Why Are You Swiping Right? The Impact of Product Orientation on Swiping Responses

Why Are You Swiping Right? The Impact of Product Orientation on Swiping Responses Abstract Many apps require consumers to evaluate products by swiping them to the right or left. This work explores whether product orientation affects the product evaluations communicated by swiping movements, compared with those made by pressing onscreen buttons. Building on stimulus-response compatibility (SRC) theory, which suggests that irrelevant product display features can activate certain behavioral responses when the product display and the behavioral response share a common dimension, this study predicts that the horizontal direction (left to right or right to left) cued by a product’s orientation should facilitate a swipe movement in the congruent direction. Five studies indicate that when people use swiping movements to evaluate objects, their evaluations are influenced by the object’s orientation, whereas evaluations conveyed through button presses reveal no orientation effect. The orientation effect for swiping responses also disappears when the objects contain a direction cue that is incongruent with their orientation, and when only one directional swipe movement is defined as a valid response option. Moreover, the effect holds for subjective evaluations but is eliminated for objective judgments, when these involve no time pressure. affordance, product orientation, touchscreen, tablet, mobile device, stimulus–response compatibility With the increasingly widespread adoption of tablets and smartphones, conventional interfaces are shifting from computer mice to touchscreens (Brasel and Gips 2014), creating new ways to interact with objects presented onscreen. For example, with the online dating app Tinder, users can like or dislike prospective dates by swiping their profile pictures to the right or left of the screen. The Stylect app helps women find the perfect shoes by showing options and encouraging them to swipe to the right if they like them or to the left if not. This conversion to touchscreen interfaces has made swiping an increasingly common movement, used to interact with products and stimuli onscreen. Accordingly, it is important to assess how the adoption of new product interaction behaviors (like swiping) affects product evaluations, especially when we consider that more than 8% of e-commerce website visits already involve tablets (Brasel and Gips 2014), and this rate seems likely to increase in the future. The current study investigates in particular how product orientation might affect liking evaluations conveyed through swiping. We propose that an object’s horizontal orientation (i.e., whether its front side is located to the left or right) primes a lateral direction, and people are more inclined to swipe in a direction congruent with the primed direction than in an incongruent direction. For example, when using the Stylect app, women may be more inclined to engage in a rightward swipe if the toe of the presented shoe is directed to the right, whereas an opposite swiping movement may be preferred if the toe of the same shoe were oriented to the left. A positive evaluation typically is conveyed with a left-to-right swipe, so rightward-oriented products may have a better chance of being liked than leftward-oriented ones. The current research contributes to two main streams of research. First, we extend research on the effects of product displays on product evaluations, by focusing on a distinct aspect of horizontal product orientation. Specifically, most extant literature dealing with horizontal product orientation addresses handle locations (Eelen, Dewitte, and Warlop 2013; Elder and Krishna 2012; Ping, Dhillon, and Beilock 2009), but we instead consider the location of the product’s front. By changing the focal element that defines the horizontal orientation, our research is relevant for a more expanded set of objects—namely, any object with a clear back and front—rather than the more restrictive set of objects with handles (see appendix A for illustrative examples). More importantly, the mechanism that explains the impact of handle location is unique to products with handles, so we also introduce a different mechanism by which product orientation—and product display features in general—affects evaluations. Specifically, we draw on stimulus–response overlap research (Kornblum, Hasbroucq, and Osman 1990) that describes how irrelevant physical features of a stimulus can interfere with proper responses to it. This body of research suggests that a behavioral response congruent with an irrelevant feature of a stimulus gets activated when both the stimulus and response vary along an overlapping dimension. We demonstrate in turn that dimensional overlap not only affects performance in response latency tasks but also serves as a source of bias for consumer evaluations. We thus specify a new route through which product displays can affect product liking. Second, our findings extend stimulus–response compatibility (SRC) research in several ways. Previous research has investigated features such as color (Stroop 1935), horizontal (Simon 1969) and vertical (Ansorge and Wuhr 2004) positions, surrounding letters (Eriksen and Eriksen 1974), pitch (Melara and Marks 1990), valence (De Houwer and Eelen 1998), and semantic meaning (De Houwer 1998). The previously studied responses include verbal responses (e.g., color naming, Stroop 1935), keypresses (most studies), and sagittal lever movements (Eder and Rothermund 2008). In a clear distinction, we investigate the potential interference of a hitherto unstudied irrelevant physical feature (i.e., horizontal orientation) and a neglected response (i.e., lateral swiping). This irrelevant feature of horizontal orientation also is less directly connected to the overlapping response dimension (direction) than is the case in typical SRC tasks. The objects do not move leftward or rightward in our study, so the overlap is more covert in nature. Finally, while typical SRC tasks involve judgments that can be established as objectively correct or incorrect, the current research considers subjective judgments like product evaluations. This shift in focus has implications for the impact of irrelevant stimulus features, especially for how this impact develops over time. In typical SRC tasks, participants try to inhibit the response activated by the influence of an irrelevant stimulus dimension when it leads to an incorrect response. Occasionally, when responses are very fast, participants might fail to do so. In contrast, subjective judgments lack an objectively correct response, rendering the issue of inhibiting an incorrect response moot. This lack of inhibition implies that SRC effects for subjective judgments are not limited to fast responses. THE INFLUENCE OF PRODUCT DISPLAY ON EVALUATIONS In addition to the way a product looks, its display also can affect its evaluation (Areni, Duhan, and Kiecker 1999; Buchanan, Simmons, and Bickart 1999; Castro, Morales, and Nowlis 2013; Cavazza and Gabrieli 2015; Fiore, Yah, and Yoh 2000; Zhu and Meyers-Levy 2009). For example, consumers evaluate a product more positively when it is placed centrally (Atalay, Bodur, and Rasolofoarison 2012; Keller, Markert, and Bucher 2015; Rodway, Schepman, and Lambert 2012; Valenzuela and Raghubir 2009) or at the top (Valenzuela and Raghubir 2015) in a visual display, as well as in environments that are less complex (Orth and Crouch 2014). Product display matters not only when products are surrounded by other products but also for (images of) products in isolation. Various studies show that product evaluations also can be affected by arbitrary product orientations. For example, products with their handle oriented to the right are rated as more attractive than their leftward-oriented counterparts (Eelen et al. 2013; Elder and Krishna 2012; Ping et al. 2009). Studies showing that handle location impacts product evaluations build on affordance theory (Gibson 1979), according to which the perception of objects can activate object interaction movements that people commonly execute when encountering the object. For instance, seeing pencils and cherries facilitates the use of a precision grip in a classification task; exposure to hammers and apples instead facilitates the use of a power grip in a similar task (Ellis and Tucker 2000; Tucker and Ellis 2001). Similarly, a handle’s position activates grasping with the corresponding hand (i.e., a handle on the left activates left-hand grasping). Depending on the person’s dominant handedness, the grasping behavior activated by the handle location also is relatively easier or more difficult to execute. The motor fluency (or lack thereof) associated with the simulated grasping behavior then affects liking for a presented, graspable object (Eelen et al. 2013; Elder and Krishna 2012; Ping et al. 2009). Merely observing an ad featuring a bowl of yogurt with a spoon presented on the right (vs. left) side of the bowl increases purchase intentions among right-handed participants, for example (Elder and Krishna 2012). Rather than handle location, we focus on a different, more general, element of product orientation that might affect product evaluations: the horizontal position of the back and front of the product. An object is leftward-oriented if its front is on the left side of the view, and rightward-oriented if the front is on the right side. From a practical perspective, this novel focus means the present research applies to any product with a clear back and front, rather than to the more restrictive set of products with handles. From a theoretical perspective, it implies that the effects of product orientation require different process explanations. Handle location primes grasping with a corresponding hand, but in the absence of a handle, no corresponding grasping behavior can be cued, nor can the (imagined) difficulty of grasping affect product evaluations. The new process explanation derives from research on stimulus–response compatibility. STIMULUS–RESPONSE COMPATIBILITY Stimulus–response compatibility refers to the impact that irrelevant stimulus features can have on behaviors if the cognitive representation of the stimulus shares features with that of the behavior (Hommel 1997). A classic demonstration of the impact of irrelevant features is offered by the well-known Stroop (1935) task in which people must name the color of the font in which a word, describing a color, is written. In congruent trials, the meaning of the color word is identical to the font color (e.g., the word red is written in red), but in incongruent trials, they differ (e.g., the word red is written in yellow). People tend to be slower and make more mistakes naming the font color in incongruent trials than in congruent ones. Ample studies support a dimensional overlap explanation of such SRC phenomena (e.g., Kornblum et al. 1990; Zhang, Zhang, and Kornblum 1999). This explanation proposes that dimensional overlap occurs when responses and stimuli vary along the same dimension. For example, in the Stroop task, the responses (naming colors) overlap on two dimensions with the stimuli; they vary along the same dimension as the relevant feature (font color) but also as the irrelevant feature (color word). While the overlap of responses with the relevant feature is no cause for concern (and in fact, is absent in many types of SRC phenomena), the overlap with the irrelevant feature causes interference, as it automatically activates responses congruent with it. Independent of the irrelevant stimulus feature (color word) that activates one response action, the relevant stimulus dimension (font color) still initiates a controlled process to identify the correct response action. In congruent trials, where the response based on the irrelevant feature is identical to the response based on the relevant feature, the automatically activated response gets verified and executed quickly (Hommel 1997; Kornblum et al. 1990; Zhang et al. 1999). For example, when people see the word red written in red, the response (saying “red”) is automatically activated, verified, and quickly executed. In incongruent trials, the response activated by the irrelevant stimulus feature is incorrect and needs to be halted, while the correct response needs to be initiated. Hence, when the word red is written in yellow, the incorrect response (saying “red”) becomes activated and needs to be inhibited in order to convey the correct response (saying “yellow”). These additional processes cause a response delay, explaining the slower response times. Because inhibition occasionally fails, the wrong response might be executed (Hommel 1997; Kornblum et al. 1990; Zhang et al. 1999). Building on dimensional overlap theory, we argue that a product’s orientation, which is irrelevant for its evaluation, can nevertheless affect its evaluation if it is conveyed through swiping. Essentially, in expressing product evaluations, (1) directional overlap may serve as a source of response activation, but (2) the process via which the activated response gets enacted is different from classical SRC tasks. We develop these two propositions further below. DIMENSIONAL OVERLAP OF STIMULUS ORIENTATION AND SWIPING RESPONSES Applications that require people to convey evaluations on mobile devices often involve swiping to the left or right or pressing a button on the left or right of a screen. However, the objects or products that app users encounter not only vary in how much one likes them but also often in their orientation. According to dimensional overlap theory, liking is, by definition, the relevant feature for an evaluation task, and product orientation an irrelevant one. Therefore, evaluations may be affected by product orientation if the specific behaviors to convey them vary in features that overlap with product orientation. Hence, we suggest that dimensional overlap may cause product orientations to activate, or prime, corresponding responses, which increases the likelihood of the congruent response and decreases the likelihood of the incongruent one. This line of reasoning implies that the proposed effect occurs at the behavioral (leftward vs. rightward swipe) rather than evaluative (like vs. dislike) level, suggesting a crucial role for space-valence mapping. Most apps map a positive response on the right side and a negative response to the left; swiping to the right virtually always conveys a liking response. Therefore, rightward-oriented products should be better liked than leftward-oriented ones. However, if the space-valence mapping reverses (i.e., the left side represents “like” and the right side represents “dislike”), we still might observe that leftward-oriented stimuli tend to elicit leftward responses; given the reversed space-valence mapping, though, it would imply a tendency to like leftward-oriented stimuli better than rightward-oriented stimuli. In most of our studies, we compare the impact of product orientation on swiping responses and button presses. Depending on the specific spatial feature that product orientation cues, it may affect either both responses or swiping responses only. Horizontal product orientation may either cue a specific lateral or left-right direction or a specific location. If product orientation cues lateral direction, it should affect swiping responses, which imply movement along the lateral direction too, but it should not affect button presses, because they imply movement in a direction that is perpendicular to the cued lateral direction. In fact, while swiping responses vary in the lateral or left-right direction (i.e., occur along the frontal axis), button presses vary in the up-down direction (i.e., occur along the sagittal axis). If, however, product orientation cues a specific left-right location, it should affect button presses, at least if they are executed with fingers of a different hand. While we pit a direction account against a location account in most of our studies, we consider it more likely that product orientation cues direction rather than location and, hence, that only swiping responses will be affected by product orientation. The idea that product orientation affects swiping responses due to dimensional overlap suggests two means to eliminate its effect. First, when a second irrelevant feature also cues lateral direction, it may cancel out the effect of product orientation, at least if the two irrelevant cues prime opposing directions. Second, stimuli and responses need to vary along the same dimension to prompt interference or facilitation (Ansorge and Wühr 2004). In our case, product orientation, and thus implied direction, varies across products. In a typical swiping task, swiping responses also vary along the direction dimension, because people can swipe right or left. If a task requires consumers to make only left-to-right swipe movements, the response direction does not vary across trials, which eliminates any overlap with direction. Therefore, we predict that a horizontal product orientation may activate a directionally congruent response if the task involves choosing to swipe to the left or to the right, but not if it involves swiping only in one direction. SELECTION OF SWIPING RESPONSES TO CONVEY PRODUCT EVALUATIONS VERSUS OBJECTIVE JUDGMENTS Even if the proposed effect of product orientation on evaluation through swiping seems a typical SRC effect, it differs in a key aspect: the absence of inhibition. In SRC tasks, people sometimes make mistakes when answering quickly, and they are very well aware when they do. Anyone who has ever served as a participant in a Stroop task, for instance, undoubtedly has had the experience of realizing immediately after responding that he or she said the color word instead of the font color. This experience reveals clearly to participants that they have been led astray by an irrelevant feature; we argue that the existence of objectively correct versus incorrect answers is crucial to participants’ efforts to inhibit their automatically activated response. But in evaluation tasks, participants are unlikely to be actively occupied with inhibiting a response that is automatically activated by product orientation. In fact, it is unlikely that participants would view their evaluations as incorrect or invalid, as many studies indicate how little insight people have into the construction of their own evaluations (Cacioppo, Priester, and Berntson 1993; Schwarz 2004, 2006; Slovic 1995; Strack, Martin, and Stepper 1988). The presence versus absence of inhibitory processes affects the temporal dynamic of the observed SRC effects. In typical SRC tasks, slower responses are much less likely to be affected by dimensional overlap, because more time results in a higher likelihood of successful inhibition. If consumers do not try to inhibit automatically activated responses, however, the effect on their responses is the same, regardless of how quickly or slowly they issue those responses. This distinction is particularly pertinent for our study setting, because real-life swiping responses take more time than is typically allowed in SRC tasks, which tend to drop responses slower than 1,500 milliseconds (De Houwer 2003; Hommel 1993; MacDonald et al. 2000). So, while no SRC effects would be observed in typical SRC tasks at these long response windows, we predict that response window is unlikely to moderate the impact of product orientation on swiping. STUDY OVERVIEW To test our propositions, we conduct five studies. Study 1 demonstrates the hypothesized effect: on a touchscreen device, rightward product orientations (i.e., toys with their front oriented to the right vs. left) tend to elicit more liking responses than leftward-oriented ones when the positive evaluations are mapped on the right side and the evaluation is conveyed through swiping. No orientation effect is observed for button pressing, though. Study 2 demonstrates the critical role of space-valence mapping: we replicate the effect of product orientation on swiping-based evaluations for the usual space-valence mapping (left = bad, right = good), but find it reverses when the space-valence mapping is reversed. Studies 3 and 4 demonstrate that the effect of product orientation on swiping is attenuated when dimensional overlap is reduced or absent. Study 3 introduces a second directional cue (i.e., implied motion) as a moderator. We replicate the effect of product orientation on evaluation when implied motion is congruent with product orientation, but find it disappears when implied motion is incongruent with product orientation. Study 4 tests the explanatory role of dimensional overlap by showing that the effect is eliminated when only one directional response is defined (i.e., left-to-right swipe movement); establishing dimensional overlap requires the responses to vary on the overlapping dimension. Finally, the influence of horizontal orientation on swiping critically depends on the presence versus absence of inhibitory processes. Study 5 shows that in conveying objective judgments, such as discriminating between toys that depict animals or not, consumers tend to override the influence of product orientation in the absence of time pressure. However, when they convey evaluations, no inhibition of the primed response occurs, irrespective of time pressure. STUDY 1: PRODUCT ORIENTATION INFLUENCES SWIPING DIRECTION The aim of this first study is to assess whether product orientation affects which swipe movement consumers prefer. We predict that when a product is oriented rightward, the tendency to swipe to the right is greater than when the product is oriented leftward. If evaluations instead require pressing a button, people do not engage in lateral directional movement, so product orientation should not affect evaluations (to the same extent) in this condition. Method In exchange for monetary compensation (€8), 58 students of Ghent University (34 women, MAge = 21.59, SD = 3.64) were randomly assigned to one of two cells in a between-subjects design (response mode: swipe vs. button). All participants were informed that the research was about gift giving. Pictures of toys appeared on the screen, and participants were instructed to indicate whether they would consider the presented toy as an attractive gift for one- to four-year-old children. A set of 30 pictures of toys (see appendix B for a sample), each with a clear leftward or rightward orientation, and their mirror-reversed version were included in this study. The pictures appeared on the screen, one by one and in a fixed order. Whether a toy was presented with its leftward or rightward orientation was determined randomly. Hence, all participants saw the exact same stimuli in the exact same order, but the orientation varied. When a picture appeared in the center of the screen, participants in the button condition could like it as a gift by pressing a green checkmark, which appeared on the right side of the screen. In the swipe condition, they instead would signal a positive evaluation by swiping the picture toward the checkmark. If participants wanted to evaluate the presented toy negatively, they pressed a red “x,” fixed on the left side of the screen, or swiped the picture to the left. Results and Discussion The 58 participants provided 1,740 responses in total. The time to make a decision varied from 6 milliseconds to 30 seconds. In all studies, we analyzed the data for all responses; we also analyzed the data after dropping implausibly fast responses (RT < 200 milliseconds) to eliminate responses that participants might have initiated before they even saw the stimulus. This cutoff is based on research showing that minimal reaction times for simple categorization, arguably easier than our tasks, are slightly above that threshold (Joubert et al. 2007). Because this exclusion involved very few observations (ranging from.15% in study 1 to.32% in study 3), it did not alter the pattern of significance in any of our studies and we therefore always present the results for all responses. Each participant offered multiple responses, so the assumption of independence of observations does not hold, and we used a multilevel model to account for the dependency in the observations from each participant. In a multilevel logistic regression (which includes a random intercept to account for variance in liking across participants and thus for dependency in the observations from the same participant), we regressed the likelihood of evaluating a toy positively on the response mode condition, the toy’s orientation, and their two-way interaction. We obtained a main effect of orientation, Wald χ²(1) = 4.85, p = .028, indicating more liking when the toy was rightward oriented (mean probability of liking, abbreviated hereafter as P = .75) rather than leftward oriented (P = .70). The only other significant effect was the expected two-way interaction effect between the response mode condition and product orientation, Wald χ²(1) = 14.44, p < .001 (figure 1). As predicted, we found an effect of orientation in the swipe condition, Wald χ²(1) = 13.45, p < .001 (Pleftward = .64, Prightward = .80), but not in the button press condition, Wald χ²(1) = 1.93, p =.16 (Pleftward = .75, Prightward = .70). FIGURE 1 View largeDownload slide MEAN ESTIMATED LIKING AS A RESULT OF PRODUCT ORIENTATION AND RESPONSE MODE FIGURE 1 View largeDownload slide MEAN ESTIMATED LIKING AS A RESULT OF PRODUCT ORIENTATION AND RESPONSE MODE These results illustrate that object orientation guides the selection of directional action and correspondingly affects product evaluations. Because this study includes a condition in which respondents expressed their evaluations through button pressing rather than swiping, we infer that product orientation affects lateral directional actions only, leaving actions with no lateral directional dimension unaffected. This favors a direction (over a location) account for the effect of product orientation. STUDY 2: THE ROLE OF SPACE-VALENCE MAPPING Most apps that require swiping movements for evaluations associate the left side with a negative evaluation and the right side with a positive evaluation; this study therefore investigates whether a similar effect occurs when positive evaluations entail leftward swiping movements. If the previously demonstrated effect originates because directionally compatible movements are facilitated by product orientations, leftward-oriented pictures should receive more positive evaluations if those positive evaluations are expressed through leftward swiping movements, rather than through pressing a button on the left side of the screen. That is, we test whether the effects obtained for swiping in study 1 reverse when the space-valence mapping reverses. Method In this laboratory study, 126 students of Ghent University (men N = 56, MAge = 23.23 years, SD = 4.97; women N = 70, MAge = 23.11 years, SD = 5.99) participated in return for monetary compensation. They sat in front of touchscreen computers and were randomly assigned to one of four between-subjects conditions. The first between-subjects manipulated factor was response mode, to express (dis)liking (swiping vs. button pressing). The second factor related to the space-valence mapping, such that half of the respondents saw a green checkmark on the right side of the screen and a red “x” on the left side (i.e., right = positive), and the other half saw a screen with the green checkmark on the left side and the red “x” on the right (i.e., left = positive). As in study 1, product orientation was manipulated within subjects: participants observed pictures of shoes with the toe oriented to the left or right. Men saw men’s shoes, and women saw women’s shoes (see appendix B). The instructions explained that participants were seated in front of a touchscreen and should complete the questionnaire by using it as an input device, rather than a mouse. They further read that 30 pictures of shoes would appear on the screen, and they should indicate whether they liked or disliked each pair. The shoes, presented in fixed order, appeared in the center of the screen, and each shoe’s rightward or leftward orientation was determined randomly. As each picture appeared, participants in the button condition could like it by pressing the green checkmark, presented on the right or left side of the screen; participants in the swipe condition were instructed to give a positive evaluation by swiping the picture to the checkmark. If they wanted to evaluate the presented shoe negatively, they pressed the red “x,” presented on the left or right side of the screen in the button condition, or swiped to the red “x” in the swipe condition. Results and Discussion Because each of the 126 participants provided 30 responses, we obtained 3,780 responses. The time to make a decision varied from 4 milliseconds to 23 seconds. We conducted a multilevel logistic regression with the binary evaluation (like/dislike) of the presented shoe as the dependent variable and response mode, shoe orientation, valence mapping, and gender, together with all their interactions, as the independent variables. We observed a main effect of gender, Wald χ2(1) = 5.48, p = .02, such that women liked the women’s shoes more (P = .44) than men liked the men’s shoes (P = .37). Yet gender did not interact with any of the remaining factors; all Wald χ2(1) < .73, and all ps > .39. We observed a main effect of valence mapping, Wald χ²(1) = 26.28, p < .001. Shoes were evaluated more favorably when the positive responses were on the right (P = .48) instead of the left (P = .33) side of the screen. This finding is in line with the observation that most people associate the right side with “good” (Casasanto and Chrysikou 2011). The valence mapping effect was qualified by an interaction with shoe orientation, Wald χ²(1) = 24.42, p < .001. When the right side represented a positive evaluation, a shoe was more liked if it was rightward (P = .53) rather than leftward (P = .43) oriented. If the right side represented a negative evaluation, a shoe was less liked if it was rightward (P = .28) rather than leftward (P = .37) oriented. In line with our theorizing, this interaction of valence mapping with shoe orientation was further qualified by response mode, Wald χ²(1) = 19.69, p < .001 (figure 2). As before, we observed a directional congruency effect for swiping, Wald χ²(1) = 39.96, p < .001, but not for button presses, Wald χ²(1) = .14, p = .70. For positive responses on the right, shoes were more preferred if they were rightward oriented (P = .58) instead of leftward oriented (P = .39, p < .001) in the swiping condition, but they were equally preferred in the button press condition (Pleftward = .46, p = .57, Prightward = .48). Conversely, if positive responses were on the left, leftward-oriented shoes were preferred more (P = .41) than rightward-oriented ones (P = .24, p < .001) in the swiping condition but equally preferred in the button press condition (Pleftward = .33, Prightward = .33, p = .91). FIGURE 2 View largeDownload slide MEAN ESTIMATED LIKING AS A RESULT OF SHOE ORIENTATION, RESPONSE MODE, AND SPACE-VALENCE MAPPING FIGURE 2 View largeDownload slide MEAN ESTIMATED LIKING AS A RESULT OF SHOE ORIENTATION, RESPONSE MODE, AND SPACE-VALENCE MAPPING These results support the idea that object orientation guides swiping responses and, correspondingly, affects reported product evaluations. Moreover, by manipulating whether positive evaluations appear on the right or left side of the screen, we illustrate that the effect is not an artifact, limited to one side of the screen. This clear evidence reveals that people are inclined to engage in congruent rather than incongruent directional actions. STUDY 3: INFLUENCE OF MULTIPLE ORIENTATION CUES ON SWIPING DIRECTION The results of studies 1 and 2 indicate that a horizontal orientation cues a response in the congruent lateral direction and thus affects swipes, but not button presses. If these effects occur because the horizontal orientation primes a congruent direction, the effect should be eliminated if a stimulus contains a directional cue in the opposite direction, because the dimensional overlap would not be unequivocally established in this case. To test this hypothesis, we used three sets of pictogram stimuli (see appendix B). The first set contained static images (e.g., someone sitting behind a desk). From a theoretical perspective, they are similar to the stimuli of studies 1 and 2. A second set of stimuli contained nonstatic images of someone moving in the same direction as implied by the orientation (e.g., a cyclist). These pictograms contain two congruent directional cues, so we anticipate replicating the effect of orientation on swiping but not button presses. A priori, it is unclear whether the combination of two congruent directional cues will strengthen the observed effects. Finally, a third set of stimuli contained nonstatic images of someone moving in the opposite direction implied by the orientation (e.g., a man falling on his back). Because these pictograms contain opposing directional cues, we expect the orientation effect for swiping to be reduced or even eliminated. In summary, for static and congruent nonstatic images, we expect an orientation effect for swiping responses but not for button presses. For incongruent nonstatic images, we expect to find no orientation effect for either swiping or button presses. Method In this study, 185 students of Ghent University (65 men, MAge = 21.63, SD = 4.18) evaluated the aesthetic appeal of 12 pictograms for monetary compensation. Each participant was seated in front of a touchscreen and evaluated the pictograms by either swiping or pressing an onscreen button. In both response mode conditions, a green checkmark (i.e., like button) was on the right side of the screen, and a red “x” (i.e., dislike button) appeared on the left. All participants evaluated all 12 pictograms, presented in a fixed order, but for each trial, whether the picture appeared in its original or mirror-reversed orientation was randomly determined; that is, the exact same pictograms were shown to all participants, but in varying orientation. Four pictograms presented human figures with a clear leftward or rightward orientation, easily inferred from static orientation cues such as head and body orientation (e.g., a person seated at a desk). Four other pictograms contained multiple congruent orientation cues, such that orientation and implied movement were in the same direction (e.g., a person looking in the same direction as walking or riding a bike). A final set of four pictograms revealed multiple, incongruent orientation cues, where orientation and implied movement were in the opposite direction (e.g., a person looking in one direction but pulling an object in the opposite direction). Results and Discussion In the multilevel logistic regression model (with a random intercept to account for variation among participants’ liking), we regressed the likelihood of evaluating a pictogram positively on response mode, pictogram orientation, and pictogram type, together with all their interactions. We obtained a main effect of pictogram type, Wald χ²(2) = 103.71, p < .001, indicating that static images tended to be somewhat liked (P = .56), as did incongruent images (P = .56), but congruent images tended to be disliked (P = .38). This main effect was qualified by a pictogram orientation × pictogram type interaction, Wald χ²(2) = 7.65, p = .022. Orientation affected liking for static images, Wald χ²(1) = 4.67, p = .031 (Pleftward = .52, Prightward = .60), and for congruent images, Wald χ²(1) = 4.45, p = .036 (Pleftward = .34, Prightward = .42) but not for incongruent images, Wald χ²(1) = .97, p = .33 (Pleftward = .58, Prightward = .54). Figure 3 depicts the responses, broken down by response mode, pictogram orientation, and pictogram type. The three-way interaction among the three independent variables did not reach a conventional level of significance, Wald χ²(2) = 5.78, p = .056, but the results are in line with our predictions. In support of our argument, the effect of orientation differs across the three types of images in the swiping condition, Wald χ²(2) = 11.34, p = .004, but not in the button press condition, Wald χ²(2) = .18, p = .91. In the button press condition, we also observed no overall effect of image orientation, Wald χ²(1) = .48, p = .49. In the swiping condition, we uncovered an effect of image orientation for static images, Wald χ²(1) = 11.13, p < .001 (Pleftward = .49, Prightward = .65), and for congruent images, Wald χ²(1) = 10.39, p < .001 (Pleftward = .29, Prightward = .46), but not for incongruent ones, Wald χ²(1) = .47, p = .49 (Pleftward = .58, Prightward = .55). FIGURE 3 View largeDownload slide MEAN ESTIMATED LIKELIHOOD OF LIKING A PRODUCT AS A RESULT OF PICTOGRAM ORIENTATION, RESPONSE MODE, AND PICTOGRAM TYPE FIGURE 3 View largeDownload slide MEAN ESTIMATED LIKELIHOOD OF LIKING A PRODUCT AS A RESULT OF PICTOGRAM ORIENTATION, RESPONSE MODE, AND PICTOGRAM TYPE At first glance, the lower likelihood of a liking response for congruent versus incongruent images may seem inconsistent with our hypotheses. However, each pictogram type was operationalized by a different set of pictograms, so comparisons of the responses to the different pictogram types clearly are confounded by the main effect of pictogram type. The congruent pictograms simply were not liked much (as also conveyed by the means in the corresponding button conditions). A more precise test of our hypotheses instead requires comparisons within each pictogram type and we have organized figure 3 accordingly. All these comparisons support our hypotheses. The results of study 3 thus show that product orientation affects swiping responses when there is a single orientation cue or when multiple directional cues are congruent, but not when a stimulus contains multiple, incongruent orientation cues. STUDY 4: ONE-DIRECTIONAL VERSUS BIDIRECTIONAL RESPONSES Study 4 has two main goals: to show that not only product evaluations but also other, more downstream variables can be affected by product orientation and to attest to the pivotal role of dimensional overlap between an irrelevant stimulus feature (orientation) and valid responses (swiping) for prompting the observed orientation effect. Study 4 therefore focuses on consumers’ expressions of willingness to pay (WTP) for products. Furthermore, SRC research establishes that dimensional overlap can be established successfully only if the response alternatives vary on the overlapping dimension. For study 4, participants convey their willingness to pay on a slider scale with a custom start position, such that the slider tool is positioned either in the center of the scale, enabling participants to drag it rightward or leftward, or on the extreme left, allowing only rightward movements of the slider. While the effect of product orientation on swiping direction and WTP should persist when the slider tool is centered on the scale, it should be eliminated when it is positioned on the extreme left. Method In exchange for monetary compensation, 184 students of Ghent University (76 men, MAge = 21.59, SD = 1.29) participated in a laboratory study in which they had to complete several unrelated tasks. For the present study, participants viewed eight pictures of gadgets, each with a front side that was unambiguously located to the left or right (see appendix B). We created two versions, leftward and rightward oriented, for each gadget, then determined at random whether the participant would see the original or mirror-reversed version. The participants indicated their willingness to pay for each gadget on a slider scale, ranging from 0 to 20 euro, by dragging the slider tool on a touchscreen to the desired position. A between-subjects manipulation altered whether the slider tool initially appeared in the middle (10 euro) or left side (0 euro) of the scale. The former condition requires consumers to swipe to the left (right) to express a lower (higher) willingness to pay, but the latter allows rightward responses only. Results and Discussion We ran a multilevel regression model regressing slider tool location (left vs. center), product orientation, and their interaction term, onto willingness to pay. We estimated the error degrees of freedom of the statistical tests using Satterthwaite’s approximation, which may produce fractional degrees of freedom, to adjust for heterogeneity in error variances across conditions (Satterthwaite 1946). The results indicated an interaction between product orientation and the initial position of the slider, F(1, 1459.63) = 4.49, p = .034. When the slider initially appeared in the middle of the scale, participants were willing to pay more for rightward-oriented products than leftward-oriented ones, F(1, 1466.57) = 9.34, p = .002 (Mleftward = 5.14, Mrightward = 6.29). When it initially appeared to the left, willingness to pay was not significantly influenced by the leftward or rightward orientation of the products, F(1, 1443.65) = .001, p = .97 (Mleftward = 5.40, Mrightward = 5.41). We also recoded willingness to pay as a binary measure of leftward swipes (i.e., WTP ≤ 10) versus rightward swipes (i.e., WTP > 10) for the condition with the centered slider. In a multilevel binary regression model, the effects of orientation and initial slider position on this dichotomous variable revealed similar patterns. Product orientation did not influence participants’ likelihood of indicating a WTP greater than 10 when the slider began on the left side of the scale, Wald χ²(1) = .21, p = .64 (Pleftward = .18, Prightward = .16), but this likelihood increased for rightward-oriented (Prightward = .22) versus leftward-oriented (Pleftward = .17) products when the slider was centered initially, χ²(1) = 3.52, p = .061. These findings confirm our expectation that response alternatives must vary on the overlapping dimension, because only when responses in both directions are possible is stimulus orientation able to cue a directionally congruent response. These results indicate that dimensional overlap underlies the effect of orientation on evaluations, according to a moderation-of-process approach (Spencer, Zanna, and Fong 2005). If variation in responses is required for SRC effects to arise, one might wonder if variation is also required on the stimulus side (see web appendix A for an empirical test). Study 4 also shows that the effect is not tied to any specific outcome measure: both product evaluations and more downstream dependent variables, such as willingness to pay, can be affected by product orientations. This finding makes sense in that we posit that the effect occurs at the behavioral, not the evaluative, level—that is, the orientation effect reflects an increased likelihood of swiping in a corresponding direction, rather than a genuine effect on evaluations. Nevertheless, orientation might not affect evaluations for all stimuli. Web appendix B does show that the effect of product orientation is very robust and is eliminated only for extremely (un)attractive stimuli. Moreover, as hypothesized, the effect of orientation may also differ for subjective versus objective judgments; this is tested in study 5. STUDY 5: INHIBITION OF ORIENTATION-CONGRUENT RESPONSES IN OBJECTIVE BUT NOT SUBJECTIVE JUDGMENTS The preceding studies show that product orientation activates a congruent response, but we suggest that the process that determines whether consumers execute the activated response action to convey their evaluations differs essentially from the process that emerges in typical SRC studies. The essential difference between the current paradigm and typical SRC studies is that we focus on subjective judgments, while typical SRC studies involve objective ones. A typical SRC paradigm establishes an unambiguous definition of the (in)correct answer that allows participants to recognize incorrect responses, which they should inhibit. Participants would select an erroneous response only if they fail to inhibit it. This is unlikely for slower responses, because they have taken more time to halt their incorrect action. So, for objective judgments (e.g., classifying a toy as an animal or a car), we should observe interference due to product orientation only under time pressure but not when no time pressure is applied. Subjective evaluations (e.g., “Do you like the presented toy as a children’s gift?”), however, are an entirely different matter. The correct responses are not clear, so consumers may not consider inhibiting any activated response, irrespective of the time frame. With study 5, we attempt to show that the influence of orientation on objective judgments that are conveyed via swiping movements is moderated by time pressure, but time pressure has no such effect for subjective evaluations. Method The study was completed by 96 students for partial course fulfillment, in the consumer laboratory of Ghent University (42 men; MAge = 20.80 years, SD = 1.41). The participants were seated in isolated cubicles in front of desktop computers connected to touchscreens. The task consisted of two parts. First, participants offered subjective evaluations, similar to the swiping condition from study 1, of the attractiveness of 30 pictures of toys. The toys appeared with a randomly determined leftward or rightward orientation. Second, the participants had to specify whether each toy represented an animal or not; 17 of the 30 stimuli (57%) were animals. The judgment type (subjective vs. objective) and product orientation (left vs. right) thus were manipulated within participants. To test whether the effect of orientation on swiping responses was moderated by time pressure in objective, but not subjective, evaluations, we included time pressure as a between-subjects factor, such that half of the participants had to offer their responses within 1,000 milliseconds (a common cutoff in analyses of response latency tasks), but the other half did not have any restriction on their response times. Results and Discussion This study yielded 5,760 data lines for analysis in that the 96 participants each completed 60 trials. However, in the time pressure condition, participants failed to respond in time to 405 trials, so we obtained 5,355 usable responses. We ran a multilevel logistic regression analysis, with the binary direction of the swipe movement (rightward swipes convey likes or identifications of animals) as the dependent variable and judgment type, toy orientation, and time pressure, together with their interactions, as independent variables. The three-way interaction was significant, Wald χ²(1) = 4.04, p = .044: toys were subjectively judged to be more attractive when oriented to the right (P = .61) rather than left (P = .56), Wald χ²(1) = 7.74, p = .005, irrespective of the required response speed, Wald χ²(1) = .303, p = .58. The effect of toy orientation on objective judgments varied with the time pressure, though, Wald χ²(1) = 3.37, p = .066. Specifically, toy orientation did not affect objective judgments in the absence of time pressure, Wald χ²(1) = .44, p = .51 (Pleftward = .59, Prightward = .57), but the correct identification was distorted by the product orientation when participants faced time pressure, Wald χ²(1) = 3.89, p = .049 (Pleftward = .55, Prightward = .60). As summarized in figure 4, these results affirmed our expectations; response speed did not alter the impact of stimulus orientation for subjective judgments, but it had a critical influence on objective judgments. This difference likely reflects participants’ failure to realize the necessity of inhibiting the response activated by stimulus orientation when conveying their liking, even though their responses were affected by the irrelevant feature of product orientation. A study in web appendix C shows that making participants aware of this potential influence eliminates it. FIGURE 4 View largeDownload slide MEAN ESTIMATED LIKELIHOOD TO SWIPE RIGHT AS A RESULT OF PRODUCT ORIENTATION, TIME PRESSURE, AND JUDGMENT TYPE FIGURE 4 View largeDownload slide MEAN ESTIMATED LIKELIHOOD TO SWIPE RIGHT AS A RESULT OF PRODUCT ORIENTATION, TIME PRESSURE, AND JUDGMENT TYPE GENERAL DISCUSSION Consumers increasingly use mobile devices to evaluate and purchase products and services, which makes it important to understand how their interactions with these devices might shape their evaluations and preferences. This article focuses on two common responses: swiping and button presses. On the basis of dimensional overlap theory (Kornblum et al. 1990; Zhang et al. 1999), we predict and find that horizontal product orientation affects swiping responses but not button presses. Specifically, consumers tend to swipe in the direction of the product orientation (studies 1 and 2). The effect of product orientation occurs at the behavioral rather than evaluative level. Consequently, when the space-valence mapping is reversed (study 2), rightward-oriented stimuli still elicit rightward swiping responses, which then signify dislike. The effect unfolds because (1) stimulus–response overlap serves as a source of response action priming (studies 3 and 4), (2) rendering consumers more likely to execute the primed response because they view their response as legitimate and do not experience the need to inhibit their orientation-congruent response (study 5). Finally, we observe orientation effects for both liking and willingness-to-pay judgments, attesting to their fundamental nature (study 4). This study contributes to extant research into how irrelevant visual product presentation cues can influence product judgments. Prior research has addressed the effect of horizontal positioning, in advertising (Chae and Hoegg 2013; Janiszewski 1990; Peracchio and Meyers-Levy 1997), retail displays (Chandon et al. 2009; Christenfeld 1995; Valenzuela and Raghubir 2009; Valenzuela, Raghubir, and Mitakakis 2013), and product packaging (Deng and Kahn 2009), to establish that this irrelevant product presentation cue provides information that consumers use in their judgments. We extend this research stream by showing that another irrelevant cue for judgment, horizontal product orientation, also can affect product evaluations if they are conveyed through swiping movements. Moreover, the mechanism that underlies the orientation effect markedly differs from the mechanisms that evoke product location effects. Moreover, the current research enriches growing literature on the role of preference assessment procedures in consumer behavior (Krüger, Mata, and Ihmels 2014; Lembregts and Pandelaere 2013; Schley, Lembregts, and Peters 2017; Weijters, Geuens, and Baumgartner 2013) and the particular impact of preference expression modalities (Klesse, Levav, and Goukens 2015; Shen, Zhang, and Krishna 2016). For example, speaking prompts more indulgent choices than manual preference expression modalities (Klesse et al. 2015), and expressing preferences via touch-based interfaces, rather than mouse-driven desktop computers, facilitates affective choices (Shen et al. 2016). Our finding extends this research by comparing button pressing to swiping, a new behavior that consumers adopt to interact with products presented on touch-based interfaces. These findings also relate to prior work on stimulus–response compatibility. The facilitation of certain motor responses by task-irrelevant visual aspects of a stimulus is well documented, such as in the Stroop effect, which describes how people are sometimes affected by the meaning of a word when asked to name the color in which it is depicted (Stroop 1935). Extending this vast research, our study differs from prior investigations in two main respects. Prior research has ignored the impact of direction cues but, as we show, product orientation also cues a directional response, which may open new lines of investigation for SRC phenomena as they relate to horizontal orientations. Some research notes how arrows (Miles and Proctor 2012; Wascher et al. 1999) and arrow primes (Danziger, Kingstone, and Ward 2001) affect response latencies, but these arrows mainly cue location, because they orient visual attention involuntarily (Tipples 2002). Further research could investigate how and why arrows differ from horizontal orientation. In addition, in showing that a response that is primed by a stimulus aspect may inform evaluative responses, rather than focusing on reduced response latencies or erroneous responses in incongruent trials, our study suggests some broader behavioral implications of stimulus aspects. They do not just affect fast, (in)correct responses; they also influence responses that convey some subjective judgment and require more time. Our research also yields insights into the mechanisms that underlie SRC effects and informs the debate about why SRC effects disappear for longer response times. Some explanations suggest that longer response times allow for the more successful execution of inhibitory processes, but another side argues that the activation of a response due to dimensional overlap decays rapidly (Hommel 1997), such that dimensional overlap no longer affects slower responses. Our findings offer support for the former position. We find that for a subjective task, the orientation effect—an SRC effect in its own right—persists even for very long response times. For subjective tasks, people sense no need to instigate inhibitory processes. The presence of an objective criterion in typical SRC tasks presumably leads participants to develop some phenomenological awareness of the bias imposed by the irrelevant feature, thereby encouraging and allowing them to inhibit their illegitimate responses. Finally, we extend extant literature on product orientation in two key ways. First, prior studies define product orientation by handle locations (Eelen et al. 2013; Elder and Krishna 2012; Ping et al. 2009), but we consider the location of the back and front. This approach not only taps into a different aspect of horizontal orientation but also applies to a broader set of products than has been covered by prior research. Second, our focus on a different orientation aspect implies a different process than has been described in prior research. The mechanism underlying the handle location effect involves the metacognitive ease of (simulated) grasping behavior. Grasping behaviors are actions that become associated with products through a learning process. Upon a consumer’s exposure to a product with a handle, a behavioral response (i.e., grasping) gets cognitively activated by the (over)learned association between the stimulus and the behavioral response, and its simulated motor fluency affects evaluative judgments (Alter and Oppenheimer 2009; Reber, Winkielman, and Schwarz 1998; Schwarz 2004). The mechanism underlying horizontal orientation, defined by back and front location, instead involves response priming due to dimensional overlap. To be clear, in the absence of a learned association between the stimulus and the response, responses still might be primed by perceptions of stimuli that share features with the cognitive representation of the behavioral response (Hommel 1997), such as when a product orientation cues a swiping response. A primed response may be easier to execute, rendering its execution more likely. However, in our studies, manipulating the horizontal orientation did not appear to influence how participants felt about a product directly; it merely affected their selected response to convey their binary like/dislike evaluation. This key difference compared with a motor fluency process is most prominent in study 2, in which we manipulated the space-valence mapping. Whether a rightward-oriented shoe was liked better than a leftward-oriented shoe depended critically on which side was associated with the “like” response, a result that is inconsistent with a motor fluency account. With regard to managerial implications, we note that many apps that ask consumers to evaluate objects offer not some random selection of objects but rather a set that users might like, learned from those users’ past responses (i.e., objects more likely to be liked). Any liking response due to object orientation thus may affect which objects appear subsequently. People often look to their own behavior to infer their attitudes and opinions too (Bem 1972), such that they might infer from their own swiping response whether they like an object or not. Such inferences may function to update their prior evaluations. Finally, online evaluation tools often keep track of consumers’ product evaluations, which enables them to inform other consumers about product popularity. Consumers in turn rely on ratings contributed by others to make purchase decisions (Chevalier and Mayzlin 2006; Zhu and Zhang 2010). The perception of success alone can be sufficient to affect product decisions, so even if popularity ratings are not representative of actual product quality, they still can guide decision making (Muchnik, Aral, and Taylor 2013; Salganik and Watts 2008). This trend highlights the importance of recognizing how product orientation can inflate the number of positive evaluations of a product, conveyed through swiping movements. The long-term consequences of a particular swiping response, induced by product orientation, on the preferences of a focal consumer or the preferences of others thus remain as questions for further research. DATA COLLECTION INFORMATION All data were collected in the consumer behavior lab at Ghent University. All studies were scheduled as part of 50 minute or 30 minute sessions in which data for other, unrelated experiments were also collected. The first author and the researchers involved in the unrelated experiments alternated in the supervision of the lab sessions. Data for study 1 were collected in October 2014. Data for study 2 were collected in September 2014. Data for study 3 were collected in October 2014. Data for study 4 were collected in March 2017. For study 5, data were collected in March 2017. Both authors were involved in the design of the studies and the analyses. The final analyses for studies 4, 5, and web appendix C that are reported in the article were conducted by the first author. The final analyses for all other studies were conducted by the second author. The authors wish to thank the editor, Gita Johar; the associate editor, Stijn van Osselaer; and three anonymous reviewers for their feedback on the previous versions of this article. Supplementary materials are included in the web appendix accompanying the online version of this article. The data package is available via doi 10.5061/dryad.rh0v0g2. This research was supported by funding through the Flemish Fund for Scientific Research to the second author (grant G020215N). APPENDIX A FRONT AND BACK VERSUS HANDLE LOCATIONS AS HORIZONTAL ORIENTATION FEATURES Example objects with the front located to the left, suggesting a leftward orientation Example objects with the front located on the left, suggesting a leftward orientation, and a handle located on the right (in other research, defined as a rightward orientation) Example objects with a handle located on the right (in other research, defined as a rightward orientation) APPENDIX B Sample stimuli Studies 1 and 5 (toys) Study 2 (women’s and men’s shoes) Men’s shoes Women’s shoes Men’s shoes Women’s shoes Men’s shoes Women’s shoes Men’s shoes Women’s shoes Study 3 (pictograms) Single orientation cue Multiple congruent orientation cues Multiple incongruent orientation cues Single orientation cue Multiple congruent orientation cues Multiple incongruent orientation cues Single orientation cue Multiple congruent orientation cues Multiple incongruent orientation cues Single orientation cue Multiple congruent orientation cues Multiple incongruent orientation cues Study 4 (gadgets) References Alter Adam L. , Oppenheimer Daniel M. ( 2009 ), “Uniting the Tribes of Fluency to Form a Metacognitive Nation,” Personality and Social Psychology Review , 13 3 , 219 – 35 . Google Scholar CrossRef Search ADS PubMed Ansorge Ulrich , Wühr Peter ( 2004 ), “A Response-Discrimination Account of the Simon Effect,” Journal of Experimental Psychology: Human Perception and Performance , 30 2 , 365 – 77 . Google Scholar CrossRef Search ADS PubMed Areni Charles S. , Duhan Dale F. , Kiecker Pamela ( 1999 ), “Point-of-Purchase Displays, Product Organization, and Brand Purchase Likelihoods,” Journal of the Academy of Marketing Science , 27 4 , 428 – 41 . Google Scholar CrossRef Search ADS Atalay A. Selin , Bodur H. Onur , Rasolofoarison Dina ( 2012 ), “Shining in the Center: Central Gaze Cascade Effect on Product Choice,” Journal of Consumer Research , 39 4 , 848 – 66 . Google Scholar CrossRef Search ADS Bem Daryl J. ( 1972 ), “Self-Perception Theory,” in Advances in Experimental Social Psychology , Vol. 6 , ed. Berkowitz Leonard , New York : Academic Press , 1 – 62 . Brasel S. Adam , Gips James ( 2014 ), “Tablets, Touchscreens, and Touchpads: How Varying Touch Interfaces Trigger Psychological Ownership and Endowment,” Journal of Consumer Psychology , 24 2 , 226 – 33 . Google Scholar CrossRef Search ADS Buchanan Lauranne , Simmons Carolyn J. , Bickart Barbara A. ( 1999 ), “Brand Equity Dilution: Retailer Display and Context Brand Effects,” Journal of Marketing Research , 36 3 , 345 – 55 . Google Scholar CrossRef Search ADS Cacioppo John T. , Priester Joseph R. , Berntson Gary G. ( 1993 ), “Rudimentary Determinants of Attitudes: II. Arm Flexion and Extension Have Differential Effects on Attitudes,” Journal of Personality and Social Psychology , 65 1 , 5 . Google Scholar CrossRef Search ADS PubMed Casasanto Daniel , Chrysikou Evangelia G. ( 2011 ), “When Left Is ‘Right’: Motor Fluency Shapes Abstract Concepts,” Psychological Science , 22 4 , 419 – 22 . Google Scholar CrossRef Search ADS PubMed Castro Iana A. , Morales Andrea C. , Nowlis Stephen M. ( 2013 ), “The Influence of Disorganized Shelf Displays and Limited Product Quantity on Consumer Purchase,” Journal of Marketing , 77 4 , 118 – 33 . Google Scholar CrossRef Search ADS Cavazza Nicoletta , Gabrielli Veronica ( 2015 ), “Affordant Shapes of Product Holder Influence Product Evaluation and Purchase Intention,” Current Psychology , 34 2 , 447 – 65 . Google Scholar CrossRef Search ADS Chae Boyoun (Grace) , Hoegg JoAndrea ( 2013 ), “The Future Looks ‘Right’: Effects of the Horizontal Location of Advertising Images on Product Attitude,” Journal of Consumer Research , 40 2 , 223 – 38 . Google Scholar CrossRef Search ADS Chandon Pierre , Hutchinson Wesley J. , Bradlow Eric T. , Young Scott H. ( 2009 ), “Does In-Store Marketing Work? Effects of the Number and Position of Shelf Facings on Brand Attention and Evaluation at the Point of Purchase,” Journal of Marketing , 73 6 , 1 – 17 . Google Scholar CrossRef Search ADS Chevalier Judith A. , Mayzlin Dina ( 2006 ), “The Effect of Word of Mouth on Sales: Online Book Reviews,” Journal of Marketing Research , 43 3 , 345 – 54 . Google Scholar CrossRef Search ADS Christenfeld Nicholas ( 1995 ), “Choices from Identical Options,” Psychological Science , 6 1 , 50 – 5 . Google Scholar CrossRef Search ADS Danziger Shai , Kingstone Alan , Ward Robert ( 2001 ), “Environmentally Defined Frames of Reference: Their Time Course and Sensitivity to Spatial Cues and Attention,” Journal of Experimental Psychology: Human Perception and Performance , 27 2 , 494 – 503 . Google Scholar CrossRef Search ADS PubMed De Houwer Jan ( 1998 ), “The Semantic Simon Effect,” Quarterly Journal of Experimental Psychology: Section A , 51 3 , 683 – 8 . Google Scholar CrossRef Search ADS De Houwer Jan ( 2003 ), “On the Role of Stimulus-Response and Stimulus-Stimulus Compatibility in the Stroop Effect,” Memory & Cognition , 31 3 , 353 – 9 . Google Scholar CrossRef Search ADS PubMed De Houwer Jan , Eelen Paul ( 1998 ), “An Affective Variant of the Simon Paradigm,” Cognition & Emotion , 12 1 , 45 – 62 . Google Scholar CrossRef Search ADS Deng Xiaoyan , Kahn Barbara E. ( 2009 ), “Is Your Product on the Right Side? The ‘Location Effect’ on Perceived Product Heaviness and Package Evaluation,” Journal of Marketing Research , 46 6 , 725 – 38 . Google Scholar CrossRef Search ADS Eder Andreas B. , Rothermund Klaus ( 2008 ), “When Do Motor Behaviors (Mis)Match Affective Stimuli? An Evaluative Coding View of Approach and Avoidance Reactions,” Journal of Experimental Psychology: General , 137 2 , 262 . Google Scholar CrossRef Search ADS PubMed Eelen Jiska , Dewitte Siegfried , Warlop Luk ( 2013 ), “Situated Embodied Cognition: Monitoring Orientation Cues Affects Product Evaluation and Choice,” Journal of Consumer Psychology , 23 4 , 424 – 33 . Google Scholar CrossRef Search ADS Elder Ryan S. , Krishna Aradhna ( 2012 ), “The ‘Visual Depiction Effect’ in Advertising: Facilitating Embodied Mental Simulation through Product Orientation,” Journal of Consumer Research , 38 6 , 988 – 1003 . Google Scholar CrossRef Search ADS Ellis Rob , Tucker Mike ( 2000 ), “Micro‐Affordance: The Potentiation of Components of Action by Seen Objects,” British Journal of Psychology , 91 4 , 451 – 71 . Google Scholar CrossRef Search ADS PubMed Eriksen Barbara A. , Eriksen Charles W. ( 1974 ), “Effects of Noise Letters upon the Identification of a Target Letter in a Nonsearch Task,” Perception & Psychophysics , 16 1 , 143 – 9 . Google Scholar CrossRef Search ADS Fiore Ann Marie , Yah Xinlu , Yoh Eunah ( 2000 ), “Effects of a Product Display and Environmental Fragrancing on Approach Responses and Pleasurable Experiences,” Psychology and Marketing , 17 1 , 27 – 54 . Google Scholar CrossRef Search ADS Gibson James J. ( 1979 ), The Ecological Approach to Visual Perception , Boston : Houghton Mifflin . Hommel Bernhard ( 1993 ), “The Relationship between Stimulus Processing and Response Selection in the Simon Task: Evidence for a Temporal Overlap,” Psychological Research , 55 4 , 280 – 90 . Google Scholar CrossRef Search ADS Hommel Bernhard ( 1997 ), “Toward an Action-Concept Model of Stimulus-Response Compatibility,” Advances in Psychology , 118 , 281 – 320 . Google Scholar CrossRef Search ADS Janiszewski Chris ( 1990 ), “The Influence of Print Advertisement Organization on Affect toward a Brand Name,” Journal of Consumer Research , 17 1 , 53 – 65 . Google Scholar CrossRef Search ADS Joubert Olivier R. , Rousselet Guillaume A. , Fize Denis , Fabre-Thorpe Michèle ( 2007 ), “Processing Scene Context: Fast Categorization and Object Interference,” Vision Research , 47 26 , 3286 – 97 . Google Scholar CrossRef Search ADS PubMed Keller Carmen , Markert Franziska , Bucher Tamara ( 2015 ), “Nudging Product Choices: The Effect of Position Change on Snack Bar Choice,” Food Quality and Preference , 41 , 41 – 3 . Google Scholar CrossRef Search ADS Klesse Anne-Kathrin , Levav Jonathan , Goukens Caroline ( 2015 ), “The Effect of Preference Expression Modality on Self-Control,” Journal of Consumer Research , 42 4 , 535 – 50 . Kornblum Sylvan , Hasbroucq Thierry , Osman Allen ( 1990 ), “Dimensional Overlap: Cognitive Basis for Stimulus-Response Compatibility—A Model and Taxonomy,” Psychological Review , 97 2 , 253 . Google Scholar CrossRef Search ADS PubMed Krüger Tobias , Mata André , Ihmels Max ( 2014 ), “The Presenter’s Paradox Revisited: An Evaluation Mode Account,” Journal of Consumer Research , 41 4 , 1127 – 36 . Google Scholar CrossRef Search ADS Lembregts Christophe , Pandelaere Mario ( 2013 ), “Are All Units Created Equal? The Effect of Default Units on Product Evaluations,” Journal of Consumer Research , 39 6 , 1275 – 89 . Google Scholar CrossRef Search ADS MacDonald Angus W. , Cohen Jonathan D. , Andrew Stenger V. , Carter Cameron S. ( 2000 ), “Dissociating the Role of the Dorsolateral Prefrontal and Anterior Cingulate Cortex in Cognitive Control,” Science , 288 5472 , 1835 – 8 . Google Scholar CrossRef Search ADS PubMed Melara Robert D. , Marks Lawrence E. ( 1990 ), “Interaction among Auditory Dimensions: Timbre, Pitch, and Loudness,” Perception & Psychophysics , 48 2 , 169 – 78 . Google Scholar CrossRef Search ADS PubMed Miles James D. , Proctor Robert W. ( 2012 ), “Correlations between Spatial Compatibility Effects: Are Arrows More Like Locations or Words?” Psychological Research , 76 6 , 777 – 91 . Google Scholar CrossRef Search ADS PubMed Muchnik Lev , Aral Sinan , Taylor Sean J. ( 2013 ), “Social Influence Bias: A Randomized Experiment,” Science , 341 6146 , 647 – 51 . Google Scholar CrossRef Search ADS PubMed Orth Ulrich R. , Crouch Roberta C. ( 2014 ), “Is Beauty in the Aisles of the Retailer? Package Processing in Visually Complex Contexts,” Journal of Retailing , 90 4 , 524 – 37 . Google Scholar CrossRef Search ADS Peracchio Laura A. , Meyers-Levy Joan ( 1997 ), “Evaluating Persuasion-Enhancing Techniques from a Resource-Matching Perspective,” Journal of Consumer Research , 24 2 , 178 – 91 . Google Scholar CrossRef Search ADS Ping Raedy M. , Dhillon Sonica , Beilock Sian L. ( 2009 ), “Reach for What You Like: The Body’s Role in Shaping Preferences,” Emotion Review , 1 2 , 140 – 50 . Google Scholar CrossRef Search ADS Reber Rolf , Winkielman Piotr , Schwarz Norbert ( 1998 ), “Effects of Perceptual Fluency on Affective Judgments,” Psychological Science , 9 1 , 45 – 8 . Google Scholar CrossRef Search ADS Rodway Paul , Schepman Astrid , Lambert Jordana ( 2012 ), “Preferring the One in the Middle: Further Evidence for the Centre‐Stage Effect,” Applied Cognitive Psychology , 26 2 , 215 – 22 . Google Scholar CrossRef Search ADS Salganik Matthew J. , Watts Duncan J. ( 2008 ), “Leading the Herd Astray: An Experimental Study of Self-Fulfilling Prophecies in an Artificial Cultural Market,” Social Psychology Quarterly , 71 4 , 338 – 55 . Google Scholar CrossRef Search ADS Satterthwaite F. E. ( 1946 ), “An Approximate Distribution of Estimates of Variance Components,” Biometrics Bulletin , 2 6 , 110 – 4 . Google Scholar CrossRef Search ADS PubMed Schley Dan R. , Lembregts Christophe , Peters Ellen ( 2017 ), “The Role of Evaluation Mode on the Unit Effect,” Journal of Consumer Psychology , 27 2 , 278 – 86 . Google Scholar CrossRef Search ADS Schwarz Norbert ( 2004 ), “Metacognitive Experiences in Consumer Judgment and Decision Making,” Journal of Consumer Psychology , 14 4 , 332 – 48 . Google Scholar CrossRef Search ADS Schwarz Norbert ( 2006 ), “Feelings, Fit, and Funny Effects: A Situated Cognition Perspective,” Journal of Marketing Research , 43 1 , 20 – 3 . Google Scholar CrossRef Search ADS Shen Hao , Zhang Meng , Krishna Aradhna ( 2016 ), “Computer Interfaces and the ‘Direct-Touch’ Effect: Can iPads Increase the Choice of Hedonic Food?” Journal of Marketing Research , 53 5 , 745 – 58 . Google Scholar CrossRef Search ADS Simon J. Richard ( 1969 ), “Reactions toward the Source of Stimulation,” Journal of Experimental Psychology , 81 1 , 174 . Google Scholar CrossRef Search ADS PubMed Slovic Paul ( 1995 ), “The Construction of Preference,” American Psychologist , 50 5 , 364 – 71 . Google Scholar CrossRef Search ADS Spencer Steven J. , Zanna Mark P. , Fong Geoffrey T. ( 2005 ), “Establishing a Causal Chain: Why Experiments Are Often More Effective than Mediational Analyses in Examining Psychological Processes,” Journal of Personality and Social Psychology , 89 6 , 845 – 51 . Google Scholar CrossRef Search ADS PubMed Strack Fritz , Martin Leonard L. , Stepper Sabine ( 1988 ), “Inhibiting and Facilitating Conditions of the Human Smile: A Nonobtrusive Test of the Facial Feedback Hypothesis,” Journal of Personality and Social Psychology , 54 5 , 768 – 77 . Google Scholar CrossRef Search ADS PubMed Stroop J. Ridley ( 1935 ), “Studies of Interference in Serial Verbal Reactions,” Journal of Experimental Psychology , 18 6 , 643 . Google Scholar CrossRef Search ADS Tipples Jason ( 2002 ), “Eye Gaze Is Not Unique: Automatic Orienting in Response to Uninformative Arrows,” Psychonomic Bulletin & Review , 9 2 , 314 – 8 . Google Scholar CrossRef Search ADS PubMed Tucker Mike , Ellis Rob ( 2001 ), “The Potentiation of Grasp Types during Visual Object Categorization,” Visual Cognition , 8 6 , 769 – 800 . Google Scholar CrossRef Search ADS Valenzuela Ana , Raghubir Priya ( 2009 ), “Position-Based Beliefs: The Center-Stage Effect,” Journal of Consumer Psychology , 19 2 , 185 – 96 . Google Scholar CrossRef Search ADS Valenzuela Ana , Raghubir Priya ( 2015 ), “Are Consumers Aware of Top–Bottom but Not of Left–Right Inferences? Implications for Shelf Space Positions,” Journal of Experimental Psychology: Applied , 21 3 , 224 – 41 . Google Scholar CrossRef Search ADS PubMed Valenzuela Ana , Raghubir Priya , Mitakakis Chrissy ( 2013 ), “Shelf Space Schemas: Myth or Reality?” Journal of Business Research , 66 7 , 881 – 8 . Google Scholar CrossRef Search ADS Wascher Edmund , Reinhard Meike , Wauschkuhn Bernd , Verleger Rolf ( 1999 ), “Spatial SR Compatibility with Centrally Presented Stimuli: An Event-Related Asymmetry Study on Dimensional Overlap,” Journal of Cognitive Neuroscience , 11 2 , 214 – 29 . Google Scholar CrossRef Search ADS PubMed Weijters Bert , Geuens Maggie , Baumgartner Hans ( 2013 ), “The Effect of Familiarity with the Response Category Labels on Item Response to Likert Scales,” Journal of Consumer Research , 40 2 , 368 – 81 . Google Scholar CrossRef Search ADS Zhang Huazhong (Harry) , Zhang Jun , Kornblum Sylvan ( 1999 ), “A Parallel Distributed Processing Model of Stimulus–Stimulus and Stimulus–Response Compatibility,” Cognitive Psychology , 38 3 , 386 – 432 . Google Scholar CrossRef Search ADS PubMed Zhu Feng , Zhang Xiaoquan ( 2010 ), “Impact of Online Consumer Reviews on Sales: The Moderating Role of Product and Consumer Characteristics,” Journal of Marketing , 74 2 , 133 – 48 . Google Scholar CrossRef Search ADS Zhu Rui , Meyers-Levy Joan ( 2009 ), “The Influence of Self-View on Context Effects: How Display Fixtures Can Affect Product Evaluations,” Journal of Marketing Research , 46 1 , 37 – 45 . Google Scholar CrossRef Search ADS © The Author(s) 2018. Published by Oxford University Press on behalf of Journal of Consumer Research, Inc. All rights reserved. For permissions, please e-mail: 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 Journal of Consumer Research Oxford University Press

Why Are You Swiping Right? The Impact of Product Orientation on Swiping Responses

Loading next page...
 
/lp/ou_press/why-are-you-swiping-right-the-impact-of-product-orientation-on-swiping-kTLqP6BVdy
Publisher
University of Chicago Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of Journal of Consumer Research, Inc. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com
ISSN
0093-5301
eISSN
1537-5277
D.O.I.
10.1093/jcr/ucy013
Publisher site
See Article on Publisher Site

Abstract

Abstract Many apps require consumers to evaluate products by swiping them to the right or left. This work explores whether product orientation affects the product evaluations communicated by swiping movements, compared with those made by pressing onscreen buttons. Building on stimulus-response compatibility (SRC) theory, which suggests that irrelevant product display features can activate certain behavioral responses when the product display and the behavioral response share a common dimension, this study predicts that the horizontal direction (left to right or right to left) cued by a product’s orientation should facilitate a swipe movement in the congruent direction. Five studies indicate that when people use swiping movements to evaluate objects, their evaluations are influenced by the object’s orientation, whereas evaluations conveyed through button presses reveal no orientation effect. The orientation effect for swiping responses also disappears when the objects contain a direction cue that is incongruent with their orientation, and when only one directional swipe movement is defined as a valid response option. Moreover, the effect holds for subjective evaluations but is eliminated for objective judgments, when these involve no time pressure. affordance, product orientation, touchscreen, tablet, mobile device, stimulus–response compatibility With the increasingly widespread adoption of tablets and smartphones, conventional interfaces are shifting from computer mice to touchscreens (Brasel and Gips 2014), creating new ways to interact with objects presented onscreen. For example, with the online dating app Tinder, users can like or dislike prospective dates by swiping their profile pictures to the right or left of the screen. The Stylect app helps women find the perfect shoes by showing options and encouraging them to swipe to the right if they like them or to the left if not. This conversion to touchscreen interfaces has made swiping an increasingly common movement, used to interact with products and stimuli onscreen. Accordingly, it is important to assess how the adoption of new product interaction behaviors (like swiping) affects product evaluations, especially when we consider that more than 8% of e-commerce website visits already involve tablets (Brasel and Gips 2014), and this rate seems likely to increase in the future. The current study investigates in particular how product orientation might affect liking evaluations conveyed through swiping. We propose that an object’s horizontal orientation (i.e., whether its front side is located to the left or right) primes a lateral direction, and people are more inclined to swipe in a direction congruent with the primed direction than in an incongruent direction. For example, when using the Stylect app, women may be more inclined to engage in a rightward swipe if the toe of the presented shoe is directed to the right, whereas an opposite swiping movement may be preferred if the toe of the same shoe were oriented to the left. A positive evaluation typically is conveyed with a left-to-right swipe, so rightward-oriented products may have a better chance of being liked than leftward-oriented ones. The current research contributes to two main streams of research. First, we extend research on the effects of product displays on product evaluations, by focusing on a distinct aspect of horizontal product orientation. Specifically, most extant literature dealing with horizontal product orientation addresses handle locations (Eelen, Dewitte, and Warlop 2013; Elder and Krishna 2012; Ping, Dhillon, and Beilock 2009), but we instead consider the location of the product’s front. By changing the focal element that defines the horizontal orientation, our research is relevant for a more expanded set of objects—namely, any object with a clear back and front—rather than the more restrictive set of objects with handles (see appendix A for illustrative examples). More importantly, the mechanism that explains the impact of handle location is unique to products with handles, so we also introduce a different mechanism by which product orientation—and product display features in general—affects evaluations. Specifically, we draw on stimulus–response overlap research (Kornblum, Hasbroucq, and Osman 1990) that describes how irrelevant physical features of a stimulus can interfere with proper responses to it. This body of research suggests that a behavioral response congruent with an irrelevant feature of a stimulus gets activated when both the stimulus and response vary along an overlapping dimension. We demonstrate in turn that dimensional overlap not only affects performance in response latency tasks but also serves as a source of bias for consumer evaluations. We thus specify a new route through which product displays can affect product liking. Second, our findings extend stimulus–response compatibility (SRC) research in several ways. Previous research has investigated features such as color (Stroop 1935), horizontal (Simon 1969) and vertical (Ansorge and Wuhr 2004) positions, surrounding letters (Eriksen and Eriksen 1974), pitch (Melara and Marks 1990), valence (De Houwer and Eelen 1998), and semantic meaning (De Houwer 1998). The previously studied responses include verbal responses (e.g., color naming, Stroop 1935), keypresses (most studies), and sagittal lever movements (Eder and Rothermund 2008). In a clear distinction, we investigate the potential interference of a hitherto unstudied irrelevant physical feature (i.e., horizontal orientation) and a neglected response (i.e., lateral swiping). This irrelevant feature of horizontal orientation also is less directly connected to the overlapping response dimension (direction) than is the case in typical SRC tasks. The objects do not move leftward or rightward in our study, so the overlap is more covert in nature. Finally, while typical SRC tasks involve judgments that can be established as objectively correct or incorrect, the current research considers subjective judgments like product evaluations. This shift in focus has implications for the impact of irrelevant stimulus features, especially for how this impact develops over time. In typical SRC tasks, participants try to inhibit the response activated by the influence of an irrelevant stimulus dimension when it leads to an incorrect response. Occasionally, when responses are very fast, participants might fail to do so. In contrast, subjective judgments lack an objectively correct response, rendering the issue of inhibiting an incorrect response moot. This lack of inhibition implies that SRC effects for subjective judgments are not limited to fast responses. THE INFLUENCE OF PRODUCT DISPLAY ON EVALUATIONS In addition to the way a product looks, its display also can affect its evaluation (Areni, Duhan, and Kiecker 1999; Buchanan, Simmons, and Bickart 1999; Castro, Morales, and Nowlis 2013; Cavazza and Gabrieli 2015; Fiore, Yah, and Yoh 2000; Zhu and Meyers-Levy 2009). For example, consumers evaluate a product more positively when it is placed centrally (Atalay, Bodur, and Rasolofoarison 2012; Keller, Markert, and Bucher 2015; Rodway, Schepman, and Lambert 2012; Valenzuela and Raghubir 2009) or at the top (Valenzuela and Raghubir 2015) in a visual display, as well as in environments that are less complex (Orth and Crouch 2014). Product display matters not only when products are surrounded by other products but also for (images of) products in isolation. Various studies show that product evaluations also can be affected by arbitrary product orientations. For example, products with their handle oriented to the right are rated as more attractive than their leftward-oriented counterparts (Eelen et al. 2013; Elder and Krishna 2012; Ping et al. 2009). Studies showing that handle location impacts product evaluations build on affordance theory (Gibson 1979), according to which the perception of objects can activate object interaction movements that people commonly execute when encountering the object. For instance, seeing pencils and cherries facilitates the use of a precision grip in a classification task; exposure to hammers and apples instead facilitates the use of a power grip in a similar task (Ellis and Tucker 2000; Tucker and Ellis 2001). Similarly, a handle’s position activates grasping with the corresponding hand (i.e., a handle on the left activates left-hand grasping). Depending on the person’s dominant handedness, the grasping behavior activated by the handle location also is relatively easier or more difficult to execute. The motor fluency (or lack thereof) associated with the simulated grasping behavior then affects liking for a presented, graspable object (Eelen et al. 2013; Elder and Krishna 2012; Ping et al. 2009). Merely observing an ad featuring a bowl of yogurt with a spoon presented on the right (vs. left) side of the bowl increases purchase intentions among right-handed participants, for example (Elder and Krishna 2012). Rather than handle location, we focus on a different, more general, element of product orientation that might affect product evaluations: the horizontal position of the back and front of the product. An object is leftward-oriented if its front is on the left side of the view, and rightward-oriented if the front is on the right side. From a practical perspective, this novel focus means the present research applies to any product with a clear back and front, rather than to the more restrictive set of products with handles. From a theoretical perspective, it implies that the effects of product orientation require different process explanations. Handle location primes grasping with a corresponding hand, but in the absence of a handle, no corresponding grasping behavior can be cued, nor can the (imagined) difficulty of grasping affect product evaluations. The new process explanation derives from research on stimulus–response compatibility. STIMULUS–RESPONSE COMPATIBILITY Stimulus–response compatibility refers to the impact that irrelevant stimulus features can have on behaviors if the cognitive representation of the stimulus shares features with that of the behavior (Hommel 1997). A classic demonstration of the impact of irrelevant features is offered by the well-known Stroop (1935) task in which people must name the color of the font in which a word, describing a color, is written. In congruent trials, the meaning of the color word is identical to the font color (e.g., the word red is written in red), but in incongruent trials, they differ (e.g., the word red is written in yellow). People tend to be slower and make more mistakes naming the font color in incongruent trials than in congruent ones. Ample studies support a dimensional overlap explanation of such SRC phenomena (e.g., Kornblum et al. 1990; Zhang, Zhang, and Kornblum 1999). This explanation proposes that dimensional overlap occurs when responses and stimuli vary along the same dimension. For example, in the Stroop task, the responses (naming colors) overlap on two dimensions with the stimuli; they vary along the same dimension as the relevant feature (font color) but also as the irrelevant feature (color word). While the overlap of responses with the relevant feature is no cause for concern (and in fact, is absent in many types of SRC phenomena), the overlap with the irrelevant feature causes interference, as it automatically activates responses congruent with it. Independent of the irrelevant stimulus feature (color word) that activates one response action, the relevant stimulus dimension (font color) still initiates a controlled process to identify the correct response action. In congruent trials, where the response based on the irrelevant feature is identical to the response based on the relevant feature, the automatically activated response gets verified and executed quickly (Hommel 1997; Kornblum et al. 1990; Zhang et al. 1999). For example, when people see the word red written in red, the response (saying “red”) is automatically activated, verified, and quickly executed. In incongruent trials, the response activated by the irrelevant stimulus feature is incorrect and needs to be halted, while the correct response needs to be initiated. Hence, when the word red is written in yellow, the incorrect response (saying “red”) becomes activated and needs to be inhibited in order to convey the correct response (saying “yellow”). These additional processes cause a response delay, explaining the slower response times. Because inhibition occasionally fails, the wrong response might be executed (Hommel 1997; Kornblum et al. 1990; Zhang et al. 1999). Building on dimensional overlap theory, we argue that a product’s orientation, which is irrelevant for its evaluation, can nevertheless affect its evaluation if it is conveyed through swiping. Essentially, in expressing product evaluations, (1) directional overlap may serve as a source of response activation, but (2) the process via which the activated response gets enacted is different from classical SRC tasks. We develop these two propositions further below. DIMENSIONAL OVERLAP OF STIMULUS ORIENTATION AND SWIPING RESPONSES Applications that require people to convey evaluations on mobile devices often involve swiping to the left or right or pressing a button on the left or right of a screen. However, the objects or products that app users encounter not only vary in how much one likes them but also often in their orientation. According to dimensional overlap theory, liking is, by definition, the relevant feature for an evaluation task, and product orientation an irrelevant one. Therefore, evaluations may be affected by product orientation if the specific behaviors to convey them vary in features that overlap with product orientation. Hence, we suggest that dimensional overlap may cause product orientations to activate, or prime, corresponding responses, which increases the likelihood of the congruent response and decreases the likelihood of the incongruent one. This line of reasoning implies that the proposed effect occurs at the behavioral (leftward vs. rightward swipe) rather than evaluative (like vs. dislike) level, suggesting a crucial role for space-valence mapping. Most apps map a positive response on the right side and a negative response to the left; swiping to the right virtually always conveys a liking response. Therefore, rightward-oriented products should be better liked than leftward-oriented ones. However, if the space-valence mapping reverses (i.e., the left side represents “like” and the right side represents “dislike”), we still might observe that leftward-oriented stimuli tend to elicit leftward responses; given the reversed space-valence mapping, though, it would imply a tendency to like leftward-oriented stimuli better than rightward-oriented stimuli. In most of our studies, we compare the impact of product orientation on swiping responses and button presses. Depending on the specific spatial feature that product orientation cues, it may affect either both responses or swiping responses only. Horizontal product orientation may either cue a specific lateral or left-right direction or a specific location. If product orientation cues lateral direction, it should affect swiping responses, which imply movement along the lateral direction too, but it should not affect button presses, because they imply movement in a direction that is perpendicular to the cued lateral direction. In fact, while swiping responses vary in the lateral or left-right direction (i.e., occur along the frontal axis), button presses vary in the up-down direction (i.e., occur along the sagittal axis). If, however, product orientation cues a specific left-right location, it should affect button presses, at least if they are executed with fingers of a different hand. While we pit a direction account against a location account in most of our studies, we consider it more likely that product orientation cues direction rather than location and, hence, that only swiping responses will be affected by product orientation. The idea that product orientation affects swiping responses due to dimensional overlap suggests two means to eliminate its effect. First, when a second irrelevant feature also cues lateral direction, it may cancel out the effect of product orientation, at least if the two irrelevant cues prime opposing directions. Second, stimuli and responses need to vary along the same dimension to prompt interference or facilitation (Ansorge and Wühr 2004). In our case, product orientation, and thus implied direction, varies across products. In a typical swiping task, swiping responses also vary along the direction dimension, because people can swipe right or left. If a task requires consumers to make only left-to-right swipe movements, the response direction does not vary across trials, which eliminates any overlap with direction. Therefore, we predict that a horizontal product orientation may activate a directionally congruent response if the task involves choosing to swipe to the left or to the right, but not if it involves swiping only in one direction. SELECTION OF SWIPING RESPONSES TO CONVEY PRODUCT EVALUATIONS VERSUS OBJECTIVE JUDGMENTS Even if the proposed effect of product orientation on evaluation through swiping seems a typical SRC effect, it differs in a key aspect: the absence of inhibition. In SRC tasks, people sometimes make mistakes when answering quickly, and they are very well aware when they do. Anyone who has ever served as a participant in a Stroop task, for instance, undoubtedly has had the experience of realizing immediately after responding that he or she said the color word instead of the font color. This experience reveals clearly to participants that they have been led astray by an irrelevant feature; we argue that the existence of objectively correct versus incorrect answers is crucial to participants’ efforts to inhibit their automatically activated response. But in evaluation tasks, participants are unlikely to be actively occupied with inhibiting a response that is automatically activated by product orientation. In fact, it is unlikely that participants would view their evaluations as incorrect or invalid, as many studies indicate how little insight people have into the construction of their own evaluations (Cacioppo, Priester, and Berntson 1993; Schwarz 2004, 2006; Slovic 1995; Strack, Martin, and Stepper 1988). The presence versus absence of inhibitory processes affects the temporal dynamic of the observed SRC effects. In typical SRC tasks, slower responses are much less likely to be affected by dimensional overlap, because more time results in a higher likelihood of successful inhibition. If consumers do not try to inhibit automatically activated responses, however, the effect on their responses is the same, regardless of how quickly or slowly they issue those responses. This distinction is particularly pertinent for our study setting, because real-life swiping responses take more time than is typically allowed in SRC tasks, which tend to drop responses slower than 1,500 milliseconds (De Houwer 2003; Hommel 1993; MacDonald et al. 2000). So, while no SRC effects would be observed in typical SRC tasks at these long response windows, we predict that response window is unlikely to moderate the impact of product orientation on swiping. STUDY OVERVIEW To test our propositions, we conduct five studies. Study 1 demonstrates the hypothesized effect: on a touchscreen device, rightward product orientations (i.e., toys with their front oriented to the right vs. left) tend to elicit more liking responses than leftward-oriented ones when the positive evaluations are mapped on the right side and the evaluation is conveyed through swiping. No orientation effect is observed for button pressing, though. Study 2 demonstrates the critical role of space-valence mapping: we replicate the effect of product orientation on swiping-based evaluations for the usual space-valence mapping (left = bad, right = good), but find it reverses when the space-valence mapping is reversed. Studies 3 and 4 demonstrate that the effect of product orientation on swiping is attenuated when dimensional overlap is reduced or absent. Study 3 introduces a second directional cue (i.e., implied motion) as a moderator. We replicate the effect of product orientation on evaluation when implied motion is congruent with product orientation, but find it disappears when implied motion is incongruent with product orientation. Study 4 tests the explanatory role of dimensional overlap by showing that the effect is eliminated when only one directional response is defined (i.e., left-to-right swipe movement); establishing dimensional overlap requires the responses to vary on the overlapping dimension. Finally, the influence of horizontal orientation on swiping critically depends on the presence versus absence of inhibitory processes. Study 5 shows that in conveying objective judgments, such as discriminating between toys that depict animals or not, consumers tend to override the influence of product orientation in the absence of time pressure. However, when they convey evaluations, no inhibition of the primed response occurs, irrespective of time pressure. STUDY 1: PRODUCT ORIENTATION INFLUENCES SWIPING DIRECTION The aim of this first study is to assess whether product orientation affects which swipe movement consumers prefer. We predict that when a product is oriented rightward, the tendency to swipe to the right is greater than when the product is oriented leftward. If evaluations instead require pressing a button, people do not engage in lateral directional movement, so product orientation should not affect evaluations (to the same extent) in this condition. Method In exchange for monetary compensation (€8), 58 students of Ghent University (34 women, MAge = 21.59, SD = 3.64) were randomly assigned to one of two cells in a between-subjects design (response mode: swipe vs. button). All participants were informed that the research was about gift giving. Pictures of toys appeared on the screen, and participants were instructed to indicate whether they would consider the presented toy as an attractive gift for one- to four-year-old children. A set of 30 pictures of toys (see appendix B for a sample), each with a clear leftward or rightward orientation, and their mirror-reversed version were included in this study. The pictures appeared on the screen, one by one and in a fixed order. Whether a toy was presented with its leftward or rightward orientation was determined randomly. Hence, all participants saw the exact same stimuli in the exact same order, but the orientation varied. When a picture appeared in the center of the screen, participants in the button condition could like it as a gift by pressing a green checkmark, which appeared on the right side of the screen. In the swipe condition, they instead would signal a positive evaluation by swiping the picture toward the checkmark. If participants wanted to evaluate the presented toy negatively, they pressed a red “x,” fixed on the left side of the screen, or swiped the picture to the left. Results and Discussion The 58 participants provided 1,740 responses in total. The time to make a decision varied from 6 milliseconds to 30 seconds. In all studies, we analyzed the data for all responses; we also analyzed the data after dropping implausibly fast responses (RT < 200 milliseconds) to eliminate responses that participants might have initiated before they even saw the stimulus. This cutoff is based on research showing that minimal reaction times for simple categorization, arguably easier than our tasks, are slightly above that threshold (Joubert et al. 2007). Because this exclusion involved very few observations (ranging from.15% in study 1 to.32% in study 3), it did not alter the pattern of significance in any of our studies and we therefore always present the results for all responses. Each participant offered multiple responses, so the assumption of independence of observations does not hold, and we used a multilevel model to account for the dependency in the observations from each participant. In a multilevel logistic regression (which includes a random intercept to account for variance in liking across participants and thus for dependency in the observations from the same participant), we regressed the likelihood of evaluating a toy positively on the response mode condition, the toy’s orientation, and their two-way interaction. We obtained a main effect of orientation, Wald χ²(1) = 4.85, p = .028, indicating more liking when the toy was rightward oriented (mean probability of liking, abbreviated hereafter as P = .75) rather than leftward oriented (P = .70). The only other significant effect was the expected two-way interaction effect between the response mode condition and product orientation, Wald χ²(1) = 14.44, p < .001 (figure 1). As predicted, we found an effect of orientation in the swipe condition, Wald χ²(1) = 13.45, p < .001 (Pleftward = .64, Prightward = .80), but not in the button press condition, Wald χ²(1) = 1.93, p =.16 (Pleftward = .75, Prightward = .70). FIGURE 1 View largeDownload slide MEAN ESTIMATED LIKING AS A RESULT OF PRODUCT ORIENTATION AND RESPONSE MODE FIGURE 1 View largeDownload slide MEAN ESTIMATED LIKING AS A RESULT OF PRODUCT ORIENTATION AND RESPONSE MODE These results illustrate that object orientation guides the selection of directional action and correspondingly affects product evaluations. Because this study includes a condition in which respondents expressed their evaluations through button pressing rather than swiping, we infer that product orientation affects lateral directional actions only, leaving actions with no lateral directional dimension unaffected. This favors a direction (over a location) account for the effect of product orientation. STUDY 2: THE ROLE OF SPACE-VALENCE MAPPING Most apps that require swiping movements for evaluations associate the left side with a negative evaluation and the right side with a positive evaluation; this study therefore investigates whether a similar effect occurs when positive evaluations entail leftward swiping movements. If the previously demonstrated effect originates because directionally compatible movements are facilitated by product orientations, leftward-oriented pictures should receive more positive evaluations if those positive evaluations are expressed through leftward swiping movements, rather than through pressing a button on the left side of the screen. That is, we test whether the effects obtained for swiping in study 1 reverse when the space-valence mapping reverses. Method In this laboratory study, 126 students of Ghent University (men N = 56, MAge = 23.23 years, SD = 4.97; women N = 70, MAge = 23.11 years, SD = 5.99) participated in return for monetary compensation. They sat in front of touchscreen computers and were randomly assigned to one of four between-subjects conditions. The first between-subjects manipulated factor was response mode, to express (dis)liking (swiping vs. button pressing). The second factor related to the space-valence mapping, such that half of the respondents saw a green checkmark on the right side of the screen and a red “x” on the left side (i.e., right = positive), and the other half saw a screen with the green checkmark on the left side and the red “x” on the right (i.e., left = positive). As in study 1, product orientation was manipulated within subjects: participants observed pictures of shoes with the toe oriented to the left or right. Men saw men’s shoes, and women saw women’s shoes (see appendix B). The instructions explained that participants were seated in front of a touchscreen and should complete the questionnaire by using it as an input device, rather than a mouse. They further read that 30 pictures of shoes would appear on the screen, and they should indicate whether they liked or disliked each pair. The shoes, presented in fixed order, appeared in the center of the screen, and each shoe’s rightward or leftward orientation was determined randomly. As each picture appeared, participants in the button condition could like it by pressing the green checkmark, presented on the right or left side of the screen; participants in the swipe condition were instructed to give a positive evaluation by swiping the picture to the checkmark. If they wanted to evaluate the presented shoe negatively, they pressed the red “x,” presented on the left or right side of the screen in the button condition, or swiped to the red “x” in the swipe condition. Results and Discussion Because each of the 126 participants provided 30 responses, we obtained 3,780 responses. The time to make a decision varied from 4 milliseconds to 23 seconds. We conducted a multilevel logistic regression with the binary evaluation (like/dislike) of the presented shoe as the dependent variable and response mode, shoe orientation, valence mapping, and gender, together with all their interactions, as the independent variables. We observed a main effect of gender, Wald χ2(1) = 5.48, p = .02, such that women liked the women’s shoes more (P = .44) than men liked the men’s shoes (P = .37). Yet gender did not interact with any of the remaining factors; all Wald χ2(1) < .73, and all ps > .39. We observed a main effect of valence mapping, Wald χ²(1) = 26.28, p < .001. Shoes were evaluated more favorably when the positive responses were on the right (P = .48) instead of the left (P = .33) side of the screen. This finding is in line with the observation that most people associate the right side with “good” (Casasanto and Chrysikou 2011). The valence mapping effect was qualified by an interaction with shoe orientation, Wald χ²(1) = 24.42, p < .001. When the right side represented a positive evaluation, a shoe was more liked if it was rightward (P = .53) rather than leftward (P = .43) oriented. If the right side represented a negative evaluation, a shoe was less liked if it was rightward (P = .28) rather than leftward (P = .37) oriented. In line with our theorizing, this interaction of valence mapping with shoe orientation was further qualified by response mode, Wald χ²(1) = 19.69, p < .001 (figure 2). As before, we observed a directional congruency effect for swiping, Wald χ²(1) = 39.96, p < .001, but not for button presses, Wald χ²(1) = .14, p = .70. For positive responses on the right, shoes were more preferred if they were rightward oriented (P = .58) instead of leftward oriented (P = .39, p < .001) in the swiping condition, but they were equally preferred in the button press condition (Pleftward = .46, p = .57, Prightward = .48). Conversely, if positive responses were on the left, leftward-oriented shoes were preferred more (P = .41) than rightward-oriented ones (P = .24, p < .001) in the swiping condition but equally preferred in the button press condition (Pleftward = .33, Prightward = .33, p = .91). FIGURE 2 View largeDownload slide MEAN ESTIMATED LIKING AS A RESULT OF SHOE ORIENTATION, RESPONSE MODE, AND SPACE-VALENCE MAPPING FIGURE 2 View largeDownload slide MEAN ESTIMATED LIKING AS A RESULT OF SHOE ORIENTATION, RESPONSE MODE, AND SPACE-VALENCE MAPPING These results support the idea that object orientation guides swiping responses and, correspondingly, affects reported product evaluations. Moreover, by manipulating whether positive evaluations appear on the right or left side of the screen, we illustrate that the effect is not an artifact, limited to one side of the screen. This clear evidence reveals that people are inclined to engage in congruent rather than incongruent directional actions. STUDY 3: INFLUENCE OF MULTIPLE ORIENTATION CUES ON SWIPING DIRECTION The results of studies 1 and 2 indicate that a horizontal orientation cues a response in the congruent lateral direction and thus affects swipes, but not button presses. If these effects occur because the horizontal orientation primes a congruent direction, the effect should be eliminated if a stimulus contains a directional cue in the opposite direction, because the dimensional overlap would not be unequivocally established in this case. To test this hypothesis, we used three sets of pictogram stimuli (see appendix B). The first set contained static images (e.g., someone sitting behind a desk). From a theoretical perspective, they are similar to the stimuli of studies 1 and 2. A second set of stimuli contained nonstatic images of someone moving in the same direction as implied by the orientation (e.g., a cyclist). These pictograms contain two congruent directional cues, so we anticipate replicating the effect of orientation on swiping but not button presses. A priori, it is unclear whether the combination of two congruent directional cues will strengthen the observed effects. Finally, a third set of stimuli contained nonstatic images of someone moving in the opposite direction implied by the orientation (e.g., a man falling on his back). Because these pictograms contain opposing directional cues, we expect the orientation effect for swiping to be reduced or even eliminated. In summary, for static and congruent nonstatic images, we expect an orientation effect for swiping responses but not for button presses. For incongruent nonstatic images, we expect to find no orientation effect for either swiping or button presses. Method In this study, 185 students of Ghent University (65 men, MAge = 21.63, SD = 4.18) evaluated the aesthetic appeal of 12 pictograms for monetary compensation. Each participant was seated in front of a touchscreen and evaluated the pictograms by either swiping or pressing an onscreen button. In both response mode conditions, a green checkmark (i.e., like button) was on the right side of the screen, and a red “x” (i.e., dislike button) appeared on the left. All participants evaluated all 12 pictograms, presented in a fixed order, but for each trial, whether the picture appeared in its original or mirror-reversed orientation was randomly determined; that is, the exact same pictograms were shown to all participants, but in varying orientation. Four pictograms presented human figures with a clear leftward or rightward orientation, easily inferred from static orientation cues such as head and body orientation (e.g., a person seated at a desk). Four other pictograms contained multiple congruent orientation cues, such that orientation and implied movement were in the same direction (e.g., a person looking in the same direction as walking or riding a bike). A final set of four pictograms revealed multiple, incongruent orientation cues, where orientation and implied movement were in the opposite direction (e.g., a person looking in one direction but pulling an object in the opposite direction). Results and Discussion In the multilevel logistic regression model (with a random intercept to account for variation among participants’ liking), we regressed the likelihood of evaluating a pictogram positively on response mode, pictogram orientation, and pictogram type, together with all their interactions. We obtained a main effect of pictogram type, Wald χ²(2) = 103.71, p < .001, indicating that static images tended to be somewhat liked (P = .56), as did incongruent images (P = .56), but congruent images tended to be disliked (P = .38). This main effect was qualified by a pictogram orientation × pictogram type interaction, Wald χ²(2) = 7.65, p = .022. Orientation affected liking for static images, Wald χ²(1) = 4.67, p = .031 (Pleftward = .52, Prightward = .60), and for congruent images, Wald χ²(1) = 4.45, p = .036 (Pleftward = .34, Prightward = .42) but not for incongruent images, Wald χ²(1) = .97, p = .33 (Pleftward = .58, Prightward = .54). Figure 3 depicts the responses, broken down by response mode, pictogram orientation, and pictogram type. The three-way interaction among the three independent variables did not reach a conventional level of significance, Wald χ²(2) = 5.78, p = .056, but the results are in line with our predictions. In support of our argument, the effect of orientation differs across the three types of images in the swiping condition, Wald χ²(2) = 11.34, p = .004, but not in the button press condition, Wald χ²(2) = .18, p = .91. In the button press condition, we also observed no overall effect of image orientation, Wald χ²(1) = .48, p = .49. In the swiping condition, we uncovered an effect of image orientation for static images, Wald χ²(1) = 11.13, p < .001 (Pleftward = .49, Prightward = .65), and for congruent images, Wald χ²(1) = 10.39, p < .001 (Pleftward = .29, Prightward = .46), but not for incongruent ones, Wald χ²(1) = .47, p = .49 (Pleftward = .58, Prightward = .55). FIGURE 3 View largeDownload slide MEAN ESTIMATED LIKELIHOOD OF LIKING A PRODUCT AS A RESULT OF PICTOGRAM ORIENTATION, RESPONSE MODE, AND PICTOGRAM TYPE FIGURE 3 View largeDownload slide MEAN ESTIMATED LIKELIHOOD OF LIKING A PRODUCT AS A RESULT OF PICTOGRAM ORIENTATION, RESPONSE MODE, AND PICTOGRAM TYPE At first glance, the lower likelihood of a liking response for congruent versus incongruent images may seem inconsistent with our hypotheses. However, each pictogram type was operationalized by a different set of pictograms, so comparisons of the responses to the different pictogram types clearly are confounded by the main effect of pictogram type. The congruent pictograms simply were not liked much (as also conveyed by the means in the corresponding button conditions). A more precise test of our hypotheses instead requires comparisons within each pictogram type and we have organized figure 3 accordingly. All these comparisons support our hypotheses. The results of study 3 thus show that product orientation affects swiping responses when there is a single orientation cue or when multiple directional cues are congruent, but not when a stimulus contains multiple, incongruent orientation cues. STUDY 4: ONE-DIRECTIONAL VERSUS BIDIRECTIONAL RESPONSES Study 4 has two main goals: to show that not only product evaluations but also other, more downstream variables can be affected by product orientation and to attest to the pivotal role of dimensional overlap between an irrelevant stimulus feature (orientation) and valid responses (swiping) for prompting the observed orientation effect. Study 4 therefore focuses on consumers’ expressions of willingness to pay (WTP) for products. Furthermore, SRC research establishes that dimensional overlap can be established successfully only if the response alternatives vary on the overlapping dimension. For study 4, participants convey their willingness to pay on a slider scale with a custom start position, such that the slider tool is positioned either in the center of the scale, enabling participants to drag it rightward or leftward, or on the extreme left, allowing only rightward movements of the slider. While the effect of product orientation on swiping direction and WTP should persist when the slider tool is centered on the scale, it should be eliminated when it is positioned on the extreme left. Method In exchange for monetary compensation, 184 students of Ghent University (76 men, MAge = 21.59, SD = 1.29) participated in a laboratory study in which they had to complete several unrelated tasks. For the present study, participants viewed eight pictures of gadgets, each with a front side that was unambiguously located to the left or right (see appendix B). We created two versions, leftward and rightward oriented, for each gadget, then determined at random whether the participant would see the original or mirror-reversed version. The participants indicated their willingness to pay for each gadget on a slider scale, ranging from 0 to 20 euro, by dragging the slider tool on a touchscreen to the desired position. A between-subjects manipulation altered whether the slider tool initially appeared in the middle (10 euro) or left side (0 euro) of the scale. The former condition requires consumers to swipe to the left (right) to express a lower (higher) willingness to pay, but the latter allows rightward responses only. Results and Discussion We ran a multilevel regression model regressing slider tool location (left vs. center), product orientation, and their interaction term, onto willingness to pay. We estimated the error degrees of freedom of the statistical tests using Satterthwaite’s approximation, which may produce fractional degrees of freedom, to adjust for heterogeneity in error variances across conditions (Satterthwaite 1946). The results indicated an interaction between product orientation and the initial position of the slider, F(1, 1459.63) = 4.49, p = .034. When the slider initially appeared in the middle of the scale, participants were willing to pay more for rightward-oriented products than leftward-oriented ones, F(1, 1466.57) = 9.34, p = .002 (Mleftward = 5.14, Mrightward = 6.29). When it initially appeared to the left, willingness to pay was not significantly influenced by the leftward or rightward orientation of the products, F(1, 1443.65) = .001, p = .97 (Mleftward = 5.40, Mrightward = 5.41). We also recoded willingness to pay as a binary measure of leftward swipes (i.e., WTP ≤ 10) versus rightward swipes (i.e., WTP > 10) for the condition with the centered slider. In a multilevel binary regression model, the effects of orientation and initial slider position on this dichotomous variable revealed similar patterns. Product orientation did not influence participants’ likelihood of indicating a WTP greater than 10 when the slider began on the left side of the scale, Wald χ²(1) = .21, p = .64 (Pleftward = .18, Prightward = .16), but this likelihood increased for rightward-oriented (Prightward = .22) versus leftward-oriented (Pleftward = .17) products when the slider was centered initially, χ²(1) = 3.52, p = .061. These findings confirm our expectation that response alternatives must vary on the overlapping dimension, because only when responses in both directions are possible is stimulus orientation able to cue a directionally congruent response. These results indicate that dimensional overlap underlies the effect of orientation on evaluations, according to a moderation-of-process approach (Spencer, Zanna, and Fong 2005). If variation in responses is required for SRC effects to arise, one might wonder if variation is also required on the stimulus side (see web appendix A for an empirical test). Study 4 also shows that the effect is not tied to any specific outcome measure: both product evaluations and more downstream dependent variables, such as willingness to pay, can be affected by product orientations. This finding makes sense in that we posit that the effect occurs at the behavioral, not the evaluative, level—that is, the orientation effect reflects an increased likelihood of swiping in a corresponding direction, rather than a genuine effect on evaluations. Nevertheless, orientation might not affect evaluations for all stimuli. Web appendix B does show that the effect of product orientation is very robust and is eliminated only for extremely (un)attractive stimuli. Moreover, as hypothesized, the effect of orientation may also differ for subjective versus objective judgments; this is tested in study 5. STUDY 5: INHIBITION OF ORIENTATION-CONGRUENT RESPONSES IN OBJECTIVE BUT NOT SUBJECTIVE JUDGMENTS The preceding studies show that product orientation activates a congruent response, but we suggest that the process that determines whether consumers execute the activated response action to convey their evaluations differs essentially from the process that emerges in typical SRC studies. The essential difference between the current paradigm and typical SRC studies is that we focus on subjective judgments, while typical SRC studies involve objective ones. A typical SRC paradigm establishes an unambiguous definition of the (in)correct answer that allows participants to recognize incorrect responses, which they should inhibit. Participants would select an erroneous response only if they fail to inhibit it. This is unlikely for slower responses, because they have taken more time to halt their incorrect action. So, for objective judgments (e.g., classifying a toy as an animal or a car), we should observe interference due to product orientation only under time pressure but not when no time pressure is applied. Subjective evaluations (e.g., “Do you like the presented toy as a children’s gift?”), however, are an entirely different matter. The correct responses are not clear, so consumers may not consider inhibiting any activated response, irrespective of the time frame. With study 5, we attempt to show that the influence of orientation on objective judgments that are conveyed via swiping movements is moderated by time pressure, but time pressure has no such effect for subjective evaluations. Method The study was completed by 96 students for partial course fulfillment, in the consumer laboratory of Ghent University (42 men; MAge = 20.80 years, SD = 1.41). The participants were seated in isolated cubicles in front of desktop computers connected to touchscreens. The task consisted of two parts. First, participants offered subjective evaluations, similar to the swiping condition from study 1, of the attractiveness of 30 pictures of toys. The toys appeared with a randomly determined leftward or rightward orientation. Second, the participants had to specify whether each toy represented an animal or not; 17 of the 30 stimuli (57%) were animals. The judgment type (subjective vs. objective) and product orientation (left vs. right) thus were manipulated within participants. To test whether the effect of orientation on swiping responses was moderated by time pressure in objective, but not subjective, evaluations, we included time pressure as a between-subjects factor, such that half of the participants had to offer their responses within 1,000 milliseconds (a common cutoff in analyses of response latency tasks), but the other half did not have any restriction on their response times. Results and Discussion This study yielded 5,760 data lines for analysis in that the 96 participants each completed 60 trials. However, in the time pressure condition, participants failed to respond in time to 405 trials, so we obtained 5,355 usable responses. We ran a multilevel logistic regression analysis, with the binary direction of the swipe movement (rightward swipes convey likes or identifications of animals) as the dependent variable and judgment type, toy orientation, and time pressure, together with their interactions, as independent variables. The three-way interaction was significant, Wald χ²(1) = 4.04, p = .044: toys were subjectively judged to be more attractive when oriented to the right (P = .61) rather than left (P = .56), Wald χ²(1) = 7.74, p = .005, irrespective of the required response speed, Wald χ²(1) = .303, p = .58. The effect of toy orientation on objective judgments varied with the time pressure, though, Wald χ²(1) = 3.37, p = .066. Specifically, toy orientation did not affect objective judgments in the absence of time pressure, Wald χ²(1) = .44, p = .51 (Pleftward = .59, Prightward = .57), but the correct identification was distorted by the product orientation when participants faced time pressure, Wald χ²(1) = 3.89, p = .049 (Pleftward = .55, Prightward = .60). As summarized in figure 4, these results affirmed our expectations; response speed did not alter the impact of stimulus orientation for subjective judgments, but it had a critical influence on objective judgments. This difference likely reflects participants’ failure to realize the necessity of inhibiting the response activated by stimulus orientation when conveying their liking, even though their responses were affected by the irrelevant feature of product orientation. A study in web appendix C shows that making participants aware of this potential influence eliminates it. FIGURE 4 View largeDownload slide MEAN ESTIMATED LIKELIHOOD TO SWIPE RIGHT AS A RESULT OF PRODUCT ORIENTATION, TIME PRESSURE, AND JUDGMENT TYPE FIGURE 4 View largeDownload slide MEAN ESTIMATED LIKELIHOOD TO SWIPE RIGHT AS A RESULT OF PRODUCT ORIENTATION, TIME PRESSURE, AND JUDGMENT TYPE GENERAL DISCUSSION Consumers increasingly use mobile devices to evaluate and purchase products and services, which makes it important to understand how their interactions with these devices might shape their evaluations and preferences. This article focuses on two common responses: swiping and button presses. On the basis of dimensional overlap theory (Kornblum et al. 1990; Zhang et al. 1999), we predict and find that horizontal product orientation affects swiping responses but not button presses. Specifically, consumers tend to swipe in the direction of the product orientation (studies 1 and 2). The effect of product orientation occurs at the behavioral rather than evaluative level. Consequently, when the space-valence mapping is reversed (study 2), rightward-oriented stimuli still elicit rightward swiping responses, which then signify dislike. The effect unfolds because (1) stimulus–response overlap serves as a source of response action priming (studies 3 and 4), (2) rendering consumers more likely to execute the primed response because they view their response as legitimate and do not experience the need to inhibit their orientation-congruent response (study 5). Finally, we observe orientation effects for both liking and willingness-to-pay judgments, attesting to their fundamental nature (study 4). This study contributes to extant research into how irrelevant visual product presentation cues can influence product judgments. Prior research has addressed the effect of horizontal positioning, in advertising (Chae and Hoegg 2013; Janiszewski 1990; Peracchio and Meyers-Levy 1997), retail displays (Chandon et al. 2009; Christenfeld 1995; Valenzuela and Raghubir 2009; Valenzuela, Raghubir, and Mitakakis 2013), and product packaging (Deng and Kahn 2009), to establish that this irrelevant product presentation cue provides information that consumers use in their judgments. We extend this research stream by showing that another irrelevant cue for judgment, horizontal product orientation, also can affect product evaluations if they are conveyed through swiping movements. Moreover, the mechanism that underlies the orientation effect markedly differs from the mechanisms that evoke product location effects. Moreover, the current research enriches growing literature on the role of preference assessment procedures in consumer behavior (Krüger, Mata, and Ihmels 2014; Lembregts and Pandelaere 2013; Schley, Lembregts, and Peters 2017; Weijters, Geuens, and Baumgartner 2013) and the particular impact of preference expression modalities (Klesse, Levav, and Goukens 2015; Shen, Zhang, and Krishna 2016). For example, speaking prompts more indulgent choices than manual preference expression modalities (Klesse et al. 2015), and expressing preferences via touch-based interfaces, rather than mouse-driven desktop computers, facilitates affective choices (Shen et al. 2016). Our finding extends this research by comparing button pressing to swiping, a new behavior that consumers adopt to interact with products presented on touch-based interfaces. These findings also relate to prior work on stimulus–response compatibility. The facilitation of certain motor responses by task-irrelevant visual aspects of a stimulus is well documented, such as in the Stroop effect, which describes how people are sometimes affected by the meaning of a word when asked to name the color in which it is depicted (Stroop 1935). Extending this vast research, our study differs from prior investigations in two main respects. Prior research has ignored the impact of direction cues but, as we show, product orientation also cues a directional response, which may open new lines of investigation for SRC phenomena as they relate to horizontal orientations. Some research notes how arrows (Miles and Proctor 2012; Wascher et al. 1999) and arrow primes (Danziger, Kingstone, and Ward 2001) affect response latencies, but these arrows mainly cue location, because they orient visual attention involuntarily (Tipples 2002). Further research could investigate how and why arrows differ from horizontal orientation. In addition, in showing that a response that is primed by a stimulus aspect may inform evaluative responses, rather than focusing on reduced response latencies or erroneous responses in incongruent trials, our study suggests some broader behavioral implications of stimulus aspects. They do not just affect fast, (in)correct responses; they also influence responses that convey some subjective judgment and require more time. Our research also yields insights into the mechanisms that underlie SRC effects and informs the debate about why SRC effects disappear for longer response times. Some explanations suggest that longer response times allow for the more successful execution of inhibitory processes, but another side argues that the activation of a response due to dimensional overlap decays rapidly (Hommel 1997), such that dimensional overlap no longer affects slower responses. Our findings offer support for the former position. We find that for a subjective task, the orientation effect—an SRC effect in its own right—persists even for very long response times. For subjective tasks, people sense no need to instigate inhibitory processes. The presence of an objective criterion in typical SRC tasks presumably leads participants to develop some phenomenological awareness of the bias imposed by the irrelevant feature, thereby encouraging and allowing them to inhibit their illegitimate responses. Finally, we extend extant literature on product orientation in two key ways. First, prior studies define product orientation by handle locations (Eelen et al. 2013; Elder and Krishna 2012; Ping et al. 2009), but we consider the location of the back and front. This approach not only taps into a different aspect of horizontal orientation but also applies to a broader set of products than has been covered by prior research. Second, our focus on a different orientation aspect implies a different process than has been described in prior research. The mechanism underlying the handle location effect involves the metacognitive ease of (simulated) grasping behavior. Grasping behaviors are actions that become associated with products through a learning process. Upon a consumer’s exposure to a product with a handle, a behavioral response (i.e., grasping) gets cognitively activated by the (over)learned association between the stimulus and the behavioral response, and its simulated motor fluency affects evaluative judgments (Alter and Oppenheimer 2009; Reber, Winkielman, and Schwarz 1998; Schwarz 2004). The mechanism underlying horizontal orientation, defined by back and front location, instead involves response priming due to dimensional overlap. To be clear, in the absence of a learned association between the stimulus and the response, responses still might be primed by perceptions of stimuli that share features with the cognitive representation of the behavioral response (Hommel 1997), such as when a product orientation cues a swiping response. A primed response may be easier to execute, rendering its execution more likely. However, in our studies, manipulating the horizontal orientation did not appear to influence how participants felt about a product directly; it merely affected their selected response to convey their binary like/dislike evaluation. This key difference compared with a motor fluency process is most prominent in study 2, in which we manipulated the space-valence mapping. Whether a rightward-oriented shoe was liked better than a leftward-oriented shoe depended critically on which side was associated with the “like” response, a result that is inconsistent with a motor fluency account. With regard to managerial implications, we note that many apps that ask consumers to evaluate objects offer not some random selection of objects but rather a set that users might like, learned from those users’ past responses (i.e., objects more likely to be liked). Any liking response due to object orientation thus may affect which objects appear subsequently. People often look to their own behavior to infer their attitudes and opinions too (Bem 1972), such that they might infer from their own swiping response whether they like an object or not. Such inferences may function to update their prior evaluations. Finally, online evaluation tools often keep track of consumers’ product evaluations, which enables them to inform other consumers about product popularity. Consumers in turn rely on ratings contributed by others to make purchase decisions (Chevalier and Mayzlin 2006; Zhu and Zhang 2010). The perception of success alone can be sufficient to affect product decisions, so even if popularity ratings are not representative of actual product quality, they still can guide decision making (Muchnik, Aral, and Taylor 2013; Salganik and Watts 2008). This trend highlights the importance of recognizing how product orientation can inflate the number of positive evaluations of a product, conveyed through swiping movements. The long-term consequences of a particular swiping response, induced by product orientation, on the preferences of a focal consumer or the preferences of others thus remain as questions for further research. DATA COLLECTION INFORMATION All data were collected in the consumer behavior lab at Ghent University. All studies were scheduled as part of 50 minute or 30 minute sessions in which data for other, unrelated experiments were also collected. The first author and the researchers involved in the unrelated experiments alternated in the supervision of the lab sessions. Data for study 1 were collected in October 2014. Data for study 2 were collected in September 2014. Data for study 3 were collected in October 2014. Data for study 4 were collected in March 2017. For study 5, data were collected in March 2017. Both authors were involved in the design of the studies and the analyses. The final analyses for studies 4, 5, and web appendix C that are reported in the article were conducted by the first author. The final analyses for all other studies were conducted by the second author. The authors wish to thank the editor, Gita Johar; the associate editor, Stijn van Osselaer; and three anonymous reviewers for their feedback on the previous versions of this article. Supplementary materials are included in the web appendix accompanying the online version of this article. The data package is available via doi 10.5061/dryad.rh0v0g2. This research was supported by funding through the Flemish Fund for Scientific Research to the second author (grant G020215N). APPENDIX A FRONT AND BACK VERSUS HANDLE LOCATIONS AS HORIZONTAL ORIENTATION FEATURES Example objects with the front located to the left, suggesting a leftward orientation Example objects with the front located on the left, suggesting a leftward orientation, and a handle located on the right (in other research, defined as a rightward orientation) Example objects with a handle located on the right (in other research, defined as a rightward orientation) APPENDIX B Sample stimuli Studies 1 and 5 (toys) Study 2 (women’s and men’s shoes) Men’s shoes Women’s shoes Men’s shoes Women’s shoes Men’s shoes Women’s shoes Men’s shoes Women’s shoes Study 3 (pictograms) Single orientation cue Multiple congruent orientation cues Multiple incongruent orientation cues Single orientation cue Multiple congruent orientation cues Multiple incongruent orientation cues Single orientation cue Multiple congruent orientation cues Multiple incongruent orientation cues Single orientation cue Multiple congruent orientation cues Multiple incongruent orientation cues Study 4 (gadgets) References Alter Adam L. , Oppenheimer Daniel M. ( 2009 ), “Uniting the Tribes of Fluency to Form a Metacognitive Nation,” Personality and Social Psychology Review , 13 3 , 219 – 35 . Google Scholar CrossRef Search ADS PubMed Ansorge Ulrich , Wühr Peter ( 2004 ), “A Response-Discrimination Account of the Simon Effect,” Journal of Experimental Psychology: Human Perception and Performance , 30 2 , 365 – 77 . Google Scholar CrossRef Search ADS PubMed Areni Charles S. , Duhan Dale F. , Kiecker Pamela ( 1999 ), “Point-of-Purchase Displays, Product Organization, and Brand Purchase Likelihoods,” Journal of the Academy of Marketing Science , 27 4 , 428 – 41 . Google Scholar CrossRef Search ADS Atalay A. Selin , Bodur H. Onur , Rasolofoarison Dina ( 2012 ), “Shining in the Center: Central Gaze Cascade Effect on Product Choice,” Journal of Consumer Research , 39 4 , 848 – 66 . Google Scholar CrossRef Search ADS Bem Daryl J. ( 1972 ), “Self-Perception Theory,” in Advances in Experimental Social Psychology , Vol. 6 , ed. Berkowitz Leonard , New York : Academic Press , 1 – 62 . Brasel S. Adam , Gips James ( 2014 ), “Tablets, Touchscreens, and Touchpads: How Varying Touch Interfaces Trigger Psychological Ownership and Endowment,” Journal of Consumer Psychology , 24 2 , 226 – 33 . Google Scholar CrossRef Search ADS Buchanan Lauranne , Simmons Carolyn J. , Bickart Barbara A. ( 1999 ), “Brand Equity Dilution: Retailer Display and Context Brand Effects,” Journal of Marketing Research , 36 3 , 345 – 55 . Google Scholar CrossRef Search ADS Cacioppo John T. , Priester Joseph R. , Berntson Gary G. ( 1993 ), “Rudimentary Determinants of Attitudes: II. Arm Flexion and Extension Have Differential Effects on Attitudes,” Journal of Personality and Social Psychology , 65 1 , 5 . Google Scholar CrossRef Search ADS PubMed Casasanto Daniel , Chrysikou Evangelia G. ( 2011 ), “When Left Is ‘Right’: Motor Fluency Shapes Abstract Concepts,” Psychological Science , 22 4 , 419 – 22 . Google Scholar CrossRef Search ADS PubMed Castro Iana A. , Morales Andrea C. , Nowlis Stephen M. ( 2013 ), “The Influence of Disorganized Shelf Displays and Limited Product Quantity on Consumer Purchase,” Journal of Marketing , 77 4 , 118 – 33 . Google Scholar CrossRef Search ADS Cavazza Nicoletta , Gabrielli Veronica ( 2015 ), “Affordant Shapes of Product Holder Influence Product Evaluation and Purchase Intention,” Current Psychology , 34 2 , 447 – 65 . Google Scholar CrossRef Search ADS Chae Boyoun (Grace) , Hoegg JoAndrea ( 2013 ), “The Future Looks ‘Right’: Effects of the Horizontal Location of Advertising Images on Product Attitude,” Journal of Consumer Research , 40 2 , 223 – 38 . Google Scholar CrossRef Search ADS Chandon Pierre , Hutchinson Wesley J. , Bradlow Eric T. , Young Scott H. ( 2009 ), “Does In-Store Marketing Work? Effects of the Number and Position of Shelf Facings on Brand Attention and Evaluation at the Point of Purchase,” Journal of Marketing , 73 6 , 1 – 17 . Google Scholar CrossRef Search ADS Chevalier Judith A. , Mayzlin Dina ( 2006 ), “The Effect of Word of Mouth on Sales: Online Book Reviews,” Journal of Marketing Research , 43 3 , 345 – 54 . Google Scholar CrossRef Search ADS Christenfeld Nicholas ( 1995 ), “Choices from Identical Options,” Psychological Science , 6 1 , 50 – 5 . Google Scholar CrossRef Search ADS Danziger Shai , Kingstone Alan , Ward Robert ( 2001 ), “Environmentally Defined Frames of Reference: Their Time Course and Sensitivity to Spatial Cues and Attention,” Journal of Experimental Psychology: Human Perception and Performance , 27 2 , 494 – 503 . Google Scholar CrossRef Search ADS PubMed De Houwer Jan ( 1998 ), “The Semantic Simon Effect,” Quarterly Journal of Experimental Psychology: Section A , 51 3 , 683 – 8 . Google Scholar CrossRef Search ADS De Houwer Jan ( 2003 ), “On the Role of Stimulus-Response and Stimulus-Stimulus Compatibility in the Stroop Effect,” Memory & Cognition , 31 3 , 353 – 9 . Google Scholar CrossRef Search ADS PubMed De Houwer Jan , Eelen Paul ( 1998 ), “An Affective Variant of the Simon Paradigm,” Cognition & Emotion , 12 1 , 45 – 62 . Google Scholar CrossRef Search ADS Deng Xiaoyan , Kahn Barbara E. ( 2009 ), “Is Your Product on the Right Side? The ‘Location Effect’ on Perceived Product Heaviness and Package Evaluation,” Journal of Marketing Research , 46 6 , 725 – 38 . Google Scholar CrossRef Search ADS Eder Andreas B. , Rothermund Klaus ( 2008 ), “When Do Motor Behaviors (Mis)Match Affective Stimuli? An Evaluative Coding View of Approach and Avoidance Reactions,” Journal of Experimental Psychology: General , 137 2 , 262 . Google Scholar CrossRef Search ADS PubMed Eelen Jiska , Dewitte Siegfried , Warlop Luk ( 2013 ), “Situated Embodied Cognition: Monitoring Orientation Cues Affects Product Evaluation and Choice,” Journal of Consumer Psychology , 23 4 , 424 – 33 . Google Scholar CrossRef Search ADS Elder Ryan S. , Krishna Aradhna ( 2012 ), “The ‘Visual Depiction Effect’ in Advertising: Facilitating Embodied Mental Simulation through Product Orientation,” Journal of Consumer Research , 38 6 , 988 – 1003 . Google Scholar CrossRef Search ADS Ellis Rob , Tucker Mike ( 2000 ), “Micro‐Affordance: The Potentiation of Components of Action by Seen Objects,” British Journal of Psychology , 91 4 , 451 – 71 . Google Scholar CrossRef Search ADS PubMed Eriksen Barbara A. , Eriksen Charles W. ( 1974 ), “Effects of Noise Letters upon the Identification of a Target Letter in a Nonsearch Task,” Perception & Psychophysics , 16 1 , 143 – 9 . Google Scholar CrossRef Search ADS Fiore Ann Marie , Yah Xinlu , Yoh Eunah ( 2000 ), “Effects of a Product Display and Environmental Fragrancing on Approach Responses and Pleasurable Experiences,” Psychology and Marketing , 17 1 , 27 – 54 . Google Scholar CrossRef Search ADS Gibson James J. ( 1979 ), The Ecological Approach to Visual Perception , Boston : Houghton Mifflin . Hommel Bernhard ( 1993 ), “The Relationship between Stimulus Processing and Response Selection in the Simon Task: Evidence for a Temporal Overlap,” Psychological Research , 55 4 , 280 – 90 . Google Scholar CrossRef Search ADS Hommel Bernhard ( 1997 ), “Toward an Action-Concept Model of Stimulus-Response Compatibility,” Advances in Psychology , 118 , 281 – 320 . Google Scholar CrossRef Search ADS Janiszewski Chris ( 1990 ), “The Influence of Print Advertisement Organization on Affect toward a Brand Name,” Journal of Consumer Research , 17 1 , 53 – 65 . Google Scholar CrossRef Search ADS Joubert Olivier R. , Rousselet Guillaume A. , Fize Denis , Fabre-Thorpe Michèle ( 2007 ), “Processing Scene Context: Fast Categorization and Object Interference,” Vision Research , 47 26 , 3286 – 97 . Google Scholar CrossRef Search ADS PubMed Keller Carmen , Markert Franziska , Bucher Tamara ( 2015 ), “Nudging Product Choices: The Effect of Position Change on Snack Bar Choice,” Food Quality and Preference , 41 , 41 – 3 . Google Scholar CrossRef Search ADS Klesse Anne-Kathrin , Levav Jonathan , Goukens Caroline ( 2015 ), “The Effect of Preference Expression Modality on Self-Control,” Journal of Consumer Research , 42 4 , 535 – 50 . Kornblum Sylvan , Hasbroucq Thierry , Osman Allen ( 1990 ), “Dimensional Overlap: Cognitive Basis for Stimulus-Response Compatibility—A Model and Taxonomy,” Psychological Review , 97 2 , 253 . Google Scholar CrossRef Search ADS PubMed Krüger Tobias , Mata André , Ihmels Max ( 2014 ), “The Presenter’s Paradox Revisited: An Evaluation Mode Account,” Journal of Consumer Research , 41 4 , 1127 – 36 . Google Scholar CrossRef Search ADS Lembregts Christophe , Pandelaere Mario ( 2013 ), “Are All Units Created Equal? The Effect of Default Units on Product Evaluations,” Journal of Consumer Research , 39 6 , 1275 – 89 . Google Scholar CrossRef Search ADS MacDonald Angus W. , Cohen Jonathan D. , Andrew Stenger V. , Carter Cameron S. ( 2000 ), “Dissociating the Role of the Dorsolateral Prefrontal and Anterior Cingulate Cortex in Cognitive Control,” Science , 288 5472 , 1835 – 8 . Google Scholar CrossRef Search ADS PubMed Melara Robert D. , Marks Lawrence E. ( 1990 ), “Interaction among Auditory Dimensions: Timbre, Pitch, and Loudness,” Perception & Psychophysics , 48 2 , 169 – 78 . Google Scholar CrossRef Search ADS PubMed Miles James D. , Proctor Robert W. ( 2012 ), “Correlations between Spatial Compatibility Effects: Are Arrows More Like Locations or Words?” Psychological Research , 76 6 , 777 – 91 . Google Scholar CrossRef Search ADS PubMed Muchnik Lev , Aral Sinan , Taylor Sean J. ( 2013 ), “Social Influence Bias: A Randomized Experiment,” Science , 341 6146 , 647 – 51 . Google Scholar CrossRef Search ADS PubMed Orth Ulrich R. , Crouch Roberta C. ( 2014 ), “Is Beauty in the Aisles of the Retailer? Package Processing in Visually Complex Contexts,” Journal of Retailing , 90 4 , 524 – 37 . Google Scholar CrossRef Search ADS Peracchio Laura A. , Meyers-Levy Joan ( 1997 ), “Evaluating Persuasion-Enhancing Techniques from a Resource-Matching Perspective,” Journal of Consumer Research , 24 2 , 178 – 91 . Google Scholar CrossRef Search ADS Ping Raedy M. , Dhillon Sonica , Beilock Sian L. ( 2009 ), “Reach for What You Like: The Body’s Role in Shaping Preferences,” Emotion Review , 1 2 , 140 – 50 . Google Scholar CrossRef Search ADS Reber Rolf , Winkielman Piotr , Schwarz Norbert ( 1998 ), “Effects of Perceptual Fluency on Affective Judgments,” Psychological Science , 9 1 , 45 – 8 . Google Scholar CrossRef Search ADS Rodway Paul , Schepman Astrid , Lambert Jordana ( 2012 ), “Preferring the One in the Middle: Further Evidence for the Centre‐Stage Effect,” Applied Cognitive Psychology , 26 2 , 215 – 22 . Google Scholar CrossRef Search ADS Salganik Matthew J. , Watts Duncan J. ( 2008 ), “Leading the Herd Astray: An Experimental Study of Self-Fulfilling Prophecies in an Artificial Cultural Market,” Social Psychology Quarterly , 71 4 , 338 – 55 . Google Scholar CrossRef Search ADS Satterthwaite F. E. ( 1946 ), “An Approximate Distribution of Estimates of Variance Components,” Biometrics Bulletin , 2 6 , 110 – 4 . Google Scholar CrossRef Search ADS PubMed Schley Dan R. , Lembregts Christophe , Peters Ellen ( 2017 ), “The Role of Evaluation Mode on the Unit Effect,” Journal of Consumer Psychology , 27 2 , 278 – 86 . Google Scholar CrossRef Search ADS Schwarz Norbert ( 2004 ), “Metacognitive Experiences in Consumer Judgment and Decision Making,” Journal of Consumer Psychology , 14 4 , 332 – 48 . Google Scholar CrossRef Search ADS Schwarz Norbert ( 2006 ), “Feelings, Fit, and Funny Effects: A Situated Cognition Perspective,” Journal of Marketing Research , 43 1 , 20 – 3 . Google Scholar CrossRef Search ADS Shen Hao , Zhang Meng , Krishna Aradhna ( 2016 ), “Computer Interfaces and the ‘Direct-Touch’ Effect: Can iPads Increase the Choice of Hedonic Food?” Journal of Marketing Research , 53 5 , 745 – 58 . Google Scholar CrossRef Search ADS Simon J. Richard ( 1969 ), “Reactions toward the Source of Stimulation,” Journal of Experimental Psychology , 81 1 , 174 . Google Scholar CrossRef Search ADS PubMed Slovic Paul ( 1995 ), “The Construction of Preference,” American Psychologist , 50 5 , 364 – 71 . Google Scholar CrossRef Search ADS Spencer Steven J. , Zanna Mark P. , Fong Geoffrey T. ( 2005 ), “Establishing a Causal Chain: Why Experiments Are Often More Effective than Mediational Analyses in Examining Psychological Processes,” Journal of Personality and Social Psychology , 89 6 , 845 – 51 . Google Scholar CrossRef Search ADS PubMed Strack Fritz , Martin Leonard L. , Stepper Sabine ( 1988 ), “Inhibiting and Facilitating Conditions of the Human Smile: A Nonobtrusive Test of the Facial Feedback Hypothesis,” Journal of Personality and Social Psychology , 54 5 , 768 – 77 . Google Scholar CrossRef Search ADS PubMed Stroop J. Ridley ( 1935 ), “Studies of Interference in Serial Verbal Reactions,” Journal of Experimental Psychology , 18 6 , 643 . Google Scholar CrossRef Search ADS Tipples Jason ( 2002 ), “Eye Gaze Is Not Unique: Automatic Orienting in Response to Uninformative Arrows,” Psychonomic Bulletin & Review , 9 2 , 314 – 8 . Google Scholar CrossRef Search ADS PubMed Tucker Mike , Ellis Rob ( 2001 ), “The Potentiation of Grasp Types during Visual Object Categorization,” Visual Cognition , 8 6 , 769 – 800 . Google Scholar CrossRef Search ADS Valenzuela Ana , Raghubir Priya ( 2009 ), “Position-Based Beliefs: The Center-Stage Effect,” Journal of Consumer Psychology , 19 2 , 185 – 96 . Google Scholar CrossRef Search ADS Valenzuela Ana , Raghubir Priya ( 2015 ), “Are Consumers Aware of Top–Bottom but Not of Left–Right Inferences? Implications for Shelf Space Positions,” Journal of Experimental Psychology: Applied , 21 3 , 224 – 41 . Google Scholar CrossRef Search ADS PubMed Valenzuela Ana , Raghubir Priya , Mitakakis Chrissy ( 2013 ), “Shelf Space Schemas: Myth or Reality?” Journal of Business Research , 66 7 , 881 – 8 . Google Scholar CrossRef Search ADS Wascher Edmund , Reinhard Meike , Wauschkuhn Bernd , Verleger Rolf ( 1999 ), “Spatial SR Compatibility with Centrally Presented Stimuli: An Event-Related Asymmetry Study on Dimensional Overlap,” Journal of Cognitive Neuroscience , 11 2 , 214 – 29 . Google Scholar CrossRef Search ADS PubMed Weijters Bert , Geuens Maggie , Baumgartner Hans ( 2013 ), “The Effect of Familiarity with the Response Category Labels on Item Response to Likert Scales,” Journal of Consumer Research , 40 2 , 368 – 81 . Google Scholar CrossRef Search ADS Zhang Huazhong (Harry) , Zhang Jun , Kornblum Sylvan ( 1999 ), “A Parallel Distributed Processing Model of Stimulus–Stimulus and Stimulus–Response Compatibility,” Cognitive Psychology , 38 3 , 386 – 432 . Google Scholar CrossRef Search ADS PubMed Zhu Feng , Zhang Xiaoquan ( 2010 ), “Impact of Online Consumer Reviews on Sales: The Moderating Role of Product and Consumer Characteristics,” Journal of Marketing , 74 2 , 133 – 48 . Google Scholar CrossRef Search ADS Zhu Rui , Meyers-Levy Joan ( 2009 ), “The Influence of Self-View on Context Effects: How Display Fixtures Can Affect Product Evaluations,” Journal of Marketing Research , 46 1 , 37 – 45 . Google Scholar CrossRef Search ADS © The Author(s) 2018. Published by Oxford University Press on behalf of Journal of Consumer Research, Inc. All rights reserved. For permissions, please e-mail: 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

Journal of Consumer ResearchOxford University Press

Published: Oct 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