. 947 – 978 . Elsevier Science BV , Amsterdam . Google Scholar CrossRef Search ADS Kenney , W.P. ( 1988 ) How to Read and Write About Fiction . Arco , New York . Kim , J. ( 2015 ) Design for Experience: Where Technology Meets Design and Strategy . Springer . Kim , T. and Biocca , F. ( 1997 ) Telepresence via television: two dimensions of telepresence may have different connections to memory and persuasion . J. Comput. Mediated Commun. , 3 . Kirk , D. , Sellen , A. , Harper , R. and Wood , K. ( 2007 , April) Understanding videowork. Proc. SIGCHI Conf. Human Factors in Computing Systems (pp. 61–70). ACM. Klatzky , R.L. ( 1998 ) Allocentric and egocentric spatial representations: definitions, distinctions, and interconnections. In Spatial Cognition . pp. 1 – 17 . Springer , Berlin Heidelberg . Kolfschoten , G.L. ( 2011 ) Cognitive load in collaboration-brainstorming. Proc. 44th Hawaii International Conference on System Sciences. IEEE. pp. 1–9. Kraus , W. ( 2006 ) The narrative negotiation of identity and belonging . Narrat. Inquiry , 16 , 103 – 111 . Google Scholar CrossRef Search ADS Lamme , V.A. ( 2003 ) Why visual attention and awareness are different . Trends Cogn. Sci. , 7 , 12 – 18 . Google Scholar CrossRef Search ADS PubMed Langer , E. ( 2014 , June 18) Vine’s most popular user clues in the ‘dinosaurs’. Retrieved March 27, 2018, from https://www.cnbc.com/id/100899790 Lehmuskallio , A. and Sarvas , R. ( 2008 , October) Snapshot video: everyday photographers taking short video-clips. Proc. 5th Nordic Conf. Human-computer interaction: building bridges (pp. 257–265). ACM. Lienhart , R.W. , ( 1999 ) Dynamic video summarization of home video. In Electronic Imaging. pp. 378 – 389 . Lim , S. and Reeves , B. ( 2009 ) Being in the game: effects of avatar choice and point of view on psychophysiological responses during play . Media Psychology , 12 , 348 – 370 . Google Scholar CrossRef Search ADS Liu , Y. and Shrum , L.J. ( 2002 ) What is interactivity and is it always such a good thing? Implications of definition, person, and situation for the influence of interactivity on advertising effectiveness . J. Advert. , 31 , 53 – 64 . Google Scholar CrossRef Search ADS Luo , Y. and Tang , X. ( 2008 ) Photo and Video Quality Evaluation: Focusing on the Subject. In Computer Vision–ECCV 2008 . pp. 386 – 399 . Springer , Berlin, Heidelberg . Google Scholar CrossRef Search ADS McCarthy , J. and Wright , P. ( 2004 ) Technology as experience . Interactions , 11 , 42 – 43 . Google Scholar CrossRef Search ADS McMillan , S.J. and Hwang , J.S. ( 2002 ) Measures of perceived interactivity: an exploration of the role of direction of communication, user control, and time in shaping perceptions of interactivity . J. Advert. , 31 , 29 – 42 . Google Scholar CrossRef Search ADS Nalder , J. ( 2013 ) Making molehills from mountains: the tilt-shift effect . Screen Educ. , 68 , 98 . Neitzel , B. ( 2005 ) Levels of play and narration. In Meister , J.C. (ed.) , Narratology Beyond Literary Criticism. Mediality, Disciplinarity. Narratologia Band 6 . pp. 45 – 64 . Walter de Gruyter , Berlin and New York . Google Scholar CrossRef Search ADS Ojala , J. , Mate , S. , Curcio , I.D. , Lehtiniemi , A. and Väänänen-Vainio-Mattila , K. ( 2014 , November) Automated creation of mobile video remixes: user trial in three event contexts. In Proc. 13th Int. Conf. Mobile and Ubiquitous Multimedia (pp. 170–179). ACM. O’Hara , K. , Black , A. and Lipson , M. ( 2006 , April) Everyday practices with mobile video telephony. Proc. SIGCHI Conf. Human Factors in computing systems (pp. 871–880). ACM. Pelli , D.G. ( 1987 ) The visual requirements of mobility, in Low Vision . pp. 134 – 146 . Springer , New York . Peng , W.T. , Chu , W.T. , Chang , C.H. , Chou , C.N. , Huang , W.J. , Chang , W.Y. and Hung , Y.P. ( 2011 ) Editing by viewing: automatic home video summarization by viewing behavior analysis . IEEE Trans. Multimed. , 13 , 539 – 550 . Google Scholar CrossRef Search ADS Prensky , M. ( 2001 ) Fun, play and games: What makes games engaging . Digital Game-Based Learn. , 5 , 1 – 05 . Puikkonen , A. , Häkkilä , J. , Ballagas , R. and Mäntyjärvi , J. ( 2009 , September). Practices in creating videos with mobile phones. Proc. 11th Int. Conf. Human-Computer Interaction with Mobile Devices and Services (p. 3). ACM. Puikkonen , A. , Ventä , L. , Häkkilä , J. and Beekhuyzen , J. ( 2008 , December) Playing, performing, reporting: a case study of mobile minimovies composed by teenage girls. Proc. 20th Australasian Conf. Computer-Human Interaction: Designing for Habitus and Habitat (pp. 140–147). ACM. Qin , H. , Patrick Rau , P.L. and Salvendy , G. ( 2009 ) Measuring player immersion in the computer game narrative . Int. J. Hum. Comput. Interact. , 25 , 107 – 133 . Google Scholar CrossRef Search ADS Rayner , K. , Inhoff , A.W. , Morrison , R.E. , Slowiaczek , M.L. and Bertera , J.H. ( 1981 ) Masking of foveal and parafoveal vision during eye fixations in reading . J. Exp. Psychol. Hum. Percept. Perform. , 7 , 167 . Google Scholar CrossRef Search ADS PubMed Riedl , M.O. and Young , R.M. ( 2010 ) Narrative planning: balancing plot and character . J. Artif. Intell. Res. , 39 , 217 – 268 . Google Scholar CrossRef Search ADS Rigoulot , S. , Delplanque , S. , Despretz , P. , Defoort-Dhellemmes , S. , Honoré , J. and Sequeira , H. ( 2008 ) Peripherally presented emotional scenes: a spatiotemporal analysis of early ERP responses . Brain Topogr. , 20 , 216 – 223 . Google Scholar CrossRef Search ADS PubMed Rigoulot , S. , D’Hondt , F. , Honore , J. and Sequeira , H. ( 2012 ) Implicit emotional processing in peripheral vision: behavioral and neural evidence . Neuropsychologia , 50 , 2887 – 2896 . Google Scholar CrossRef Search ADS PubMed Roskos-Ewoldsen , B. , Davies , J. and Roskos-Ewoldsen , D.R. ( 2004 ) Implications of the mental models approach for cultivation theory . Communications , 29 , 345 – 363 . Google Scholar CrossRef Search ADS Schank , R.C. and Berman , T.R. ( 2002 ) The pervasive role of stories in knowledge and action. In Green , M.C. , Strange , J.J. and Brock , T.C. (eds) , Narrative Impact: Social and Cognitive Foundations . pp. 287 – 313 . Lawrence Erlbaum Associates . Schuurink , E.L. and Toet , A. ( 2010 ) Effects of third-person perspective on affective appraisal and engagement: findings from SECOND LIFE . Simulat. Gaming , 41 , 724 – 742 . Google Scholar CrossRef Search ADS Slater , M.D. , Rouner , D. and Long , M. ( 2006 ) Television dramas and support for controversial public policies: effects and mechanisms . J. Commun. , 56 , 235 – 252 . Google Scholar CrossRef Search ADS Somers , M.R. ( 1994 ) The narrative constitution of identity: a relational and network approach . Theory Soc. , 23 , 605 – 649 . Google Scholar CrossRef Search ADS Steeples , C. ( 2002 , January) Voice annotation of multimedia artifacts: reflective learning in distributed professional communities. In System Sciences, 2002. HICSS. Proc. 35th Annual Hawaii International Conference on (10 pp.). IEEE. Stern , B.B. ( 1991 ) Who talks advertising? Literary theory and narrative ‘point of view . J. Advert. , 20 , 9 – 22 . Google Scholar CrossRef Search ADS Steuer , J. , Biocca , F. and Levy , M.R. ( 1995 ) Defining virtual reality: dimensions determining telepresence . Commun. Age Virtual Real. , 33 – 56 . Sundar , S.S. ( 2004 ) Theorizing interactivity’s effects . Inform. Soc. , 20 , 385 – 389 . Google Scholar CrossRef Search ADS Tamborini , R. , Eastin , M. , Lachlan , K. , Skalski , P. , Fediuk , T. and Brady , R. ( 2001 ) Hostile thoughts, presence and violent virtual video games. In 51st Annual Convention of the International Communication Association, Washington, DC. Taylor , S.E. and Thompson , S.C. ( 1982 ) Stalking the elusive’ vividness’ effect . Psychol. Rev. , 89 , 155 . Google Scholar CrossRef Search ADS Tench , D. ( 1989 ) Narrative presence and narrative structure in Verga’s early novels . J. Narrat. Technique , 205 – 216 . Vogeley , K. and Fink , G.R. ( 2003 ) Neural correlates of the first-person-perspective . Trends Cogn. Sci. , 7 , 38 – 42 . Google Scholar CrossRef Search ADS PubMed Webb , N.A. and Griffin , M.J. ( 2003 ) Eye movement, vection, and motion sickness with foveal and peripheral vision . Aviat. Space Environ. Med. , 74 , 622 – 625 . Google Scholar PubMed Wignall , J. ( 2012 ) Digital Photography FAQs . John Wiley & Sons . Wilson , E. ( 1994 ) The Theater Experience . McGraw-Hill , Blacklick, OH . Wilson , B. and Henseler , J. ( 2007 ) Modeling reflective higher-order constructs using three approaches with PLS path modeling: A Monte Carlo comparison. Australian and New Zealand Marketing Academy Conference. Yarow , J. ( 2013 ) Vine sharing plummets after Instagram launches video. http://www.businessinsider.com/chart-of-the-day-vine-sharing-plummets-after-instagram-launches-video-2013-6 Zillmann , D. ( 1995 ) Mechanisms of emotional involvement with drama . Poetics , 23 , 33 – 51 . Google Scholar CrossRef Search ADS Zsombori , V. , Frantzis , M. , Guimaraes , R.L. , Ursu , M.F. , Cesar , P. , Kegel , I. , Craigie , R. and Bulterman , D.C. (2011) Automatic generation of video narratives from shared UGC. In Proc. 22nd ACM Conference on Hypertext and Hypermedia. ACM, pp. 325–334. APPENDIX Nar A example View largeDownload slide View largeDownload slide This condition has six scenes of first-person perspective with foveal vision continuously. Nar B example View largeDownload slide View largeDownload slide This condition has six scenes of first-person perspective with peripheral vision continuously. Nar C example View largeDownload slide View largeDownload slide This condition has six scenes of first-person perspective with intersecting foveal and peripheral vision. Nar D example View largeDownload slide View largeDownload slide This condition has six scenes of third-person perspective with foveal vision continuously. Nar E example View largeDownload slide View largeDownload slide This condition has six scenes of third-person perspective with peripheral vision continuously. Nar F example View largeDownload slide View largeDownload slide This condition has six scenes of third-person perspective with intersecting foveal and peripheral vision. Nar G example View largeDownload slide View largeDownload slide This condition has six scenes of foveal vision with intersecting one-person and third-person perspective. Nar H example View largeDownload slide View largeDownload slide This condition has six scenes of peripheral vision with intersecting one-person and third-person perspective. Nar I example View largeDownload slide View largeDownload slide This condition has mixed scene effects with intersecting one-person perspective X foveal vision and third-person perspective X peripheral vision. Nar J example View largeDownload slide View largeDownload slide This condition has mixed scene effects with intersecting one-person perspective X peripheral vision and third-person perspective X foveal vision. Author notes Handling Editor: Dr Sharon Tettegah © The Author(s) 2018. Published by Oxford University Press on behalf of The British Computer Society. All rights reserved. For Permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Interacting with Computers Oxford University Press

The New Snapshot Narrators: Changing Your Visions and Perspectives!

Loading next page...
 
/lp/ou_press/the-new-snapshot-narrators-changing-your-visions-and-perspectives-R5TbEon5Y0
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of The British Computer Society. All rights reserved. For Permissions, please email: journals.permissions@oup.com
ISSN
0953-5438
eISSN
1873-7951
D.O.I.
10.1093/iwc/iwy012
Publisher site
See Article on Publisher Site

Abstract

Abstract The emergence of high-performance mobile devices and communication networks has provided people with new ways of producing video content. In the past, people recorded videos using professional devices such as camcorders, but now they use their smartphones, wearable video recorders and drones. Because of this change, there is strong interest among users in Internet of Things environments in the context of how video is captured, edited, shared and interacted with. In particular, trends show a shift from professional-edited video content to user-generated video content, with the increasing popularity of video-sharing social media and the emergence of new video recording devices. However, there have been very few studies that have explored how a snapshot video can be edited. Only a few researchers have studied the effects of video on social interaction among users, and they have failed to consider how video content creation can facilitate social interaction. Therefore, we conducted experimental research to discover the impact of scene format (narrative focus and perspective) of everyday videos on narrative engagement and social interaction. We conducted two studies: (i) single condition effect of narrative focus and narrative perspective, and (ii) mixed condition effect of narrative focus and narrative perspective. The results indicated that the single narrative focus and narrative perspective affects narrative engagement and its four sub-constructs. In addition, the mixed narrative focus and narrative perspective affected narrative engagement, and the effects of interaction between them were determined. According to narrative engagement patterns, the tendencies of social interactions, including various system features, were different. The implications and limitations of the study’s results are discussed in the final section of the article. RESEARCH HIGHLIGHTS The effects of four sub-constructs of narrative engagement on social interactivity in video-sharing social media are presented. Scene format of a video strongly influences the narrative conveyed to others. Effects of narrative focus and narrative perspective on shared video are stronger when the scene format condition of the video is mixed. The effect of narrative engagement on viewers’ social interaction behaviors is explained. The main and interaction effects of scene formats on social interactivity in terms of the four sub-constructs of narrative engagement are verified. 1. INTRODUCTION Advances in technology over the past few years, including high-performance mobile devices and high-bandwidth communication networks, have provided new ways of producing video content (Kirk et al., 2007). In the past, people recorded videos using professional devices such as camcorders, but now they use their smartphones. This means that it is easier than ever before to produce a video. As a result, mobile video traffic on the Internet increased by 50% from 2012 to 2014 (Cisco, 2013). In particular, the amount of mobile video data has increased 16-fold. This massive growth in video traffic accounts for nearly two-thirds of all Internet traffic. Anderson (2010) said that the phenomena surrounding video usage are becoming integrated with social media and transcending traditional media. Because of this change, users in mobile environments are highly interested in how video is captured, edited, shared and engaged with. In particular, recent trends indicate a shift from professional-edited video content to user-generated video content, with the increasing popularity of online video-sharing applications such as Vine, and the emergence of new video recording devices, such as smart glasses and drones. User-generated video content has various properties. These are captured spontaneously, edited, and shared instantly, and are often meaningful in the context of the sharer’s experience. Owing to these properties, user-generated video provides a detailed record of the contextual and situational factors that become the background of the sharer’s experience (Steeples, 2002). Thus, editing after shooting is necessary to communicate the sharer’s experience, even if this is regarded as a cumbersome and inefficient task. In particular, most users have been to special events, such as a concert or sports event, where numerous people in the crowd hold up their mobile phones to record the event (Ojala et al., 2014); they often edit and share videos taken this way. For example, they use the video recording and editing functions of Instagram: users record videos while touching the record button. If they want to pause, they simply take their finger off the screen; if they want to continue recording, they just put their finger back on the screen. All the scenes thus recorded are automatically connected as if they were recorded continuously in one take. In this way, users can record only the scene at a desired moment on the fly. They know that the edited video can help them share their experiences more vividly with viewers than a non-edited video. In addition, video has an innate quality and richness that can never be fully captured via text (Kellogg et al., 1997). This means that the sharer can deliver their personal perspective of the experience to a viewer by video. Thus, unlike cinema, TV or animation, user-generated video content can lead to more impressive narratives in a short running time. Owing to these characteristics of videos, many studies on video usage have recently been conducted in the mobile environment context. O’Hara et al. (2006) verified a range of underlying motivations and values in various contexts in terms of the social practices surrounding video consumption on mobile devices. Puikkonen et al. (2008, 2009) investigated mobile video work in real life and identified video usage patterns. In particular, a video taken of everyday life is referred to as a snapshot video. Lehmuskallio and Sarvas (2008) described snapshot videos as lacking concern for techniques that characterize amateur or professional videography; instead they are rather quick, aesthetically poor exposures of real life. He said that a snapshot video is usually captured within and outside familiar contexts, disregarding professional video recording techniques. While a snapshot video is quite popular among users today, there are very few studies that have explored how a snapshot video is edited. A few researchers have conducted studies on the effect of video on social interaction among users, but they have neglected the methods of producing video content for facilitating social interaction. Some studies have described video as an effective medium for delivering impressions to others. For example, video has been described as a medium for giving one’s impressions about the presence of others (Chung et al., 2015; Daft and Lengel, 1986). Some researchers have argued that the effects of mediums such as a video can vary depending on the relationships between the communications channel and the type of medium (Dennis and Valacich, 1999; Hiltz et al., 2000). Communicating with others on low synchronicity (e.g. e-mail or bulletin board) may be appropriate for conveyance of information, whereas communicating with others on high synchronicity (e.g. face-to-face or video interaction) may be more desirable for convergence on shared meaning (Dennis et al., 2008). However, the way video sharers make videos to encourage further social interaction with others is yet to be investigated in the field of human–computer interaction (Bornoe and Barkhuus, 2010; Peng et al., 2011; Zsombori et al., 2011). In addition, it is difficult to deliver the video sharer’s experience without appropriate editing, because video can provide a richer narrative compared with other media, such as text and photo (Girgensohn et al., 2000; Hua et al., 2004; Lienhart, 1999). Accordingly, a well-edited video in terms of narrative is important to deliver the sharer’s experience and facilitate social interaction among users through the snapshot video. This study focuses on narrative structure, because a snapshot video has a meaningful story that is strongly linked to the sharer’s personal experience. The experience can be structuralized by a narrative in which people present their story (Chatman, 1975; Somers, 1994). Narratives include not only the sharer’s actions and feelings but also reflections about his/her actions and feelings (Bruner and Lucariello, 1989). The main purpose of making the narrative is to convey what has happened or what is happening to them through structuralization to show their experience to others. Therefore, it is essential to edit the sharer’s video based on the principle of narrative structure. There are two important types of video narrative structures in this context: narrative focus and narrative perspective. Narrative focus is what the video sharer, as a narrator, focuses on regarding his/her experience. The ‘focus’ on narrative can be controlled by a highlight effect. A video scene can be focused on the overall scenery, but it can also be focused on highlighted scenery as the central event of experience. Narrative perspective refers to the video sharer’s perspective in each video scene. The ‘perspective’ of the narrative can be controlled by point of view, i.e. videos can have a first-person or third-person point of view. Each video scene can have the same point of view continuously, but can also shift between different points of view. Our aim in this study is to investigate the narrative structure of video for facilitating social interaction among users. In particular, we aim to verify the structural effects of video, including narrative focus and narrative perspective, when delivering the sharer’s experience to a viewer. The effect of video narratives can be explained by the concept of narrative engagement. In our previous study, we verified that the duration of a video can affect the viewer’s narrative engagement. Narrative engagement is defined as the outcome of a convergent process where all mental systems and capacities of the viewer are focused on the sharer’s narrative being played out in the video (Jang et al., 2016). In our previous study, we found that the optimum running time and number of cut scenes to facilitate viewer’s narrative engagement with a snapshot video is 24 s and 6 scenes, respectively, in a mobile environment. However, we could not understand why the viewer tends to engage with the video narrative based on the structural effect of the narrative, nor the tendency viewers to use social features. Therefore, we modified the duration of video stimuli and manipulated the scene format in order to verify the effect of narrative structure in this study. In addition, we focused not only on the detailed differentiation of narrative engagement by the narrative structure of videos but also on the tendency toward social interactive behavior through the socially functional system of our study. For the experiments, we improved our experimental system to provide participants with a more advanced video recording system than in the previous study. In our previous experimental system, we created the video stimuli manually, and the participants used the low-fidelity version of the socially interactive system in the limited mobile environment. In the current experimental system, we developed an automatic synchronous video recording system connected with Google Glass and a drone. Before the main experiment, participants were given access to our video recording system, which they used to share their videos by posting these on our experimental social media. The videos were used to determine what people are usually interested in. From this, we finally decided on the event, story and narrative of video stimuli, and manipulating these to control external factors, such as contents, characters and events. The next section describes our theoretical background and hypotheses in terms of the narrative structure of snapshot videos, including narrative engagement, narrative focus and narrative perspective related to point of view. Section 3 explains the method used in our study. Section 4 shows the results based on the statistical analysis. Section 5 presents a general discussion, and Section 6 explains the limitations and implications of this study. 2. THEORETICAL BACKGROUND 2.1. Narrative structure and engagement According to the narrative paradigm, a narrative is a way of constructing stories (Fisher, 1985). Schank and Berman (2002) defined narrative as the detailed reconstruction of the human experience. In this perspective, McCarthy and Wright (2004) said that it is important to decide what the beginning and end of the story are, and how the experience is structuralized by narrative. In other words, while a story is the event that occurred itself, a narrative is the reconstructed story. These reconstructions provide us with the meaning behind the event (Bruner and Lucariello, 1989; Green and Brock, 2000). The primary goal of structuring a narrative is to communicate a specific message to the audience and to impress it upon them (Neitzel, 2005; Qin et al., 2009). Audiences understand the narrative flow of causal relations among events having a linear time sequence (Abolafia, 2010; Bordwell et al., 1997). Through this understanding, the narrative affects the audience’s beliefs, attitudes and behavioral intentions (Appel and Richter, 2007; Brock et al., 2002; Slater et al., 2006). A well-structured narrative draws the attention of viewers, who feel and imagine that they are in narrative themselves (Green and Brock, 2000). The status of a viewer’s immersion in the narrative is explained by the concept of narrative engagement (Busselle and Bilandzic, 2009). According to Busselle and Bilandzic (2009), narrative engagement can be explained by four sub-constructs: attentional focus, narrative understanding, narrative presence and emotional engagement. Attentional focus is defined as the degree of concentration on a specific target without distraction. Narrative understanding is defined as the degree of understanding of the narrative. Narrative presence is defined as the extent of feeling present in the narrated world. Emotional engagement is defined as the level of sympathy for characters in the narrative. These sub-constructs are based on a mental model created for understanding narratives. People understand a narrative by interpreting its background, characters and situations through their knowledge of the real world (Graesser et al., 2002; Roskos-Ewoldsen et al., 2004). In this process, people understand the situation in the narrative or empathize with the emotions of characters, given their absorption with the characters (Cohen, 2001; Zillmann, 1995). In addition, people feel immersed in the narrative, which is referred to as telepresence (Green and Brock, 2000; Green, 2004). Consequently, the viewer has fun and understands what the narrative is saying through immersion in the narrative (Prensky, 2001). This study clarifies the effects of narrative structure on each sub-construct of narrative engagement. We already know that people in the mobile environment receive the sharer’s experience from a snapshot video through narrative, and then conduct socially interactive behavior depending on their level of narrative engagement (Jang et al., 2016). However, we could not verify how the viewer of the video engages with the sharer’s thoughts, feelings and situations through video narrative. For the purpose of the present study, we assumed that the change in narrative structure caused by the format of scenes in videos affects viewers’ narrative engagement. 2.2. Narrative focus Narrative focus is defined as the structural range of focus on the narrative. Sharers of the narrative can focus on the totality of actions, events, characters and settings, but they can also focus on just one or a combination of these (Kraus, 2006). Stage lighting for drama is a good example of the narrative focus effect. The lighting sometimes lights up the whole stage using floodlights, but can also light a specific area of the stage using a spotlight. These lighting methods direct viewers’ attention to the central event occurring on the whole stage to emphasize important points in the narrative (Wilson, 1994). The narrative focus effect can also be explained by selective focusing. The effect of selective focusing is that the background or foreground of a photo is used to highlight the subject of the photo (Nalder, 2013; Wignall, 2012). As people pay attention to narrative in a snapshot video, their eyes respond to scene changes, including physiological changes in the retina. The retina is made up of rod and cone cells. Cones detect the detailed information of objects. Because they are at the center of the retina, the angle of sight in order to clearly see an object is only 5° (Gould et al., 2007). The areas of high and low acuity in the central vision are called foveal vision and peripheral vision, respectively (Rayner et al., 1981; Webb and Griffin, 2003). In this study, we adopted methods of controlling narrative focus in each video scene through two types of vision: foveal vision and peripheral vision. The video scene for foveal vision has a focusing effect such that it emphasizes a specific area of the sharer’s experience by calculating the degree of central acuity in their eyes. In contrast, the video scene for peripheral vision has high clarity of visibility by high resolution without any other blurred areas in the whole scene. 2.3. Narrative perspective Narrative perspective is explained as the structural sight of delivering a situation to viewers. Brooks and Warren (1959) divided the narrative perspective into four types/parts depending on who the narrator is. The most commonly used criteria is point of view. According to this method of presenting a situation, the narrative perspective can take the form of a first-person point of view, first-person narrator’s point of view, third-person narrator’s point of view, and omniscient point of view (Klatzky, 1998; Stern, 1991; Vogeley and Fink, 2003). A first-person point of view is a narration in which the central character of the story describes the events of a narrative directly. A first-person narrator’s point of view is a narration in which the narrator is present in the story itself and describes the events. A third-person narrator’s point of view is one in which the narrator is outside of the story and describes the events objectively as an observer. Finally, the omniscient point of view offers narration in which the narrator, who is outside of the story, describes not only the events but also the inner thoughts of characters in the story (Galyean, 1995). We assumed that the snapshot videos were taken using wearable glasses or drones. A video scene recorded using wearable glasses is equivalent to a first-person point of view, with a narrator describing his or her story directly. When this point of view is recorded using wearable glasses, we can know that the sharer’s eyes move toward their central experience, which is subjective and emotional. A video scene taken by a drone is equivalent to a third-person narrator’s point of view. When this point of view is recorded by the drone, we can know the sharer’s surrounding environment. In this study, we controlled the narrative perspective in each video scene through two types of perspectives: first-person perspective and third-person perspective. The video scene with the first-person perspective contains images recorded at/from the sharer’s own eye level. On the other hand, the video scene with the third-person perspective has a bird’s eye-view recorded by means of a helicam. 2.4. Active social interaction Social interaction has been described by the concept of interactivity. Many scholars have studied interactivity as a technical property belonging to computer-mediated environments (Heeter, 2000; Sundar, 2004). In general, interaction features have been considered in terms of functionalities, such as frequency (Liu and Shrum, 2002), controllability (Betrancourt, 2005; Coyle and Thorson, 2001), complexity (Heeter, 2000) and so on. However, these have some limitations. First, the functional features of mobile environments may reveal the potential interactivity but not the actual social interaction. The potential interaction between users is facilitated when they engage perceptually or physically with the virtual environment (Bucy, 2004). Second, in interactions between people in face-to-face communications, it is possible for users to interact with others more deeply when a particular feature is encountered in a mobile environment. This means that existing functions of systems, such as ‘Like’ buttons, need to be considered from a cognitive perspective; we also need to interpret social features according to differences in social interaction behavior. Some scholars have studied this as a user’s perceptual experience (Bucy, 2004; McMillan and Hwang, 2002). This perspective describes social interaction as a kind of expression of psychological state by users during the interaction with a medium. In particular, because the nature of an interaction depends on the process, it is important to understand social interaction as a cognitive level of interaction. This is revealed by the effort level of the action (Heeter, 1989) as an active social interaction in which people use social features to interact with others more deeply. While many studies have applied the concept of active social interaction, few have investigated the features of socially active behavior. Moreover, there has been little research on video and system features that affect active social interaction via narrative engagement. Therefore, we defined active social interaction as a socially engaging behavior, and we applied four system features: click a ‘Like’ button, write a ‘Comment’, ‘Recall’ his or her friend, and ‘Share’ the video to others with comments. These were applied by the concept of effort level of action, with ‘Like’ relatively lighter and ‘Give’ relatively deeper in terms of level of social interaction. We applied these features according to relational effort using system features for measuring the amount of effort people actually exert in socially active behavior. 3. RESEARCH HYPOTHESIS Based on the theoretical background and previous studies, we set hypotheses on how narrative focus and narrative perspective affect attentional focus, narrative understanding, narrative presence and emotional engagement, as the detailed concepts of narrative engagement. We also set hypotheses on how the differences between single and mixed conditions affect narrative engagement. 3.1. The effect of narrative focus on narrative engagement and four sub-constructs A video having a low vision effect makes viewers feel discomfort and cognitive dissonance. In particular, the viewer feels the artificial situations owing to the forced attention, and experiences physical discomfort (Riedl and Young, 2010). These feelings and physical discomfort induce cognitive side-effects among viewers, including dizziness, unsteadiness and disorientation, when they are restricted to using foveal vision (Alfano and Michel, 1990). Likewise, in a mobile environment, videos with foveal vision compel viewers to focus on the central event of the sharer’s experience. This can trigger viewers’ lack of attentional focus to the snapshot video because of their lack of freedom to watch what they want. This lack of freedom leads to low interest in the snapshot video; consequently, viewers may give up or shift their attention to other videos on social media. Therefore, Hypothesis 1. All video scenes having peripheral vision have more positive effects on attentional focus than those having only foveal vision. When viewers are compelled to watch a video with their foveal vision, their task performance in relation to the video is low. For instance, viewers’ game performance decreased when foveal vision was 10°, and their shopping performance decreased when foveal vision was 4° (Pelli, 1987). Dolezal’s (1982) study required participants to wear glasses that induced restricted foveal vision in real life environments. The participants were surprised when they suddenly caught sight of the object because of the lack of recognition of the surrounding environment. These results indicate that people can experience lack of recognition and understanding of surrounding environments when experiencing these through foveal vision. Likewise, we assumed that videos having scenes of foveal vision trigger a lack of narrative understanding among viewers. Videos having scenes of peripheral vision can increase viewers’ willingness to focus on what they want to watch (Lamme, 2003). Conversely, videos with scenes of foveal vision can lead to viewers’ lack of narrative understanding of snapshot videos because of the limited information to understand narrative. This means that the limitation of vision leads to a low understanding of snapshot videos; consequently, the viewer may miss the video’s narrative or ask the sharer for clarification on social media. Therefore, Hypothesis 2. All video scenes having peripheral vision have more positive effects on narrative understanding than those having only foveal vision. Vividness is important to provide a sense of presence, which refers to the ability to perceive (Kim, 2015). This is explained as an abundance of stimulation that is provided by a medium to a person (Steuer et al., 1995). Multimedia content, such as photos and videos, provide more vividness than text because of the different amounts of abundant information (Taylor and Thompson, 1982). This can be manipulated by two technical factors: the breadth and depth of stimulation (Biocca, 1992). In particular, the depth of stimulation is the most effective means of increasing stimulation. Depth is the quality of the perception of the stimulation. For example, videos with high resolution can provide greater depth of perception. Likewise, videos with scenes of peripheral vision can provide a greater sense of presence than videos with foveal vision only. Foveal vision induces people to focus on the central event of the sharer’s experience, but it also triggers the poor depth of perception about the surrounding environment of the event. On the other hand, peripheral vision provides visual clarity of the whole scene and enables viewers to perceive surroundings in with great depth of perception. This means that videos having peripheral vision can provide a strong sense of presence; consequently, the viewer feels as though they are involved in the sharer’s experience. Hypothesis 3. All video scenes having peripheral vision have more positive effects on narrative presence than those having only foveal vision. Foveal vision disrupts the processing of emotional information (Calvo and Lang, 2005; De Cesarei et al., 2009; Rigoulot et al., 2012). According to Rigoulot et al. (2012), emotional information is processed automatically, even when it is displayed in low visual conditions. This result indicates that peripheral vision affects human cognition of emotional information in the surrounding environment. The reaction time for the emotional information of peripheral vision was faster than when the information was presented in foveal vision (Rigoulot et al., 2008). This means that people are sensitive to their peripheral vision in order to feel a situation comprehensively. Likewise, videos having scenes of foveal vision can trigger a lack of emotional engagement in the viewer. In particular, the processing of emotional scenes could be influenced by the characteristics of scenes, such as greater luminance, contrast or amount of color saturation (Calvo and Lang, 2005). This means that foveal vision perceives less vivid emotional information than peripheral vision; consequently, the viewer may find it difficult to empathize with the narrative’s emotional flow. Therefore, Hypothesis 4. All video scenes having peripheral vision have more positive effects on emotional engagement than those having only foveal vision. Overall, we assumed that video scenes having peripheral vision can strongly facilitate viewers’ narrative engagement. Hypothesis 5. All video scenes having peripheral vision have more positive effects on narrative engagement than those having only foveal vision. 3.2. The effect of narrative perspective on narrative engagement and four sub-constructs First-person narration has the advantage of expression, whereby it can communicate faithfully represented thoughts to others (Lim and Reeves, 2009). In the game-playing environment, the first-person perspective appears to enhance the immersion effect, and people feel the dynamics and focus on the game’s narrative (Farrar et al., 2006; Tamborini et al., 2001). These feelings of trust and immersion facilitate attention, which is being drawn to the main event in the narrative. Likewise, we assumed that snapshot videos having a first-person perspective provide viewers with more dynamic feelings. First-person video narratives cannot include the sharer as a character in the events in the video; therefore, the viewer directs his/her attention to the sharer’s experience by putting his/her ego at the center of the video narrative. This means that a first-person video attracts viewer’s attention more than that of a third-person video; consequently, the viewer tends to exhibit more attentional focus. Therefore, Hypothesis 6. All video scenes with a first-person perspective have more positive effects on attentional focus than those using a third-person perspective. The degree of consumers’ comprehension of the message of advertisements is influenced by the difference in perspectives on message delivery (Stern, 1991). This results from the amount of information on video. A snapshot video using the first-person perspective gives limited information for viewers to understand the narrative of the video, which leads to a lack of understanding of the video. However, a snapshot video using the third-person gives viewers gives viewers a sufficient amount of information for them to understand the narrative. Likewise, snapshot videos on social media often have various narrative elements, including surrounding events with no artificial editing (Bornoe and Barkhuus, 2010). This is the reason why the perspective of the video affects the amount of information delivered in the narrative. The overall situation of the sharer’s experience is shown by the third-person perspective, which has a richer narrative than the first-person perspective. Consequently, the viewer can obtain increased understanding of the narrative. Therefore, Hypothesis 7. All video scenes having a third-person perspective have more positive effects on narrative understanding than those with a first-person perspective. The first-person element in games, such as shooting games, is more dynamic in engendering feelings of presence (Farrar et al., 2006). This results from the loss of awareness of self and surroundings (Busselle and Bilandzic, 2009). However, the first-person view makes people focus on the game situation. This effect can disturb them by requiring them to construct an alternate world mentally in a short span of time. Thus, the construction of an alternate world is gradually achieved by gamers. Likewise, it can be speculated that first-person snapshots can disturb the feeling of presence because of the impossibility of constructing an alternate world in a short time. In contrast, third-person videos can make people construct their mental model by imagining the sharer’s experience. It has vividness because people receive situational and contextual information to feel presence (Kim, 2015). This means that the feeling of narrative presence requires an understanding of the whole situation (Schuurink and Toet, 2010; Tench, 1989); consequently, the third-person view of videos can provide more narrative presence than the first-person view. Therefore, Hypothesis 8. All videos having the third-person perspective have more positive effects on narrative presence than having the first-person perspective. People can feel empathy toward others’ emotions by means of various factors, including voice, facial expressions, gestures and movement (Castellano et al., 2008). They construct the mental model of another person’s world by grasping whole situations, at which point they become assimilated emotionally (Fischer et al., 2004; Janssen et al., 2014). In particular, vivid information on emotion helps them engage with the emotional narrative and achieve empathy. Likewise, third-person videos can induce more emotional engagement than first-person videos. The third-person provides varied information, including the whole atmosphere of the event, surrounding situation, and the sharer’s behaviors and facial expressions. This means that the third-person in a snapshot video creates an omniscient viewpoint; consequently, viewers can engage with the sharer’s emotions naturally. Therefore, Hypothesis 9. All video scenes having the third-person perspective have more positive effects on emotional engagement than those having the first-person perspective. Despite the effects of the first-person perspective on attentional focus, we assumed that the overall effects of videos having the third-person perspective can more strongly facilitate viewers’ narrative engagement than videos having the first-person perspective. Hypothesis 10. All video scenes having the third-person perspective have more positive effects on narrative engagement than those having the first-person perspective. 3.3. The effect of mixed narrative focus on narrative engagement People usually communicate their willingness to share experiences with others using highlighting (Baudisch et al., 2003; Luo and Tang, 2008). If the sharer imposes his or her ideas continuously onto others, however, they feel restricted because they have no freedom to imagine for understanding narrative. On the other hand, if the sharer gradually highlights his/her ideas to others, they make a cognitive effort to understand the narrative (Kolfschoten, 2011). In this respect, mixed vision of videos has the advantage of engaging viewers more than single vision. Foveal vision has a lack of surrounding information, while peripheral vision has relatively abundant information. Because of these differences in the video narrative, scenes that change between foveal and peripheral visions ‘increase viewers’ cognitive load for engaging with the video narrative. In particular, viewers can grasp the core event of the narrative in foveal vision, and infer the whole narrative of the sharer’s experience in peripheral vision. This natural process has a more positive effect on engaging with the narrative. Therefore, Hypothesis 11. Videos with mixed vision have more positive effects on narrative engagement than having only a single vision. 3.4. The effect of mixed narrative perspective on narrative engagement The point-of-view determines the range of content and the expressions for narrating events and characters (Black et al., 1979). The first-person point of view is commonly used to express personal attitudes and values. This point of view reveals intimate feelings, moods, state of mind, and is instantly interesting (Eliot, 1957; Stern, 1991). The third-person point of view provides information to an audience about the whole situation. This point of view reveals everything, including surrounding information, character’s behavior, moods with facial expressions and various factors of the situation (Kenney, 1988). However, using a single point-of-view has some disadvantages when describing the narrative. The first-person point of view lacks information variety in which the narrator plausibly talks about others as well as he/she does about him/herself; eventually, audiences may become bored by the limited and self-centered point of view. On the other hand, the third-person point of view lacks naturality when delivering the narrator’s real experience. The third-person point of view is unnatural because it frequently shows obvious situations. In this respect, the mixed point-of-view video is more engaging than a single point-of-view video. The viewer can feel the dynamics and perceive the sharer’s personal experiences more vividly through the first-person. In addition, the viewer can feel the whole atmosphere of the video narrative and receive the sharer’s situational experience more obviously through the third-person (Lim and Reeves, 2009). Mixing these in a video enables us to understand the narrative and empathize more with the character’s situation in the video. Therefore, Hypothesis 12. A mixed point-of-view video has more positive effects on narrative engagement than a single point-of-view video. 3.5. The interaction effect of scene format on narrative engagement We can assume that the mixed point-of-view affects narrative engagement more positively when the vision of the video is mixed. As referred to earlier, mixed vision can facilitate viewers’ engagement more than single vision. This effect is caused by the cognitive effort of the viewer in response to the lack of information from the foveal and peripheral vision. Then, if the mixed point-of-view is provided, viewers can better understand the sharer’s situation by bridging the gap between each scene. Thus, we hypothesize that the positive impacts of mixed point-of-view on viewers’ narrative engagement will be stronger when the vision of the video is also mixed. Therefore, Hypothesis 13. The effect of the mixed point-of-view on narrative engagement will strengthen if the vision of the video is mixed. 4. STUDY This study sought to verify how narrative focus and narrative perspective affect the four sub-constructs of narrative engagement, and the tendency toward social interaction in a mobile environment. 4.1. Study design The purpose of this study is to verify the effect of narrative focus as a structural range, and narrative perspective as a structural sight, on narrative engagement. In addition, we wanted to know how narrative engagement affects viewers’ tendency towards social interaction. We believe that these are important in the design of wearable devices and new video-sharing social media. The study is divided into two parts. To examine hypotheses 1–8, we set a 2 × 2, i.e. (narrative focus: foveal vision/peripheral vision) × (narrative perspective: first person/third person), within-subject experimental design as study 1. To examine hypotheses 9 and 10, we set a 2 × 2, i.e. (narrative focus: single vision/mixed vision) × (narrative perspective: single person/mixed person), within-subject experiment design as study 2 (Table 1). TABLE 1. Study design. Study 2 Single vision (F and P) Mixed vision (F–P or P–F) Study 1 Foveal vision (F) Peripheral vision (P) Single person (1 and 3) The first-person (1) 1F-1F-1F-1F-1F-1F 1P-1P-1P-1P-1P-1P 1F-1P-1F-1P-1F-1P The third-person (3) 3F-3F-3F-3F-3F-3F 3P-3P-3P-3P-3P-3P 3P-3F-3P-3F-3P-3F Mixed-person (1–3 or 3-1) 1F-3F-1F-3F-1F-3F 3P-1P-3P-1P-3P-1P 1F-3P-1F-3P-1F-3P 3P-1F-3P-1F-3P-1F Study 2 Single vision (F and P) Mixed vision (F–P or P–F) Study 1 Foveal vision (F) Peripheral vision (P) Single person (1 and 3) The first-person (1) 1F-1F-1F-1F-1F-1F 1P-1P-1P-1P-1P-1P 1F-1P-1F-1P-1F-1P The third-person (3) 3F-3F-3F-3F-3F-3F 3P-3P-3P-3P-3P-3P 3P-3F-3P-3F-3P-3F Mixed-person (1–3 or 3-1) 1F-3F-1F-3F-1F-3F 3P-1P-3P-1P-3P-1P 1F-3P-1F-3P-1F-3P 3P-1F-3P-1F-3P-1F TABLE 1. Study design. Study 2 Single vision (F and P) Mixed vision (F–P or P–F) Study 1 Foveal vision (F) Peripheral vision (P) Single person (1 and 3) The first-person (1) 1F-1F-1F-1F-1F-1F 1P-1P-1P-1P-1P-1P 1F-1P-1F-1P-1F-1P The third-person (3) 3F-3F-3F-3F-3F-3F 3P-3P-3P-3P-3P-3P 3P-3F-3P-3F-3P-3F Mixed-person (1–3 or 3-1) 1F-3F-1F-3F-1F-3F 3P-1P-3P-1P-3P-1P 1F-3P-1F-3P-1F-3P 3P-1F-3P-1F-3P-1F Study 2 Single vision (F and P) Mixed vision (F–P or P–F) Study 1 Foveal vision (F) Peripheral vision (P) Single person (1 and 3) The first-person (1) 1F-1F-1F-1F-1F-1F 1P-1P-1P-1P-1P-1P 1F-1P-1F-1P-1F-1P The third-person (3) 3F-3F-3F-3F-3F-3F 3P-3P-3P-3P-3P-3P 3P-3F-3P-3F-3P-3F Mixed-person (1–3 or 3-1) 1F-3F-1F-3F-1F-3F 3P-1P-3P-1P-3P-1P 1F-3P-1F-3P-1F-3P 3P-1F-3P-1F-3P-1F 4.2. Participants We recruited participants who were active users of video-sharing social media, such as Instagram (usually its video-sharing feature), Vine, Roadmovie and so on (Constine, 2017; Langer, 2014). In particular, those with experience in using video editing features, like scene filters, could apply. We chose our participants based on the criterion of level of experience in using video-sharing social media, such as Instagram, Vine and RoadMovie, which were the most popular video-sharing social media among users at the time of the experiment. We also considered the active use of video-sharing social media as a level of experience. The level of experience with video-sharing social media was set according to the ratio of shares or likes of other videos by the user. This criterion is important when choosing participants for our experiment because this indicates their adaptability to the experimental system. If a participant is not experienced in using video-sharing social media, he/she may report strange feelings when faced with our experimental system. Through these criteria and considerations, we recruited participants from university online communities and Facebook. All potential participants self-reported their usage of video-sharing social media. In general, participants usually shared at least one daily personal video a day and liked at least five videos made by others per day. To confirm their usage, we collected data on their ‘sharing’ and ‘liking’ behaviors from their personal social media accounts for 3 months. Finally, 52 undergraduate students were chosen for our study (i.e. studies 1 and 2). The subjects were between the ages of 20 and 28, with average age of 24.2. There were 27 (43.5%) males and 35 (56.5%) females. They voluntarily participated in the experiment and received monetary compensation equivalent to USD 15. 4.3. Stimuli manipulation The narrative focus was manipulated by ‘bokeh’, which is the aesthetic quality of the blur produced in the out-of-focus parts of an image caused by a lens (Buhler and Wexler, 2002). This effect occurs in parts of the scene that lie outside the depth of field. When we made the video stimuli, we applied the sharer’s eye-tracking data to adjust the effect to their foveal vision. The application of foveal vision can provide the central event of the sharer’s experience. We adopted two scene manipulations as the structural range of narrative focus: foveal vision and peripheral vision. Foveal vision has a low-level depth of field, whereas peripheral vision has a high-level depth of field. The narrative perspective was manipulated by points of view. We adopted two scene manipulations as the structural sight of the narrative perspective: the first-person point of view and third-person point of view. The first-person point of view is a visual representation method with an egocentric reference frame which shows that one’s own multimodal experiential space is centered on one’s own body (Vogeley and Fink, 2003). On the other hand, the third-person point of view is a visual representation method with an allocentric reference frame that has a perspective other than that of the narrator (Aguirre and D’Esposito, 1999). 4.4. Narrative of video stimuli Video stimuli consisted of fourteen separate video narratives. As a side effect, the narrative effect of video stimuli may vary depending on the interest of participants in their content. To avoid side effects due to contents, raw video contents were recorded by following the same storyline. One of our researchers played the role of the heroine, and she rode various amusement park rides over the course of one day. Video narrative subjects include ‘Air Soccer’, ‘Drum Basket’, ‘Flume Ride’, ‘Roller-Coaster’, ‘Gyro Drop’, ‘Swing Tree’, ‘Bungee Drop’, ‘Suspended Swinging Coaster’, ‘Merry-Go-Round’, ‘Dance Parade’, ‘Cups’, ‘Ferris Wheel’, ‘Mini Basket Ball’ and ‘Dodgems’. Among them, ‘Flume Ride’, ‘Gyro Drop’, ‘Suspended Swinging Coaster’ and ‘Ferris Wheel’ were used for the preliminary study. Ten video narratives were selected and used according to our study process, including study 1 and study 2. All videos had a running time of 24 s and 6 cut scenes. All of the video contents contained a narrative structure, including a beginning, middle and end of scene. In total, there were 100 (10 video narratives × 10 video structures) video stimuli selected randomly for each condition per participant (see Appendix) (Table 2). TABLE 2. Narrative contents. TABLE 2. Narrative contents. 4.5. Measurement We already verified the measurement of narrative engagement and its four sub-constructs (i.e. attentional focus, narrative understanding, narrative presence and emotional engagement) in our previous study (Jang et al., 2016). We applied these to the present experiment, and the phrasing of the questions was modified according to the wordings from the preliminary study. In addition, all the questions were checked against the original versions of measurement, including Appel and Richter (2007), Kim and Biocca (1997), and Cohen (2001). To get more validation of measurement, we adopted an additional item in each questionnaire for the four sub-constructs. Finally, we set questionnaires for the four constructs, including three-, four- or five-question items, and measured the responses with a seven-point Likert scale (see Table 3). TABLE 3. Measurement items. Variables Item Measurement References from Original Source Attentional focus AF1 While watching, I was fully concentrated on the narrative. Jang et al. (2016) Busselle and Bilandzic (2009) AF2 While watching, I was fully focused in the narrative. AF3 While watching, I had forgotten about other thoughts. New item for this study Narrative understanding NU1 This video’ narrative was naturally understandable. Jang et al. (2016) Busselle and Bilandzic (2009) NU2 It was difficult to understand the narrative of this video. NU3 I could easily follow the flow of the narrative in this video Appel and Richter (2007) NU4 I had a hard time recognizing the narrative of this video. NU5 While watching, I understood why the narrative of video was proceed. New item for this study Narrative presence NP1 While watching the video, I lost track of time. Jang et al. (2016) Kim and Biocca (1997) NP2 While watching the video, I completely forgot that I was in the middle of the experiment. Busselle and Bilandzic (2009) NP3 While watching the video, I forgot my everyday concerns. New item for this study Emotional engagement EE1 While watching the video, I felt I knew exactly what the sharer was going through emotionally. Jang et al. (2016) Cohen (2001) EE2 I never really shared the emotions of the sharer. EE3 While watching the video, I could feel the emotions the sharer portrayed. Busselle and Bilandzic (2009) EE4 This video affected the emotional changes to me. New item for this study Variables Item Measurement References from Original Source Attentional focus AF1 While watching, I was fully concentrated on the narrative. Jang et al. (2016) Busselle and Bilandzic (2009) AF2 While watching, I was fully focused in the narrative. AF3 While watching, I had forgotten about other thoughts. New item for this study Narrative understanding NU1 This video’ narrative was naturally understandable. Jang et al. (2016) Busselle and Bilandzic (2009) NU2 It was difficult to understand the narrative of this video. NU3 I could easily follow the flow of the narrative in this video Appel and Richter (2007) NU4 I had a hard time recognizing the narrative of this video. NU5 While watching, I understood why the narrative of video was proceed. New item for this study Narrative presence NP1 While watching the video, I lost track of time. Jang et al. (2016) Kim and Biocca (1997) NP2 While watching the video, I completely forgot that I was in the middle of the experiment. Busselle and Bilandzic (2009) NP3 While watching the video, I forgot my everyday concerns. New item for this study Emotional engagement EE1 While watching the video, I felt I knew exactly what the sharer was going through emotionally. Jang et al. (2016) Cohen (2001) EE2 I never really shared the emotions of the sharer. EE3 While watching the video, I could feel the emotions the sharer portrayed. Busselle and Bilandzic (2009) EE4 This video affected the emotional changes to me. New item for this study AF, attentional focus; NU, narrative understanding; NP, narrative presence; EE, emotional engagement. TABLE 3. Measurement items. Variables Item Measurement References from Original Source Attentional focus AF1 While watching, I was fully concentrated on the narrative. Jang et al. (2016) Busselle and Bilandzic (2009) AF2 While watching, I was fully focused in the narrative. AF3 While watching, I had forgotten about other thoughts. New item for this study Narrative understanding NU1 This video’ narrative was naturally understandable. Jang et al. (2016) Busselle and Bilandzic (2009) NU2 It was difficult to understand the narrative of this video. NU3 I could easily follow the flow of the narrative in this video Appel and Richter (2007) NU4 I had a hard time recognizing the narrative of this video. NU5 While watching, I understood why the narrative of video was proceed. New item for this study Narrative presence NP1 While watching the video, I lost track of time. Jang et al. (2016) Kim and Biocca (1997) NP2 While watching the video, I completely forgot that I was in the middle of the experiment. Busselle and Bilandzic (2009) NP3 While watching the video, I forgot my everyday concerns. New item for this study Emotional engagement EE1 While watching the video, I felt I knew exactly what the sharer was going through emotionally. Jang et al. (2016) Cohen (2001) EE2 I never really shared the emotions of the sharer. EE3 While watching the video, I could feel the emotions the sharer portrayed. Busselle and Bilandzic (2009) EE4 This video affected the emotional changes to me. New item for this study Variables Item Measurement References from Original Source Attentional focus AF1 While watching, I was fully concentrated on the narrative. Jang et al. (2016) Busselle and Bilandzic (2009) AF2 While watching, I was fully focused in the narrative. AF3 While watching, I had forgotten about other thoughts. New item for this study Narrative understanding NU1 This video’ narrative was naturally understandable. Jang et al. (2016) Busselle and Bilandzic (2009) NU2 It was difficult to understand the narrative of this video. NU3 I could easily follow the flow of the narrative in this video Appel and Richter (2007) NU4 I had a hard time recognizing the narrative of this video. NU5 While watching, I understood why the narrative of video was proceed. New item for this study Narrative presence NP1 While watching the video, I lost track of time. Jang et al. (2016) Kim and Biocca (1997) NP2 While watching the video, I completely forgot that I was in the middle of the experiment. Busselle and Bilandzic (2009) NP3 While watching the video, I forgot my everyday concerns. New item for this study Emotional engagement EE1 While watching the video, I felt I knew exactly what the sharer was going through emotionally. Jang et al. (2016) Cohen (2001) EE2 I never really shared the emotions of the sharer. EE3 While watching the video, I could feel the emotions the sharer portrayed. Busselle and Bilandzic (2009) EE4 This video affected the emotional changes to me. New item for this study AF, attentional focus; NU, narrative understanding; NP, narrative presence; EE, emotional engagement. We measured the active social interaction by counting actual usage behavior. In order to apply the effort level of action, we multiplied the usage count and the weighted value, in which ‘Like’ is 1.3, ‘Comment’ is 1.6, ‘Recall’ is 1.9, and ‘Give’ is 2.2. The minimum value for active social interaction is 0 if the participant does not use the social features of a video stimulus, whereas the maximum value for active social interaction is 7 if the participant uses all social features. If the participant uses ‘Like’ and ‘Give’, the count is the same, but the value (1, same) of active social interaction is different (1.3 and 2.2). 4.6. Experimental system and procedure Before the experiment, we explained to participants the concept of video recording through Google Glass and a drone, and provided them with samples of Google Glass and drone devices (all devices were developer versions) to aid in with their understanding of our experiment stimuli. In addition, participants were asked to view the heroine in the video stimuli as the sharer, and that she is an acquaintance. To provide an environment similar to an actual situation, we set the experimental system as multi-platform, including mobile phones and Google Glass. Participants could use their own mobile phones and installed the experimental application as a prototype video-sharing social media. They tried out this application to understand how our prototype video-sharing social media operated. In the prototype, we provided various video contents recorded by Google Glass and a drone. After letting participants try out our prototype, we conducted the main experiment in a controlled laboratory. The experimental system was automated and did not require researchers’ intervention. All descriptions of the experimental process were provided by a text-to-speech system. The researchers observed participant behaviors through two-way mirrors (Fig. 1). FIGURE 1. View largeDownload slide Experimental system. FIGURE 1. View largeDownload slide Experimental system. To conduct the survey relevant to the experiment, we applied Typeform (typeform.com) in our experimental system. Typeform is an online survey tool that provides an optimal mobile survey environment, as it is implemented in a mobile environment. Each participant was asked to access the experiment’s URL through our prototype application. We assigned ID numbers to each participant in order to give each one a new set of random combinations of nine video stimuli in the controlled laboratory environment. The environment was set via PC connected to Google Glass for watching video stimuli. All sets of video stimuli were assigned randomly. Participants were informed about the research purpose, description and experimental procedures through the automated text-to-speech system. The full-scale experimental procedure is as follows. In the beginning, participants watched the video stimuli using Google Glass; then, they were allowed to use four social features of the mobile experimental system. Participants had to write down why they did or did not use system features in response to a certain stimulus in our controlled experiment. 4.7. Measurement model We used PASW Statistics 18 to conduct analysis of variance and regression analysis to test the hypotheses. We also used the partial least squares method of structural equation modeling (Smart PLS 2.0) to test the measurement model. Reliability was measured with Cronbach’s alpha and composite reliability, both of which must exceed 0.70 (Fornell and Larcker, 1981). Table 5 indicates that all values exceeded the required minimum of 0.70. Convergent validity was verified when the standardized factor loading of each construct exceeds 0.70, with a t-value >1.96 as well as when the average variance extracted is >0.50 (Arnold and Reynolds, 2003). Table 5 shows that all standardized factor loadings exceed the required minimum of 0.70. In addition, all average variance extracted values exceed the required minimum of 0.50 (Table 4). TABLE 4. Convergent validity and reliability. Construct Items Factor t-value Composition Reliability AVE Cronbach’s α Attentional focus AF1 0.943 51.948 0.928 0.764 0.894 AF2 0.952 51.575 AF3 0.909 44.997 Narrative understanding NU1 0.841 27.385 0.930 0.726 0.906 NU2 0.802 13.535 NU3 0.920 57.546 NU4 0.872 27.948 NU5 0.822 15.215 Narrative presence NP1 0.920 43.334 0.913 0.725 0.871 NP2 0.896 25.023 NP3 0.920 32.171 Emotional engagement EE1 0.917 52.168 0.931 0.771 0.900 EE2 0.860 19.177 EE3 0.919 12.685 EE4 0.812 18.694 Like Lik 0.719 1 1 1 Comment Com 0.761 1 1 1 Recall Rec 0.798 1 1 1 Give Giv 0.780 1 1 1 Construct Items Factor t-value Composition Reliability AVE Cronbach’s α Attentional focus AF1 0.943 51.948 0.928 0.764 0.894 AF2 0.952 51.575 AF3 0.909 44.997 Narrative understanding NU1 0.841 27.385 0.930 0.726 0.906 NU2 0.802 13.535 NU3 0.920 57.546 NU4 0.872 27.948 NU5 0.822 15.215 Narrative presence NP1 0.920 43.334 0.913 0.725 0.871 NP2 0.896 25.023 NP3 0.920 32.171 Emotional engagement EE1 0.917 52.168 0.931 0.771 0.900 EE2 0.860 19.177 EE3 0.919 12.685 EE4 0.812 18.694 Like Lik 0.719 1 1 1 Comment Com 0.761 1 1 1 Recall Rec 0.798 1 1 1 Give Giv 0.780 1 1 1 TABLE 4. Convergent validity and reliability. Construct Items Factor t-value Composition Reliability AVE Cronbach’s α Attentional focus AF1 0.943 51.948 0.928 0.764 0.894 AF2 0.952 51.575 AF3 0.909 44.997 Narrative understanding NU1 0.841 27.385 0.930 0.726 0.906 NU2 0.802 13.535 NU3 0.920 57.546 NU4 0.872 27.948 NU5 0.822 15.215 Narrative presence NP1 0.920 43.334 0.913 0.725 0.871 NP2 0.896 25.023 NP3 0.920 32.171 Emotional engagement EE1 0.917 52.168 0.931 0.771 0.900 EE2 0.860 19.177 EE3 0.919 12.685 EE4 0.812 18.694 Like Lik 0.719 1 1 1 Comment Com 0.761 1 1 1 Recall Rec 0.798 1 1 1 Give Giv 0.780 1 1 1 Construct Items Factor t-value Composition Reliability AVE Cronbach’s α Attentional focus AF1 0.943 51.948 0.928 0.764 0.894 AF2 0.952 51.575 AF3 0.909 44.997 Narrative understanding NU1 0.841 27.385 0.930 0.726 0.906 NU2 0.802 13.535 NU3 0.920 57.546 NU4 0.872 27.948 NU5 0.822 15.215 Narrative presence NP1 0.920 43.334 0.913 0.725 0.871 NP2 0.896 25.023 NP3 0.920 32.171 Emotional engagement EE1 0.917 52.168 0.931 0.771 0.900 EE2 0.860 19.177 EE3 0.919 12.685 EE4 0.812 18.694 Like Lik 0.719 1 1 1 Comment Com 0.761 1 1 1 Recall Rec 0.798 1 1 1 Give Giv 0.780 1 1 1 Discriminant validity was measured with the criterion that the square root of the average variance extracted for each construct should be greater than the corresponding correlation coefficients (Fornell and Larcker, 1981). All square roots of each corresponding average variance extracted exceeded the corresponding correlation coefficients, as shown in Table 5. The results indicate that the factor loadings of the items of each construct show correlation, meaning that the items of each construct explain their corresponding constructs in a statistically significant manner. In addition, the factor loading of each item on its construct was higher than those on the other constructs. This fact indicates that the items being part of their corresponding constructs is statistically supported. TABLE 5. Discriminant validity. Attentional focus Narrative understanding Narrative presence Emotional engagement Like Comment Recall Give Attentional focus 0.874 Narrative understanding 0.522 0.852 Narrative presence 0.578 0.454 0.851 Emotional engagement 0.586 0.513 0.605 0.878 Like 0.412 0.371 0.418 0.479 1 Comment 0.332 0.224 0.324 0.376 0.460 1 Recall 0.351 0.275 0.364 0.423 0.444 0.318 1 Give 0.276 0.279 0.305 0.383 0.453 0.260 0.545 1 Attentional focus Narrative understanding Narrative presence Emotional engagement Like Comment Recall Give Attentional focus 0.874 Narrative understanding 0.522 0.852 Narrative presence 0.578 0.454 0.851 Emotional engagement 0.586 0.513 0.605 0.878 Like 0.412 0.371 0.418 0.479 1 Comment 0.332 0.224 0.324 0.376 0.460 1 Recall 0.351 0.275 0.364 0.423 0.444 0.318 1 Give 0.276 0.279 0.305 0.383 0.453 0.260 0.545 1 The boldface figures on the diagonal are the square root of AVE. TABLE 5. Discriminant validity. Attentional focus Narrative understanding Narrative presence Emotional engagement Like Comment Recall Give Attentional focus 0.874 Narrative understanding 0.522 0.852 Narrative presence 0.578 0.454 0.851 Emotional engagement 0.586 0.513 0.605 0.878 Like 0.412 0.371 0.418 0.479 1 Comment 0.332 0.224 0.324 0.376 0.460 1 Recall 0.351 0.275 0.364 0.423 0.444 0.318 1 Give 0.276 0.279 0.305 0.383 0.453 0.260 0.545 1 Attentional focus Narrative understanding Narrative presence Emotional engagement Like Comment Recall Give Attentional focus 0.874 Narrative understanding 0.522 0.852 Narrative presence 0.578 0.454 0.851 Emotional engagement 0.586 0.513 0.605 0.878 Like 0.412 0.371 0.418 0.479 1 Comment 0.332 0.224 0.324 0.376 0.460 1 Recall 0.351 0.275 0.364 0.423 0.444 0.318 1 Give 0.276 0.279 0.305 0.383 0.453 0.260 0.545 1 The boldface figures on the diagonal are the square root of AVE. Cross-loadings between constructs are shown in Table 6. Discriminant validity was measured with the criterion that the square root of the average variance extracted should be greater than the corresponding correlation coefficients. All square roots of each corresponding average variance extracted exceeded the corresponding correlation coefficients. TABLE 6. Cross-loadings. AF NU NP EE LIK COM REC GIV Attentional focus AF1 0.915 0.545 0.674 0.555 0.413 0.306 0.349 0.287 AF2 0.922 0.517 0.664 0.543 0.411 0.274 0.342 0.267 AF3 0.917 0.438 0.761 0.541 0.352 0.289 0.318 0.23 Narrative understanding NU1 0.515 0.843 0.485 0.494 0.379 0.253 0.304 0.263 NU2 0.352 0.800 0.265 0.354 0.244 0.150 0.143 0.173 NU3 0.528 0.920 0.449 0.476 0.365 0.192 0.269 0.288 NU4 0.413 0.870 0.308 0.380 0.244 0.149 0.237 0.213 NU5 0.386 0.823 0.389 0.457 0.321 0.197 0.193 0.235 Narrative presence NP1 0.768 0.445 0.908 0.596 0.393 0.300 0.371 0.288 NP2 0.652 0.375 0.888 0.497 0.381 0.319 0.308 0.23 NP3 0.726 0.374 0.885 0.509 0.281 0.233 0.285 0.225 Emotional engagement EE1 0.533 0.475 0.526 0.916 0.391 0.339 0.344 0.309 EE2 0.428 0.418 0.424 0.859 0.360 0.287 0.339 0.316 EE3 0.505 0.466 0.498 0.918 0.440 0.342 0.346 0.327 EE4 0.578 0.436 0.657 0.813 0.483 0.345 0.448 0.389 Like LIK 0.412 0.371 0.418 0.479 1 0.460 0.444 0.453 Comment COM 0.332 0.224 0.324 0.376 0.460 1 0.318 0.260 Recall REC 0.351 0.275 0.364 0.423 0.444 0.318 1 0.545 Give GIV 0.276 0.279 0.305 0.383 0.453 0.260 0.545 1 AF NU NP EE LIK COM REC GIV Attentional focus AF1 0.915 0.545 0.674 0.555 0.413 0.306 0.349 0.287 AF2 0.922 0.517 0.664 0.543 0.411 0.274 0.342 0.267 AF3 0.917 0.438 0.761 0.541 0.352 0.289 0.318 0.23 Narrative understanding NU1 0.515 0.843 0.485 0.494 0.379 0.253 0.304 0.263 NU2 0.352 0.800 0.265 0.354 0.244 0.150 0.143 0.173 NU3 0.528 0.920 0.449 0.476 0.365 0.192 0.269 0.288 NU4 0.413 0.870 0.308 0.380 0.244 0.149 0.237 0.213 NU5 0.386 0.823 0.389 0.457 0.321 0.197 0.193 0.235 Narrative presence NP1 0.768 0.445 0.908 0.596 0.393 0.300 0.371 0.288 NP2 0.652 0.375 0.888 0.497 0.381 0.319 0.308 0.23 NP3 0.726 0.374 0.885 0.509 0.281 0.233 0.285 0.225 Emotional engagement EE1 0.533 0.475 0.526 0.916 0.391 0.339 0.344 0.309 EE2 0.428 0.418 0.424 0.859 0.360 0.287 0.339 0.316 EE3 0.505 0.466 0.498 0.918 0.440 0.342 0.346 0.327 EE4 0.578 0.436 0.657 0.813 0.483 0.345 0.448 0.389 Like LIK 0.412 0.371 0.418 0.479 1 0.460 0.444 0.453 Comment COM 0.332 0.224 0.324 0.376 0.460 1 0.318 0.260 Recall REC 0.351 0.275 0.364 0.423 0.444 0.318 1 0.545 Give GIV 0.276 0.279 0.305 0.383 0.453 0.260 0.545 1 AF, attentional focus; NU, narrative understanding; NP, narrative presence; EE, emotional engagement; LIK, like; COM, comment; REC, recall; GIV, give. Bold italic figures indicate statistically significant values. TABLE 6. Cross-loadings. AF NU NP EE LIK COM REC GIV Attentional focus AF1 0.915 0.545 0.674 0.555 0.413 0.306 0.349 0.287 AF2 0.922 0.517 0.664 0.543 0.411 0.274 0.342 0.267 AF3 0.917 0.438 0.761 0.541 0.352 0.289 0.318 0.23 Narrative understanding NU1 0.515 0.843 0.485 0.494 0.379 0.253 0.304 0.263 NU2 0.352 0.800 0.265 0.354 0.244 0.150 0.143 0.173 NU3 0.528 0.920 0.449 0.476 0.365 0.192 0.269 0.288 NU4 0.413 0.870 0.308 0.380 0.244 0.149 0.237 0.213 NU5 0.386 0.823 0.389 0.457 0.321 0.197 0.193 0.235 Narrative presence NP1 0.768 0.445 0.908 0.596 0.393 0.300 0.371 0.288 NP2 0.652 0.375 0.888 0.497 0.381 0.319 0.308 0.23 NP3 0.726 0.374 0.885 0.509 0.281 0.233 0.285 0.225 Emotional engagement EE1 0.533 0.475 0.526 0.916 0.391 0.339 0.344 0.309 EE2 0.428 0.418 0.424 0.859 0.360 0.287 0.339 0.316 EE3 0.505 0.466 0.498 0.918 0.440 0.342 0.346 0.327 EE4 0.578 0.436 0.657 0.813 0.483 0.345 0.448 0.389 Like LIK 0.412 0.371 0.418 0.479 1 0.460 0.444 0.453 Comment COM 0.332 0.224 0.324 0.376 0.460 1 0.318 0.260 Recall REC 0.351 0.275 0.364 0.423 0.444 0.318 1 0.545 Give GIV 0.276 0.279 0.305 0.383 0.453 0.260 0.545 1 AF NU NP EE LIK COM REC GIV Attentional focus AF1 0.915 0.545 0.674 0.555 0.413 0.306 0.349 0.287 AF2 0.922 0.517 0.664 0.543 0.411 0.274 0.342 0.267 AF3 0.917 0.438 0.761 0.541 0.352 0.289 0.318 0.23 Narrative understanding NU1 0.515 0.843 0.485 0.494 0.379 0.253 0.304 0.263 NU2 0.352 0.800 0.265 0.354 0.244 0.150 0.143 0.173 NU3 0.528 0.920 0.449 0.476 0.365 0.192 0.269 0.288 NU4 0.413 0.870 0.308 0.380 0.244 0.149 0.237 0.213 NU5 0.386 0.823 0.389 0.457 0.321 0.197 0.193 0.235 Narrative presence NP1 0.768 0.445 0.908 0.596 0.393 0.300 0.371 0.288 NP2 0.652 0.375 0.888 0.497 0.381 0.319 0.308 0.23 NP3 0.726 0.374 0.885 0.509 0.281 0.233 0.285 0.225 Emotional engagement EE1 0.533 0.475 0.526 0.916 0.391 0.339 0.344 0.309 EE2 0.428 0.418 0.424 0.859 0.360 0.287 0.339 0.316 EE3 0.505 0.466 0.498 0.918 0.440 0.342 0.346 0.327 EE4 0.578 0.436 0.657 0.813 0.483 0.345 0.448 0.389 Like LIK 0.412 0.371 0.418 0.479 1 0.460 0.444 0.453 Comment COM 0.332 0.224 0.324 0.376 0.460 1 0.318 0.260 Recall REC 0.351 0.275 0.364 0.423 0.444 0.318 1 0.545 Give GIV 0.276 0.279 0.305 0.383 0.453 0.260 0.545 1 AF, attentional focus; NU, narrative understanding; NP, narrative presence; EE, emotional engagement; LIK, like; COM, comment; REC, recall; GIV, give. Bold italic figures indicate statistically significant values. In this research, narrative engagement is the second-order construct. The method of analyzing a second-order construct involves including at least three first-order factors that are measured by more than two items for each construct. In addition, all statistically supported values have to exceed each criterion. Narrative engagement has four first-order constructs (attentional focus, narrative understanding, narrative presence and emotional engagement), and each of them has more than two items. We verified the statistical values for measurement items which fulfilled the necessary conditions for analyzing second-order constructs. All items used for generating the second-order construct had to be converted into a single value. For this, we used latent variable scores obtained from partial least squares analysis (Wilson and Henseler, 2007). Additional data on statistical considerations for the second-order construct are shown in Table 7. TABLE 7. Additional data for analyzing NE as the second-order construct. Second-order construct Items First-order construct First-order items t-value Narrative engagement NE1 Attentional focus AF1 24.768 NE2 AF2 22.124 NE3 AF3 20.062 NE4 Narrative understanding NU1 12.396 NE5 NU2 6.361 NE6 NU3 12.096 NE7 NU4 8.030 NE8 NU5 7.902 NE9 Narrative presence NP1 19.480 NE10 NP2 11.789 NE11 NP3 14.089 NE12 Emotional engagement EE1 14.008 NE13 EE2 8.921 NE14 EE3 12.685 NE15 EE4 16.400 Second-order construct Items First-order construct First-order items t-value Narrative engagement NE1 Attentional focus AF1 24.768 NE2 AF2 22.124 NE3 AF3 20.062 NE4 Narrative understanding NU1 12.396 NE5 NU2 6.361 NE6 NU3 12.096 NE7 NU4 8.030 NE8 NU5 7.902 NE9 Narrative presence NP1 19.480 NE10 NP2 11.789 NE11 NP3 14.089 NE12 Emotional engagement EE1 14.008 NE13 EE2 8.921 NE14 EE3 12.685 NE15 EE4 16.400 NE, narrative engagement; AF, attentional focus; NU, narrative understanding; NP, narrative presence; EE, emotional engagement. TABLE 7. Additional data for analyzing NE as the second-order construct. Second-order construct Items First-order construct First-order items t-value Narrative engagement NE1 Attentional focus AF1 24.768 NE2 AF2 22.124 NE3 AF3 20.062 NE4 Narrative understanding NU1 12.396 NE5 NU2 6.361 NE6 NU3 12.096 NE7 NU4 8.030 NE8 NU5 7.902 NE9 Narrative presence NP1 19.480 NE10 NP2 11.789 NE11 NP3 14.089 NE12 Emotional engagement EE1 14.008 NE13 EE2 8.921 NE14 EE3 12.685 NE15 EE4 16.400 Second-order construct Items First-order construct First-order items t-value Narrative engagement NE1 Attentional focus AF1 24.768 NE2 AF2 22.124 NE3 AF3 20.062 NE4 Narrative understanding NU1 12.396 NE5 NU2 6.361 NE6 NU3 12.096 NE7 NU4 8.030 NE8 NU5 7.902 NE9 Narrative presence NP1 19.480 NE10 NP2 11.789 NE11 NP3 14.089 NE12 Emotional engagement EE1 14.008 NE13 EE2 8.921 NE14 EE3 12.685 NE15 EE4 16.400 NE, narrative engagement; AF, attentional focus; NU, narrative understanding; NP, narrative presence; EE, emotional engagement. 4.8. Study 1 First, we conducted a 2 × 2, i.e. (narrative focus: foveal vision (F)/peripheral vision (P)) × (narrative perspective: first person (1)/third person (3)), within-subject experiment (Table 1). The making of mixed conditions, including vision and perspective, were decided by the results of participants’ opinions. We did not know what order of mixed vision would encourage participants to engage with a video narrative without a sense of difference (FP or PF). Mixed-person stimuli faced the same problem as well (13 or 31). Furthermore, the mixed condition between vision and perspective can be made up of four stimuli cases (1F-3P/3F-1P/1P-3F/3P-1F, three times this pattern per video). Therefore, we first verified the single condition effect of narrative focus and perspective on the detailed narrative engagement, including attentional focus, narrative understanding, narrative presence and emotional engagement. 4.8.1. Hypothesis testing To verify the Hypotheses 1–10, we conducted multivariate analysis of variance with attentional focus, narrative understanding, narrative presence and emotional engagement. We also converted four sub-constructs into narrative engagement. Then, we analyzed Hypotheses 5 and 10. Effects of single narrative focus Hypothesis 1 assumed that all video scenes having peripheral vision had more positive effects on attentional focus than those having only foveal vision. The results showed a significant main effect of narrative focus on attentional focus (F(1, 206) = 8.391, P < 0.05). Hypothesis 2 assumed that all video scenes having a peripheral vision had more positive effects on narrative understanding than those having only foveal vision. However, the results did not show a main effect of narrative focus on narrative understanding (F(1, 206) = 1.979, P > 0.05). Hypothesis 3 assumed that all video scenes having a peripheral vision had more positive effects on narrative presence than those having only foveal vision. The results showed a significant main effect of narrative focus on narrative presence (F(1, 206) = 12.659, P < 0.001). Hypothesis 4 assumed that all video scenes having a peripheral vision had more positive effects on emotional engagement than those having only foveal vision. The results showed a significant main effect of narrative focus on emotional engagement (F(1, 206) = 5.462, P < 0.05). Hypothesis 5 assumed that all video scenes having a peripheral vision had more positive effects on narrative engagement than those having only foveal vision. The results showed a significant main effect of narrative focus on narrative engagement (F (1, 206) = 9.783, P < 0.05). Overall, Hypotheses 1, 3 and 4 were supported, but Hypothesis 2 was not supported. Hypothesis 5, as the sum total of the effects of four sub-constructs was supported (Fig. 2). FIGURE 2. View largeDownload slide The effect of single narrative focus on four sub-constructs of narrative engagement. *P < 0.05, **P < 0.001. FIGURE 2. View largeDownload slide The effect of single narrative focus on four sub-constructs of narrative engagement. *P < 0.05, **P < 0.001. Post-hoc for analyzing the tendency of active social interaction by the effect of narrative focus All social features, including ‘Like’, ‘Comment’, ‘Recall’ and ‘Give’, showed statistically significant difference between foveal and peripheral vision: ‘Like’, F(1, 206) = 14.187, P < 0.001; ‘Comment’, F(1, 206) = 5.063, P < 0.05; ‘Recall’, F(1, 206) = 21.508, P < 0.001; and ‘Give’, F(1, 206) = 9.524, P < 0.05. The results showed that the value of active social interaction is almost twice more on the condition of peripheral vision without ‘Comment’. With this result, the cumulative count of social interaction showed statistically significant difference. Peripheral vision facilitated more social behavior (count = 361.7) than foveal vision (count = 201.3) (F(1, 206) = 23.325) (Figs 3 and 4). FIGURE 3. View largeDownload slide The effect of single narrative focus on four features of social interaction in video. *P< 0.05, **P < 0.001. FIGURE 3. View largeDownload slide The effect of single narrative focus on four features of social interaction in video. *P< 0.05, **P < 0.001. FIGURE 4. View largeDownload slide The effect of single narrative focus on the amount of social interaction in video. **P < 0.001. FIGURE 4. View largeDownload slide The effect of single narrative focus on the amount of social interaction in video. **P < 0.001. Effects of single narrative perspective Hypothesis 6 assumed that all video scenes having the first-person perspective had more positive effects on attentional focus than those having the third-person perspective. The results showed a significant main effect of narrative perspective on attentional focus (F(1, 206) = 8.581, P < 0.05) but the hypothesis was not supported. Hypothesis 7 assumed that all video scenes having the third-person perspective had more positive effects on narrative understanding than those having the first-person perspective. The results showed a significant main effect of narrative perspective on narrative understanding (F(1, 206) = 14.751, P < 0.001). Hypothesis 8 assumed that all video scenes having the third-person perspective had more positive effects on narrative presence than those having the first-person perspective. However, the results did not show a main effect of narrative perspective on narrative presence (F(1, 206) = 3.795, P > 0.05). Hypothesis 9 assumed that all video scenes having the third-person perspective had more positive effects on emotional engagement than those having the first-person perspective. However, the results did not show a main effect of narrative perspective on emotional engagement (F(1, 206) = 2.087, P > 0.05). Hypothesis 10 assumed that all video scenes having the third-person perspective had more positive effects on narrative engagement than those having the first-person perspective. The results showed a significant main effect of narrative perspective on narrative engagement (F(1, 206) = 5.502, P < 0.05). Overall, only Hypothesis 7 was supported, while Hypotheses 6, 8, and 9 were not supported. Hypothesis 10, as the sum total of the effects of four sub-constructs, was supported (Fig. 5). FIGURE 5. View largeDownload slide The effect of single narrative perspective on four sub-constructs of narrative engagement. *P < 0.05, **P < 0.001. FIGURE 5. View largeDownload slide The effect of single narrative perspective on four sub-constructs of narrative engagement. *P < 0.05, **P < 0.001. Post-hoc for analyzing the tendency of active social interaction by the effect of narrative perspective All social features, including ‘Like’, ‘Comment’, ‘Recall’ and ‘Give’, did not show statistically significant difference: ‘Like’, F(1, 206) = 0.709, P > 0.05; ‘Comment’, F (1, 206) = 2.818, P > 0.05; ‘Recall’, F (1, 206) = 0.347, P > 0.05; and ‘Give’, F (1, 206) = 0.823, P > 0.05. The results showed that the value of active social interaction is less changed by the different perspective conditions of video stimuli. With this result, the cumulative count of social interaction showed no statistically significant difference. The social behavior counts for the third-person (count = 286.2) was almost the same as that for the first-person (count = 272.8) (F(1, 206) = 0.072) (Figs 6 and 7). FIGURE 6. View largeDownload slide The effect of single narrative perspective on four features of social interaction in video. FIGURE 6. View largeDownload slide The effect of single narrative perspective on four features of social interaction in video. FIGURE 7. View largeDownload slide The effect of single narrative perspective on the amount of social interaction in video. FIGURE 7. View largeDownload slide The effect of single narrative perspective on the amount of social interaction in video. The results of Hypotheses 1–10 testing are shown in Table 8 as the detailed data. TABLE 8. Results of hypothesis testing. IV DV SS dF MS F η2 Sig. Result NF AF 9.233 1 9.233 8.391 .039 .004 H1 Supported NU 2.089 1 2.089 1.97 .010 .161 H2 Not supported NP 11.614 1 11.614 12.659 .058 .000 H3 Supported EE 4.970 1 4.970 5.462 .026 .020 H4 Supported NE 9.398 1 9.398 9.783 .047 .002 H5 Supported NPR AF 9.433 1 9.433 8.581 .040 .004 H6 Not supported NU 14.667 1 14.667 14.751 .067 .000 H7 Supported NP 3.629 1 3.629 3.795 .018 .053 H8 Not supported EE 1.930 1 1.930 2.087 .010 .150 H9 Not supported NE 5.201 1 5.201 5.502 .026 .020 H10 Supported IV DV SS dF MS F η2 Sig. Result NF AF 9.233 1 9.233 8.391 .039 .004 H1 Supported NU 2.089 1 2.089 1.97 .010 .161 H2 Not supported NP 11.614 1 11.614 12.659 .058 .000 H3 Supported EE 4.970 1 4.970 5.462 .026 .020 H4 Supported NE 9.398 1 9.398 9.783 .047 .002 H5 Supported NPR AF 9.433 1 9.433 8.581 .040 .004 H6 Not supported NU 14.667 1 14.667 14.751 .067 .000 H7 Supported NP 3.629 1 3.629 3.795 .018 .053 H8 Not supported EE 1.930 1 1.930 2.087 .010 .150 H9 Not supported NE 5.201 1 5.201 5.502 .026 .020 H10 Supported IV, independent variable; DV, dependent variable; AF, attentional focus; NU, narrative understanding; NP, narrative presence; EE, emotional engagement; NE, narrative engagement; SS, sum of squares; dF, degree of freedom; MS, mean square; F, F-ratio; Sig., significance; η2, eta squared; B, unstandardized coefficients; β, beta coefficient; VIF, variance inflation factor. Bold figures indicate statistically significant values. TABLE 8. Results of hypothesis testing. IV DV SS dF MS F η2 Sig. Result NF AF 9.233 1 9.233 8.391 .039 .004 H1 Supported NU 2.089 1 2.089 1.97 .010 .161 H2 Not supported NP 11.614 1 11.614 12.659 .058 .000 H3 Supported EE 4.970 1 4.970 5.462 .026 .020 H4 Supported NE 9.398 1 9.398 9.783 .047 .002 H5 Supported NPR AF 9.433 1 9.433 8.581 .040 .004 H6 Not supported NU 14.667 1 14.667 14.751 .067 .000 H7 Supported NP 3.629 1 3.629 3.795 .018 .053 H8 Not supported EE 1.930 1 1.930 2.087 .010 .150 H9 Not supported NE 5.201 1 5.201 5.502 .026 .020 H10 Supported IV DV SS dF MS F η2 Sig. Result NF AF 9.233 1 9.233 8.391 .039 .004 H1 Supported NU 2.089 1 2.089 1.97 .010 .161 H2 Not supported NP 11.614 1 11.614 12.659 .058 .000 H3 Supported EE 4.970 1 4.970 5.462 .026 .020 H4 Supported NE 9.398 1 9.398 9.783 .047 .002 H5 Supported NPR AF 9.433 1 9.433 8.581 .040 .004 H6 Not supported NU 14.667 1 14.667 14.751 .067 .000 H7 Supported NP 3.629 1 3.629 3.795 .018 .053 H8 Not supported EE 1.930 1 1.930 2.087 .010 .150 H9 Not supported NE 5.201 1 5.201 5.502 .026 .020 H10 Supported IV, independent variable; DV, dependent variable; AF, attentional focus; NU, narrative understanding; NP, narrative presence; EE, emotional engagement; NE, narrative engagement; SS, sum of squares; dF, degree of freedom; MS, mean square; F, F-ratio; Sig., significance; η2, eta squared; B, unstandardized coefficients; β, beta coefficient; VIF, variance inflation factor. Bold figures indicate statistically significant values. 4.9. Study 2 In study 2, we conducted a 2 × 2, i.e. (narrative focus: single vision (F or P)/mixed vision (F–P or P–F)) × (narrative perspective: single person (1 or 3)/mixed-person (1–3 or 3-1)), within-subject experiment (Table 1). To verify Hypotheses 11 and 12, we provided mixed conditions of narrative focus and perspective to our participants. The method for making the mixed condition, including vision and perspective, was decided by the results of study 1 and participants’ opinions. Participants reported that the scene with 1F condition felt more natural than 1P, and the scene with 3P felt more natural than 3F. Therefore, we set the criteria for making the mixed condition video stimuli: the first-person combined with foveal vision, the third-person combined with peripheral vision, and the first video scene started through the natural combination (1F or 3P). According to these criteria, we set the mixed condition video stimuli as 1F-3F-1F-3F-1F-3F, 3P-1P-3P-1P-3P-1P, 1F-1P-1F-1P-1F-1P and 3P-3F-3P-3F-3P-3F. In addition, we made two double mixed condition (mixed vision × mixed-person) video stimuli, including 1F-3P (repeated three times) and 3P-1F (repeated three times). Therefore, we finally set the six mixed condition videos, including two double mixed conditions. 4.9.1. Single vs. Mixed vision and Single vs. Mixed point-of-view effect on narrative engagement To verify Hypotheses 11 and 12, we conducted a planned contrast analysis with narrative engagement. Hypothesis 11 assumed that mixed vision videos having foveal and peripheral conditions had more positive effects on narrative engagement than those having only single vision. The results showed a statistically significant effect of narrative focus on narrative engagement (t = −4.091, df = 516, P = 0.000). Hypothesis 12 assumed that mixed point-of-view videos (having the first and third-person perspectives) had more positive effects on narrative engagement than those having only a single perspective. The results showed a statistically significant effect of narrative perspective on narrative engagement (t = −3.701, df = 516, P = 0.000). To test Hypothesis 13 as an interaction effect, we conducted two-way analysis of variance. The interaction effect between two mixed conditions on narrative engagement was significant (F(1, 206) = 8.137, P < 0.05). This means that the effect of the mixed point-of-view on narrative engagement will strengthen as the vision of the video is mixed (Figure 8). FIGURE 8. View largeDownload slide The interaction effect between two mixed conditions (mixed-point of view × mixed vision) on narrative engagement. FIGURE 8. View largeDownload slide The interaction effect between two mixed conditions (mixed-point of view × mixed vision) on narrative engagement. 4.9.2. Post-hoc for analyzing the tendency of active social interaction between single and mixed vision or single and mixed point-of-view All social features, including ‘Like’, ‘Comment’, ‘Recall’ and ‘Give’ showed statistically significant difference between single and mixed vision: ‘Like’, F(1, 206) = 0.709, P > 0.05, ‘Comment’, F(1, 206) = 2.818, P > 0.05, ‘Recall’, F(1, 206) = 0.347, P > 0.05, and ‘Give’, F(1, 206) = 0.823, P > 0.05. The results showed that the active social interaction values changed less on the different perspective conditions of video stimuli. With this result, the cumulative count of social interaction showed no statistically significant difference. The social behavior count in the third-person (count = 286.2) was almost the same as that in the first-person (count = 272.8) (F(1, 206) = 0.072). 5. CONCLUSION AND DISCUSSION The primary goal of this study is to verify how the scene formats in snapshot videos affect the four detailed narrative engagement factors: attentional focus, narrative understanding, narrative presence and emotional engagement. In addition, we assumed that the method of creating narrative structure through scene formats is important to facilitate viewers’ active social interaction. To verify this theoretically and practically, scene formats related to narrative structure were considered not only by narrative theory, including the concepts of narrative focus and narrative perspective, but also through the practical possibility of technical realization, including ‘bokeh’ effects and perspective taking. Using Google Glass and a drone, we took snapshot videos inspired by our scenario of a person’s daily experiences. We edited fourteen narratives for our experiment using snapshot videos. In terms of narrative focus, the results indicated that peripheral vision leads to more attentional focus, narrative presence and emotional engagement than foveal vision. Narrative understanding was not differentiated between peripheral and foveal vision. In addition, we found that peripheral vision effects on attentional focus, narrative presence and emotional engagement facilitated more active social interaction among viewers than foveal vision. Furthermore, we found that mixed vision videos lead to more narrative engagement than single vision videos. In terms of narrative perspective, the results indicated that the third-person perspective leads to more attentional focus and narrative understanding than the first-person perspective. Narrative presence and emotional engagement were not differentiated between the first- and third-person perspectives. In addition, we found that differences in perspective between the first and third person did not occur during active social interaction. Meanwhile, we found that a mixed-person videos lead to more narrative engagement than a single-person video. Based on the study results, three interesting issues are worth discussing in more detail. Firstly, as a narrative structure, the effect of narrative focus fulfills the three conditions of narrative engagement. First, peripheral vision facilitates more attentional engagement (i.e. attentional focus) than foveal vision. It is important to note that videos having vivid scenes can create the environment for attentional focus. Second, viewers feel more narrative presence with peripheral vision than foveal vision. This means that delivering the sharer’s experience by video requires the vivid expression of scenes to provide dynamics to viewers. Finally, emotional engagement is triggered more by peripheral vision than foveal vision. Contrary to foveal vision, which concentrates on the central information of the event, peripheral vision has emotional information for triggering emotional engagement. It is important to note that these structural effects of detailed narrative engagement affect active social interaction. Secondly, as a narrative structure, the effect of narrative perspective fulfills the two conditions of narrative engagement. First, the third-person perspective facilitates more attentional engagement (i.e. attentional focus) than the first-person perspective. It is important to note that videos showing the whole situation encourage viewers to engage with the sharer’s narrative more effectively. Second, viewers understand the narrative better in the third person rather than the first person. This means that viewers can understand the sharer’s experience better through the third person than the first person. In addition, it is important to note that these structural effects on detailed narrative engagement affect active social interaction. Thirdly, viewers try to interact with others while engaging with the sharers’ narrative through the narrative structure of snapshot videos. In particular, the leverage of narrative presence and emotional engagement makes a difference in active social interaction. This result implies that viewers communicate with others, using snapshot videos as the medium of active social interaction, beyond the simplified interaction of clicking ‘Like’ or writing a ‘Comment’. 6. LIMITATIONS AND IMPLICATIONS 6.1. Limitations and further study In our study, part of the experimental system was showing video stimuli with Google Glass. This can provide participants with real experience in using new video-sharing social media with wearable devices. However, our experimental system, including the effects of narrative focus and narrative perspective, felt unnatural to participants. In order to aid in participants’ adaptation, we gave them sufficient time to become familiar with Google Glass. However, they felt somewhat dizzy when watching our video stimuli with Google Glass. Furthermore, aside from wearing Google Glass to watch our video stimuli, they also participated in our survey and used social features step by step. These processes may increase the cognitive load of participants and cause confusion. Therefore, future works should adopt a more natural experimental system. Nevertheless, this study has various theoretical and practical implications. 6.2. Theoretical implications Theoretically, this study verified the effects of four detailed concepts of narrative engagement (attentional focus, narrative understanding, narrative presence and emotional engagement) through empirical data. In particular, we examined how social interaction actively depended on the tendency of detailed engaging effects. Our prior study provided a theoretical framework on narrative engagement in terms of the temporal format of snapshot videos. In addition, we already verified the effects of narrative engagement on social interaction. However, our prior study had limitations in explaining how people try to interact actively with others in terms of the quality of socially interactive behavior. Accordingly, we extended our previous study by using system features, applying the concept of effort level of action in which ‘Like’ is relatively lighter and ‘Give’ is relatively heavier in terms of level of social interaction. In particular, we set concepts of narrative structure by summarizing past studies on narratives and related video effects. From the perspective of human–computer interaction research, two different effects related to narrative structure can be described, as mentioned earlier. From the perspective of narrative structure, we logically understood and applied 2D constructs to explain the structural narrative effects of the sharer’s experience: structural range and structural sight of the narrative. This result indicates theoretically that people are affected by changes in narrative structure during active social interaction. 6.3. Practical implications Practically, this research points to the effects of two structural changes, including narrative focus and narrative perspective, on active social interaction. First, in order to increase narrative engagement, we believe that the peripheral focusing effect should be considered. We know that creating this scene structure has a high impact on attentional focus, narrative presence and emotional engagement. This means that designers and developers who want to create an immersive video-sharing social media application or platform can focus on making an automated scene filter to facilitate users’ attention, realistic feelings and emotional assimilation. For example, designers can provide a sharpened filter for making snapshot videos more vivid to viewers. Instagram has provided similar features to make scenes clearer using the ‘sharpen’ filter. In addition, in order to increase narrative engagement, we believe that the third-person point of view should be considered. We know that creating this scene structure has a high impact on attentional focus and narrative understanding. This means that designers and developers who want to make an immersive video-sharing social medium can focus on the use of recording devices to facilitate users’ attention to, and understanding of, narratives. For example, developers can consider a wearable drone to help sharers record and share selfie snapshot videos from the third-person point of view. Nixie, a wearable drone device, can take photos or videos, and then flies back to users automatically. It provides a new perspective in which viewers can pay attention to what the sharer does. For this reason, we can assume that social media like Facebook should consider a third-person point of view. Because the dynamics of timelines are too fast to focus on and understand a specific video, it is necessary to attract viewers’ interest without the content of the video narrative. In addition, this research points to the effects of mixed structural changes on narrative engagement. In order to increase narrative engagement, we know that the mixed condition of narrative focus should be considered. Contrary to the single condition of narrative focus, the mixed condition provides viewers with room to imagine the sharer’s situation for themselves. Therefore, developers who want to make a video editing application can consider automatic mixed scene filters that provide scene effects automatically in order to edit snapshot videos more attractively. For example, we can create the effect of foveal vision by using eye-tracking data that can provide the sharer’s focusing spot. If the sharer’s smartglass is capable of performing eye-tracking, the mixed scene vision of a snapshot video can be edited automatically using the sharer’s eye-tracking data to apply the scene filter. For this reason, we suggest that social media like Snapchat consider the vision effects when shared video is recorded in a short time. If Spectacles, the smartglasses launched by Snapchat, is capable of collecting eye-tracking data, it will be an effective way of editing video scenes automatically by adopting the concept of narrative focus. Furthermore, we know that the mixed condition of narrative perspective should be considered. Contrary to the single condition, the mixed condition provides viewers with room to interpret the sharer’s situation for themselves. Therefore, developers who want to make video recording devices for the Internet of Things environment can consider the automatic mixed scene perspective to provide the scene perspective automatically by collaborative video-taking environment. For example, we can take the third-person point of view by using others’ video data. A location-based cloud video-sharing platform such as Vyclone can be provided to make the sharer’s own narrative by using one’s own and others’ video data simultaneously, and the mixed scene perspective of the snapshot video can be edited automatically. Finally, we know that new social features are needed for interacting with others more deeply. People are already used to clicking the ‘Like’ button to express their empathy. However, viewers may want to interact more deeply and directly with others because of their desire to share their own dramatic experiences as a result of the sharer’s snapshot video. Therefore, designers should consider deeper expressive social features for video-sharing social media. For example, we can reply to the timeline of shared snapshot videos directly and then share it with others. In addition, the viewer can immediately talk about the sharer’s snapshot video with others without making any posts. ACKNOWLEDGEMENTS The authors thank Minji Kim who kindly proofread the entire draft and all of our research team members. Be a good storyteller with wearable devices and drones! FUNDING This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (NRF-2016R1D1A1B02015987). REFERENCES Abolafia , M.Y. ( 2010 ) Narrative construction as sensemaking: how a central bank thinks . Organ. Stud. , 31 , 349 – 367 . Google Scholar CrossRef Search ADS Aguirre , , G.K. and D'Esposito , M. ( 1999 ) Topographical disorientation: a synthesis and taxonomy . Brain , 122 , 1613 – 1628 . Google Scholar CrossRef Search ADS PubMed Alfano , P.L. and Michel , G.F. ( 1990 ) Restricting the field of view: perceptual and performance effects . Percept. Motor Skills , 70 , 35 – 45 . Google Scholar CrossRef Search ADS Anderson , J. ( 2010 ) Understanding the Changing Needs of the US Online Consumer, 2010. Forrester Research. Appel , M. and Richter , T. ( 2007 ) Persuasive effects of fictional narratives increase over time . Media Psychol. , 10 , 113 – 134 . Arnold , M.J. and Reynolds , K.E. ( 2003 ) Hedonic shopping motivations . J. Retail. , 79 , 77 – 95 . Google Scholar CrossRef Search ADS Baudisch , P. , DeCarlo , D. , Duchowski , A.T. and Geisler , W.S. ( 2003 ) Focusing on the essential: considering attention in display design . Commun. ACM , 46 , 60 – 66 . Google Scholar CrossRef Search ADS Betrancourt , M. ( 2005 ) The animation and interactivity principles in multimedia learning. In Mayer , R. (ed.) , The Cambridge Handbook of Multimedia Learning . pp. 287 – 296 . Cambridge University Press , Cambridge . Google Scholar CrossRef Search ADS Biocca , F. ( 1992 ) Communication within virtual reality: creating a space for research . J. Commun. , 42 , 5 – 22 . Google Scholar CrossRef Search ADS Black , J.B. , Turner , T.J. and Bower , G.H. ( 1979 ) Point of view in narrative comprehension, memory, and production . J. Verb. Learn. Verb. Behav. , 18 , 187 – 198 . Google Scholar CrossRef Search ADS Bordwell , D. , Thompson , K. and Ashton , J. ( 1997 ) Film Art: An Introduction . McGraw-Hill , New York . Bornoe , N. and Barkhuus , L. ( 2010 , April) Video microblogging: Your 12 seconds of fame. In CHI'10 Extended Abstracts on Human Factors in Computing Systems. ACM, pp. 3325–3330. Brock , T.C. , Strange , J.J. and Green , M.C. ( 2002 ) Power beyond reckoning. In Green , M. C. , Strange , J. J. and Brock , T. C. (eds) , Narrative impact: Social and Cognitive Foundations . pp. 1 – 16 . Lawrence Erlbaum , Mahwah, NJ . Brooks , C. and Warren , R.P. ( 1959 ) Understanding Fiction , Vol. 5 , Appleton-Century-Crofts , New York . Bruner , J. and Lucariello , J. ( 1989 ) Monologue as narrative recreation of the world, in Narratives from the Crib. pp. 73–97. Bucy , E.P. ( 2004 ) Second generation net news: interactivity and information accessibility in the online environment . Int. J. Media Manag. , 6 , 102 – 113 . Google Scholar CrossRef Search ADS Buhler , J. and Wexler , D. , 2002 . A phenomenological model for bokeh rendering. In ACM SIGGRAPH 2002 Conference Abstracts and Applications. ACM, pp. 142 – 142 . Busselle , R. and Bilandzic , H. ( 2009 ) Measuring narrative engagement . Media Psychol. , 12 , 321 – 347 . Google Scholar CrossRef Search ADS Calvo , M.G. and Lang , P.J. ( 2005 ) Parafoveal semantic processing of emotional visual scenes . J. Exp. Psychol. , 31 , 502 . Castellano , G. , Kessous , L. and Caridakis , G. ( 2008 ) Emotion recognition through multiple modalities: face, body gesture, speech. In Peter , C. and Beale , R. (eds) , Affect and Emotion in Human-Computer Interaction . pp. 92 – 103 . Springer , Berlin, Heidelberg . Google Scholar CrossRef Search ADS Chatman , S. ( 1975 ) Towards a theory of narrative . New Literary Hist. , 6 , 295 – 318 . Google Scholar CrossRef Search ADS Chung , N. , Han , H. and Koo , C. ( 2015 ) Adoption of travel information in user-generated content on social media: the moderating effect of social presence . Behav. Inform. Technol. , 34 , 902 – 919 . Google Scholar CrossRef Search ADS Cohen , J. ( 2001 ) Defining identification: a theoretical look at the identification of audiences with media characters . Mass Commun. Soc. , 4 , 245 – 264 . Google Scholar CrossRef Search ADS Constine , J. ( 2017 , June 20) Instagram Stories hits 250M daily users, adds Live video replays. Retrieved March 27, 2018, from https://techcrunch.com/2017/06/20/instagram-live-video-replays/ Coyle , J.R. and Thorson , E. ( 2001 ) The effects of progressive levels of interactivity and vividness in web marketing sites . J. Advert. , 30 , 65 – 77 . Google Scholar CrossRef Search ADS Daft , R.L. and Lengel , R.H. ( 1986 ) Organizational information requirements, media richness and structural design . Manag. Sci. , 32 , 554 – 571 . Google Scholar CrossRef Search ADS De Cesarei , A. , Codispoti , M. and Schupp , H.T. ( 2009 ) Peripheral vision and preferential emotion processing . NeuroReport , 20 , 1439 – 1443 . Google Scholar CrossRef Search ADS PubMed Dennis , A.R. , Fuller , R.M. and Valacich , J.S. ( 2008 ) Media, tasks, and communication processes: a theory of media synchronicity . MIS Quarterly , 32 , 575 – 600 . Google Scholar CrossRef Search ADS Dennis , A.R. and Valacich , J.S. ( 1999 , January) Rethinking media richness: towards a theory of media synchronicity. In Systems Sciences, 1999. HICSS-32. Proc. 32nd Annual Hawaii International Conference on (pp. 10-pp). IEEE. Dolezal , H. ( 1982 ) Living in a World Transformed: Perceptual and Performatory Adaptation to Visual Distortion. Academic Press. Eliot , T.S. ( 1957 ) The social function of poetry . On Poetry Poets , 19 , 33 . Farrar , K.M. , Krcmar , M. and Nowak , K.L. ( 2006 ) Contextual features of violent video games, mental models, and aggression . J. Commun. , 56 , 387 – 405 . Google Scholar CrossRef Search ADS Fischer , A.H. , Rotteveel , M. , Evers , C. and Manstead , A.S. ( 2004 ) Emotional assimilation: how we are influenced by others’ emotions . Curr. Psychol. Cogn. , 22 , 223 – 246 . Fisher , W.R. ( 1985 ) The narrative paradigm: an elaboration . Commun. Monogr. , 52 , 347 – 367 . Google Scholar CrossRef Search ADS Fornell , C. and Larcker , D.F. ( 1981 ) Evaluating structural equation models with unobservable variables and measurement error . J. Market. Res. , 18 , 39 – 50 . Google Scholar CrossRef Search ADS Galyean , T.A. , III . ( 1995 ) Narrative guidance of interactivity. Doctoral dissertation, Massachusetts Institute of Technology. Girgensohn , A. , Boreczky , J. , Chiu , P. , Doherty , J. , Foote , J. , Golovchinsky , G. , Uchihashi , S. and Wilcox , L. 2000 . A semi-automatic approach to home video editing. Proc. 13th Annual ACM Symposium on User Interface Software and Technology. ACM, pp. 81–89. Gould , S. , Arfvidsson , J. , Kaehler , A. , Sapp , B. , Messner , M. , Bradski , G.R. , Baumstarck , P. , Chung , S. and Ng , A.Y. ( 2007 ) Peripheral-foveal vision for real-time object recognition and tracking in video . IJCAI , 7 , 2115 – 2121 . Graesser , A.C. , Olde , B. and Klettke , B. ( 2002 ) How does the mind construct and represent stories . Narrative impact: Social and cognitive foundations , 229 – 262 . Green , M.C. ( 2004 ) Transportation into narrative worlds: the role of prior knowledge and perceived realism . Discourse Processes , 38 , 247 – 266 . Google Scholar CrossRef Search ADS Green , M.C. and Brock , T.C. ( 2000 ) The role of transportation in the persuasiveness of public narratives . J. Personal. Soc. Psychol. , 79 , 701 . Google Scholar CrossRef Search ADS Heeter , C. ( 1989 ) Classifying mediated communication systems . Commun. Yearbook , 12 , 477 – 489 . Heeter , C. ( 2000 ) Interactivity in the context of designed experiences . J. Interact. Advert. , 1 , 3 – 14 . Google Scholar CrossRef Search ADS Hiltz , S.R. , Coppola , N. , Rotter , N. , Toroff , M. and Benbunan-Fich , R. ( 2000 ) Measuring the importance of collaborative learning for the effectiveness of ALN: a multi-measure . Online Educ. , 1 , 101 – 119 . Cisco . Visual Networking Index: Global Mobile Data Traffic Forecast Update 2014–2019 White Paper. (n.d.). Retrieved June 24, 2015, from http://www.cisco.com/c/en/us/solutions/collateral/service-provider/visual-networking-index-vni/white_paper_c11-520862.html Hua , X.S. , Lu , L. and Zhang , H.J. ( 2004 ) Optimization-based automated home video editing system . IEEE Trans. Circuits Syst. Video Technol. , 14 , 572 – 583 . Google Scholar CrossRef Search ADS Jang , J. , Kim , J. , Shin , H. , Aum , H. and Kim , J. ( 2016 ) Effects of temporal format of everyday video on narrative engagement and social interactivity . Interact. Comput. , 28 , 718 – 736 . iwv043. Google Scholar CrossRef Search ADS Janssen , J.H. , IJsselsteijn , W.A. and Westerink , J.H. ( 2014 ) How affective technologies can influence intimate interactions and improve social connectedness . Int. J. Hum. Comput. Stud. , 72 , 33 – 43 . Google Scholar CrossRef Search ADS Kellogg , W.A. , Bellamy , R.K.E. and Van Deusen , M. ( 1997 ) A practical guide to working with edited video. In Handbook of Human-Computer Interaction ( 2 edn ). pp. 947 – 978 . Elsevier Science BV , Amsterdam . Google Scholar CrossRef Search ADS Kenney , W.P. ( 1988 ) How to Read and Write About Fiction . Arco , New York . Kim , J. ( 2015 ) Design for Experience: Where Technology Meets Design and Strategy . Springer . Kim , T. and Biocca , F. ( 1997 ) Telepresence via television: two dimensions of telepresence may have different connections to memory and persuasion . J. Comput. Mediated Commun. , 3 . Kirk , D. , Sellen , A. , Harper , R. and Wood , K. ( 2007 , April) Understanding videowork. Proc. SIGCHI Conf. Human Factors in Computing Systems (pp. 61–70). ACM. Klatzky , R.L. ( 1998 ) Allocentric and egocentric spatial representations: definitions, distinctions, and interconnections. In Spatial Cognition . pp. 1 – 17 . Springer , Berlin Heidelberg . Kolfschoten , G.L. ( 2011 ) Cognitive load in collaboration-brainstorming. Proc. 44th Hawaii International Conference on System Sciences. IEEE. pp. 1–9. Kraus , W. ( 2006 ) The narrative negotiation of identity and belonging . Narrat. Inquiry , 16 , 103 – 111 . Google Scholar CrossRef Search ADS Lamme , V.A. ( 2003 ) Why visual attention and awareness are different . Trends Cogn. Sci. , 7 , 12 – 18 . Google Scholar CrossRef Search ADS PubMed Langer , E. ( 2014 , June 18) Vine’s most popular user clues in the ‘dinosaurs’. Retrieved March 27, 2018, from https://www.cnbc.com/id/100899790 Lehmuskallio , A. and Sarvas , R. ( 2008 , October) Snapshot video: everyday photographers taking short video-clips. Proc. 5th Nordic Conf. Human-computer interaction: building bridges (pp. 257–265). ACM. Lienhart , R.W. , ( 1999 ) Dynamic video summarization of home video. In Electronic Imaging. pp. 378 – 389 . Lim , S. and Reeves , B. ( 2009 ) Being in the game: effects of avatar choice and point of view on psychophysiological responses during play . Media Psychology , 12 , 348 – 370 . Google Scholar CrossRef Search ADS Liu , Y. and Shrum , L.J. ( 2002 ) What is interactivity and is it always such a good thing? Implications of definition, person, and situation for the influence of interactivity on advertising effectiveness . J. Advert. , 31 , 53 – 64 . Google Scholar CrossRef Search ADS Luo , Y. and Tang , X. ( 2008 ) Photo and Video Quality Evaluation: Focusing on the Subject. In Computer Vision–ECCV 2008 . pp. 386 – 399 . Springer , Berlin, Heidelberg . Google Scholar CrossRef Search ADS McCarthy , J. and Wright , P. ( 2004 ) Technology as experience . Interactions , 11 , 42 – 43 . Google Scholar CrossRef Search ADS McMillan , S.J. and Hwang , J.S. ( 2002 ) Measures of perceived interactivity: an exploration of the role of direction of communication, user control, and time in shaping perceptions of interactivity . J. Advert. , 31 , 29 – 42 . Google Scholar CrossRef Search ADS Nalder , J. ( 2013 ) Making molehills from mountains: the tilt-shift effect . Screen Educ. , 68 , 98 . Neitzel , B. ( 2005 ) Levels of play and narration. In Meister , J.C. (ed.) , Narratology Beyond Literary Criticism. Mediality, Disciplinarity. Narratologia Band 6 . pp. 45 – 64 . Walter de Gruyter , Berlin and New York . Google Scholar CrossRef Search ADS Ojala , J. , Mate , S. , Curcio , I.D. , Lehtiniemi , A. and Väänänen-Vainio-Mattila , K. ( 2014 , November) Automated creation of mobile video remixes: user trial in three event contexts. In Proc. 13th Int. Conf. Mobile and Ubiquitous Multimedia (pp. 170–179). ACM. O’Hara , K. , Black , A. and Lipson , M. ( 2006 , April) Everyday practices with mobile video telephony. Proc. SIGCHI Conf. Human Factors in computing systems (pp. 871–880). ACM. Pelli , D.G. ( 1987 ) The visual requirements of mobility, in Low Vision . pp. 134 – 146 . Springer , New York . Peng , W.T. , Chu , W.T. , Chang , C.H. , Chou , C.N. , Huang , W.J. , Chang , W.Y. and Hung , Y.P. ( 2011 ) Editing by viewing: automatic home video summarization by viewing behavior analysis . IEEE Trans. Multimed. , 13 , 539 – 550 . Google Scholar CrossRef Search ADS Prensky , M. ( 2001 ) Fun, play and games: What makes games engaging . Digital Game-Based Learn. , 5 , 1 – 05 . Puikkonen , A. , Häkkilä , J. , Ballagas , R. and Mäntyjärvi , J. ( 2009 , September). Practices in creating videos with mobile phones. Proc. 11th Int. Conf. Human-Computer Interaction with Mobile Devices and Services (p. 3). ACM. Puikkonen , A. , Ventä , L. , Häkkilä , J. and Beekhuyzen , J. ( 2008 , December) Playing, performing, reporting: a case study of mobile minimovies composed by teenage girls. Proc. 20th Australasian Conf. Computer-Human Interaction: Designing for Habitus and Habitat (pp. 140–147). ACM. Qin , H. , Patrick Rau , P.L. and Salvendy , G. ( 2009 ) Measuring player immersion in the computer game narrative . Int. J. Hum. Comput. Interact. , 25 , 107 – 133 . Google Scholar CrossRef Search ADS Rayner , K. , Inhoff , A.W. , Morrison , R.E. , Slowiaczek , M.L. and Bertera , J.H. ( 1981 ) Masking of foveal and parafoveal vision during eye fixations in reading . J. Exp. Psychol. Hum. Percept. Perform. , 7 , 167 . Google Scholar CrossRef Search ADS PubMed Riedl , M.O. and Young , R.M. ( 2010 ) Narrative planning: balancing plot and character . J. Artif. Intell. Res. , 39 , 217 – 268 . Google Scholar CrossRef Search ADS Rigoulot , S. , Delplanque , S. , Despretz , P. , Defoort-Dhellemmes , S. , Honoré , J. and Sequeira , H. ( 2008 ) Peripherally presented emotional scenes: a spatiotemporal analysis of early ERP responses . Brain Topogr. , 20 , 216 – 223 . Google Scholar CrossRef Search ADS PubMed Rigoulot , S. , D’Hondt , F. , Honore , J. and Sequeira , H. ( 2012 ) Implicit emotional processing in peripheral vision: behavioral and neural evidence . Neuropsychologia , 50 , 2887 – 2896 . Google Scholar CrossRef Search ADS PubMed Roskos-Ewoldsen , B. , Davies , J. and Roskos-Ewoldsen , D.R. ( 2004 ) Implications of the mental models approach for cultivation theory . Communications , 29 , 345 – 363 . Google Scholar CrossRef Search ADS Schank , R.C. and Berman , T.R. ( 2002 ) The pervasive role of stories in knowledge and action. In Green , M.C. , Strange , J.J. and Brock , T.C. (eds) , Narrative Impact: Social and Cognitive Foundations . pp. 287 – 313 . Lawrence Erlbaum Associates . Schuurink , E.L. and Toet , A. ( 2010 ) Effects of third-person perspective on affective appraisal and engagement: findings from SECOND LIFE . Simulat. Gaming , 41 , 724 – 742 . Google Scholar CrossRef Search ADS Slater , M.D. , Rouner , D. and Long , M. ( 2006 ) Television dramas and support for controversial public policies: effects and mechanisms . J. Commun. , 56 , 235 – 252 . Google Scholar CrossRef Search ADS Somers , M.R. ( 1994 ) The narrative constitution of identity: a relational and network approach . Theory Soc. , 23 , 605 – 649 . Google Scholar CrossRef Search ADS Steeples , C. ( 2002 , January) Voice annotation of multimedia artifacts: reflective learning in distributed professional communities. In System Sciences, 2002. HICSS. Proc. 35th Annual Hawaii International Conference on (10 pp.). IEEE. Stern , B.B. ( 1991 ) Who talks advertising? Literary theory and narrative ‘point of view . J. Advert. , 20 , 9 – 22 . Google Scholar CrossRef Search ADS Steuer , J. , Biocca , F. and Levy , M.R. ( 1995 ) Defining virtual reality: dimensions determining telepresence . Commun. Age Virtual Real. , 33 – 56 . Sundar , S.S. ( 2004 ) Theorizing interactivity’s effects . Inform. Soc. , 20 , 385 – 389 . Google Scholar CrossRef Search ADS Tamborini , R. , Eastin , M. , Lachlan , K. , Skalski , P. , Fediuk , T. and Brady , R. ( 2001 ) Hostile thoughts, presence and violent virtual video games. In 51st Annual Convention of the International Communication Association, Washington, DC. Taylor , S.E. and Thompson , S.C. ( 1982 ) Stalking the elusive’ vividness’ effect . Psychol. Rev. , 89 , 155 . Google Scholar CrossRef Search ADS Tench , D. ( 1989 ) Narrative presence and narrative structure in Verga’s early novels . J. Narrat. Technique , 205 – 216 . Vogeley , K. and Fink , G.R. ( 2003 ) Neural correlates of the first-person-perspective . Trends Cogn. Sci. , 7 , 38 – 42 . Google Scholar CrossRef Search ADS PubMed Webb , N.A. and Griffin , M.J. ( 2003 ) Eye movement, vection, and motion sickness with foveal and peripheral vision . Aviat. Space Environ. Med. , 74 , 622 – 625 . Google Scholar PubMed Wignall , J. ( 2012 ) Digital Photography FAQs . John Wiley & Sons . Wilson , E. ( 1994 ) The Theater Experience . McGraw-Hill , Blacklick, OH . Wilson , B. and Henseler , J. ( 2007 ) Modeling reflective higher-order constructs using three approaches with PLS path modeling: A Monte Carlo comparison. Australian and New Zealand Marketing Academy Conference. Yarow , J. ( 2013 ) Vine sharing plummets after Instagram launches video. http://www.businessinsider.com/chart-of-the-day-vine-sharing-plummets-after-instagram-launches-video-2013-6 Zillmann , D. ( 1995 ) Mechanisms of emotional involvement with drama . Poetics , 23 , 33 – 51 . Google Scholar CrossRef Search ADS Zsombori , V. , Frantzis , M. , Guimaraes , R.L. , Ursu , M.F. , Cesar , P. , Kegel , I. , Craigie , R. and Bulterman , D.C. (2011) Automatic generation of video narratives from shared UGC. In Proc. 22nd ACM Conference on Hypertext and Hypermedia. ACM, pp. 325–334. APPENDIX Nar A example View largeDownload slide View largeDownload slide This condition has six scenes of first-person perspective with foveal vision continuously. Nar B example View largeDownload slide View largeDownload slide This condition has six scenes of first-person perspective with peripheral vision continuously. Nar C example View largeDownload slide View largeDownload slide This condition has six scenes of first-person perspective with intersecting foveal and peripheral vision. Nar D example View largeDownload slide View largeDownload slide This condition has six scenes of third-person perspective with foveal vision continuously. Nar E example View largeDownload slide View largeDownload slide This condition has six scenes of third-person perspective with peripheral vision continuously. Nar F example View largeDownload slide View largeDownload slide This condition has six scenes of third-person perspective with intersecting foveal and peripheral vision. Nar G example View largeDownload slide View largeDownload slide This condition has six scenes of foveal vision with intersecting one-person and third-person perspective. Nar H example View largeDownload slide View largeDownload slide This condition has six scenes of peripheral vision with intersecting one-person and third-person perspective. Nar I example View largeDownload slide View largeDownload slide This condition has mixed scene effects with intersecting one-person perspective X foveal vision and third-person perspective X peripheral vision. Nar J example View largeDownload slide View largeDownload slide This condition has mixed scene effects with intersecting one-person perspective X peripheral vision and third-person perspective X foveal vision. Author notes Handling Editor: Dr Sharon Tettegah © The Author(s) 2018. Published by Oxford University Press on behalf of The British Computer Society. All rights reserved. For Permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Interacting with ComputersOxford University Press

Published: Jul 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off