TY - JOUR
AU - Hancock, Jeffrey T
AB - Abstract A salient issue for online romantic relationships is the possibility of deception, but it is unclear how lies are communicated before daters meet. We collected mobile dating deceptions from the discovery phase, a conversation period after daters match on profiles but before a face-to-face interaction. Study 1 found that nearly two-thirds of lies were driven by impression management, particularly self-presentation and availability management goals. Study 2 found that approximately 7% of messages were deceptive, and content patterns were consistent with Study 1. Across studies, the participant’s lying rate was correlated with the perceived lying rate of the partner. We discuss the implications of these data in relation to impression management, deception theory, and online dating research. The use of mobile dating apps to facilitate romantic relationships has increased at an unprecedented rate. According to Smith and Anderson (2016), nearly one quarter of young adults use online dating or mobile applications to form romantic relationships, and the frequency nearly doubles for those on same-sex dating platforms (Blackwell, Birnholtz, & Abbott, 2015). Mobile dating apps, such as Tinder or Hinge, are appealing because they use location-based technology to connect people with similar interests and preferences, but these platforms still elicit many of the same interpersonal concerns of face-to-face communication and traditional online dating. Chief among these concerns is deception, or the deliberate attempt by one communicator to produce a false belief in another (Levine, 2014; Vrij, 2008). Online dating deception research has often considered how people misrepresent social information (Corriero & Tong, 2016), how online appearances compare to offline appearances (Hancock & Toma, 2009), and the relationship between personality characteristics and deception (Ranzini & Lutz, 2017). It is unclear, though, how often deception is used in the period of message exchange after daters have matched on profiles, but before they have decided to meet in person. In two studies, we evaluated the frequency of deception from actual messages exchanged between daters on mobile dating applications and categorized the deceptive message content to understand what people lied about in their conversations. Impression management in online dating deception Evidence for deception in online dating has been documented across three types of dating profile information (Toma, 2015): (a) physical attributes, (b) personal interests, and (c) photos. For example, deceptions related to physical attributes tend to differ by gender to meet societal ideals by the opposite sex. Men often exaggerate their height to be more attractive to women and women often underestimate their weight to be more attractive to men (Toma, Hancock, & Ellison, 2008). Details about a person’s interests, activities, or likes are also a crucial part of online dating and, in general, people may use deception to make the self appear more attractive or compelling (Kashy & DePaulo, 1996). Profile deceptions can help the dater construct a version of the self that persuades prospective dates; in this sense, the dating profile is an advertisement. Daters can promote the most marketable self with their profile, and deception is one strategy to accomplish this (Toma & Hancock, 2012). Profile photos are also carefully selected for the most attractive or attention-grabbing portrayals of the self, which may not be recent or entirely authentic at the time. These photos often fit gendered self-presentation goals as well, with women showing younger versions of themselves and men displaying wealth and status (Hancock & Toma, 2009). Embellishing a dating profile, overstating interests, and adjusting photos are broadly considered impression management strategies for the online dater (Ellison, Heino, & Gibbs, 2006). More generally, people sometimes try to manage the impressions of others by engaging in practices of deceptive self-presentation (see DePaulo, 1992), where deception is considered a tactic to strategically self-enhance (Weiss & Feldman, 2006, p. 1071). For example, in a non-romantic setting, Feldman, Forrest, and Happ (2002) had a pair of participants engage in a 10-minute face-to-face conversation that was secretly taped. One subject in the pair was instructed to appear either (a) likeable, (b) competent, or (c) to become familiar with the other participant without any prescribed self-presentation goal. Participants who were instructed to appear likeable or competent told more lies than people who were not given self-presentation instructions. Given that impression management goals can trigger deception tactics, when online dating is the setting of interest, there are clear impression management objectives (e.g., to appear attractive to the partner). Indeed, the probability of deception may be amplified because of the high-stakes nature of appearing attractive to potential mates (see Walczyk, Harris, Duck, & Mulay, 2014). Another study on face-to-face job interview deception had job candidates respond to questions about classroom performance, accomplishments, or extracurricular activities (Weiss & Feldman, 2006). After the interview, participants reviewed their videotaped interaction for any inaccurate statements. The data revealed that impression management lies were pervasive for the interviewee and often communicated to self-enhance (e.g., over 90% of lies were told to achieve an impression management goal). Job interviews and online dating are similar settings (see Ellison, Hancock, & Toma, 2012; Heino, Ellison, & Gibbs, 2010), as people are incentivized to present their best or most attractive self, the discourse structure is dyadic, and people can adjust their communication patterns based on positive or negative feedback. The parallels between these settings, along with prior evidence suggesting that impression management, self-presentation, and deception are tightly linked (DePaulo, 1992; Feldman et al., 2002; Toma & Hancock, 2012; Weiss & Feldman, 2006), suggest that these deception types will likely play an important role in mobile dating conversations. To our knowledge, limited research has evaluated the frequency and content of actual deceptive messages from mobile dating conversations. In two studies, we evaluated how often people lie to their mobile dating partner and whether impression management strategies are particularly salient in the message content. We focused on mobile dating instead of web-based online dating due to the emphasis on texting conversations after people match on dating profiles and the pervasiveness of these platforms (Blackwell et al., 2015; Corriero & Tong, 2016; Smith & Anderson, 2016). The discovery phase There are typically three important periods in a mobile dating relationship (Markowitz, Hancock, & Tong, 2018). The first, which we call the profile stage, refers to the period when mobile daters individually develop a profile to create the best depiction of the self. Similar to an online dating profile, each dater can provide photos from social networking sites, a brief self-description in a free-response section of the profile, and other characteristics (e.g., employment, interests) to make the self appear attractive. After the profile stage, daters attempt to connect with others based on the profile data. On most dating apps, this is the matching stage, where people scroll through profiles and, based on the profile data (e.g., photos, small text descriptions, common connections that may be available on the app), decide on the potential to connect with the date or reject that opportunity. Finally, if two people have established mutual interest in the profile stage and they match, they can communicate via text. We call this period of message exchange the discovery phase, because it represents a time when conversation plays a crucial role in the decision to meet the other person face-to-face (Markowitz et al., in press). Recall, prior evidence suggests that online dating profiles often contain traces of deception in the text descriptions (Toma & Hancock, 2012) and photos (Hancock & Toma, 2009). After daters match on profiles, however, they should still use deception as a resource during the discovery phase because impression management remains a crucial part of the “getting to know you” process of online conversations (see Bazarova, Taft, Choi, & Cosley, 2013). For example, daters may lie about their mutual interests, daily activities, or intentions for dating, to reach the ultimate goal of meeting the person face-to-face. Therefore, deception is not just an important interpersonal dynamic for profiles, but it carries into discovery phase conversations to fulfill certain impression management goals. Self-presentation and availability management Two distinct impression management goals should be reflected during the discovery phase. First, daters should be less concerned with promoting their physical attributes, given that the profile stage has passed and both people have matched. Instead, consistent with prior research suggesting that people use deception to accomplish a variety of self-presentation and self-enhancement objectives (DePaulo, 1992; Feldman et al., 2002; Weiss & Feldman, 2006), daters should be more interested in coming across as romantically available and desirable (e.g., acting witty, promiscuous, interesting, or appearing attractive; Ellison et al., 2006). This distinction between physical attribute lies and romantic availability lies is similar to the difference between search goods and experience goods in consumer behavior research (Klein, 1998). Search goods are typically objects with value that can be assessed before purchase (e.g., the ripeness of a piece of fruit, or data that people can gather from a person’s online dating profile), while the characteristics of experience goods are often valued after consumption (e.g., a bottle of wine, or the dateability of a person that can be assessed through conversation). In the discovery phase, we should observe verbal deceptions related to appearing interpersonally striking and available, rather than deceptions that are related to physical attributes. For example, daters may mention false details about friends to appear social or discuss their job to suggest that they have a stable occupation, though the frequency of these self-presentation lies is uncertain. With our first research question, we asked: RQ1: How often are deceptions produced in the discovery phase of mobile dating conversations to accomplish self-presentation goals? A second impression management goal during the discovery phase should be to decide whether to meet, avoid, or continue talking to the match. People can accomplish this by drawing on the affordances of mobile technology to manage their social availability. For example, the asynchronicity of texting allows people to delay a response if they require more time to form an utterance. The portability of mobile communication also allows people to answer messages in different locations and obscure their whereabouts. Prior work examining text messages between friends, colleagues, and family suggests that people draw on these affordances to often lie about their location, activities, and their internal feelings, because it is face-threatening to tell someone that you do not want to see them. These availability management messages are called “butler lies,” such as the excuse of “sorry my phone was dead,” after not acknowledging a prior message (Hancock et al., 2009). Butler lies are deceptive because they deliberately instill a false belief in another person (e.g., the phone may not have been dead), but they are social in nature and less malicious than bald-faced lies (e.g., the person did not want to talk at the moment and, therefore, lied about his or her availability). People may also use deception as an availability management strategy, particularly in new romantic relationships, because appearing always available can be an undesirable trait. These lies help to control how often people communicate and how desperate they appear (Ellison et al., 2006). In everyday text messaging between known partners, butler lies are common, comprising nearly 20-40% of the texting lies (see French, Smith, Birnholtz, & Hancock, 2015). Availability management lies are also found across cultures, when people are motivated to avoid others for conflict, punishment, embarrassment, or impoliteness reasons (Levine, Ali, Dean, Abdulla, & Garcia-Ruano, 2016). Given that one objective during the discovery phase is to decide whether to meet the other person face-to-face, managing impressions about one’s availability should be another important goal. Our second research question asks: RQ2: How often are butler lies used during discovery phase conversations to accomplish availability management goals? Perception of lies In addition to our interests in discovering the frequency and content of impression management lies, an important question is whether a participant believed that his or her partner lied during the discovery phase. Most people assume that their interpersonal interactions are predominantly honest (Levine & McCornack, 1992; McCornack & Parks, 1986). This perspective is connected to the truth bias, which proposes that people infer honesty independent of message veracity in most social situations (see Truth-Default Theory; Levine, 2014). The truth bias is one of the strongest and most pronounced effects in deception research (Bond & DePaulo, 2006), and likely exists because people often do not have enough data to make accurate veracity judgements (see ALIED Theory; Street, 2015). In the absence of verifiable details or clues to reveal deceit, people use contextual information (e.g., prior knowledge, social information about communicators) to estimate if a person is lying or telling the truth (Street, 2015). Since most people believe that lying is socially undesirable (see Bok, 1999), they often assume that their partner is communicating honestly because enough valid counter-information does not typically exist. Therefore, the truth bias likely guides how people judge the veracity of their partner in the discovery phase, and we explored the perception of lies in this conversation period. RQ3: How often are partner messages perceived as deceptive? Is a participant’s perception of their partner’s lying behavior influenced by their own lying behavior? Prior work on egocentrism biases suggests that we judge others based on our own actions, often called the false consensus effect (Epley, 2015). People perceive the behavior of others in highly predictable ways, such that they set a reference point based on their own actions and then adjust that anchor to suggest how others will act. For example, in a seminal paper by Ross, Greene, and House (1977, Study 3), college students were asked to advertise for a campus restaurant by wearing a sign. Students viewed their own actions as typical; those who agreed or disagreed to wear the sign believed that others would largely match their preference as well. This study, and a long tradition of research on false consensus effects, demonstrates that people estimate the actions of others based on their own behavior (see Epley, 2015). We examined a false consensus effect for deception and expected that the more people lie, the more they will believe that their partner lied as well. H1: The rate of perceived partner lying will be correlated with participant’s lying rate. Study 1 Method One hundred ten participants were recruited from Amazon Mechanical Turk and paid for a 30-minute study on mobile dating apps and texting. All participants must have met another person on a dating app and communicated with the same person over text messaging. We asked participants to provide the mobile dating and text messaging conversations with a partner to understand if lying rates were different across these systems. This also allowed us to compare the mobile dating deception rates with a similar system that has been studied in deception research (Birnholtz, Guillory, Hancock, & Bazarova, 2010; French et al., 2015). Participant exclusion criteria included failure to comply with the study’s guidelines or failure to complete all aspects of the procedure (e.g., importing nonsensical placeholders instead of actual messages). This study was approved by the authors’ Institutional Review Board. Given the sensitive nature of the data and because we collected personal information from dating conversations, we encrypted the data files and stored them on a secure server. Participants We removed five participants from the dataset after these individuals met our exclusion criteria (final N = 105). Males (n = 54, M = 28.1 years), females (n = 47; M = 26.4 years), those designating an “other” gender (n = 3; M = 25.3 years), and those not wanting to report a gender (n = 1, M = 19 years) reported using a variety of dating apps. Most participants used Tinder (n = 78), with other apps having lower frequencies (OkCupid, n = 9; Grindr, n = 7; Bumble, n = 4; MeetMe, n = 3; Plenty of Fish, n = 2; Her, n = 1; Jackd, n = 1). OkCupid and Plenty of Fish are online dating sites with mobile dating apps, but these data were included because participants were instructed to provide data from their connection on the mobile part of the service only. Procedure After a home screen instructed people about the nature of the study and they agreed to participate, subjects were told that they would import messages from a recent connection that started on a mobile dating app and also transferred to text messaging. Each person imported a total of approximately 20 messages, 10 sent on the mobile dating app and 10 sent on the texting system. Instructions and examples were provided to describe the message importing process. Crucially, instructions were also provided about what deception is and how it is a normative part of everyday conversation (see Supplementary Materials; Levine, 2014). Participants were told to start with the most recent message and not the first exchange from each system. This helped to limit the number of introductory (e.g., “Hello”) and perfunctory messages. In the same message timeframe, participants were also asked to estimate how many times they believe that they were lied to, which allowed us to test H1. After each message was imported, participants rated the deceptiveness of their statements on a 5-point scale (1, not deceptive at all; 2, somewhat deceptive; 3, moderately deceptive; 4, largely deceptive; 5, extremely deceptive). After deceptiveness ratings, we asked participants how much they liked their mobile dating partner on a 5-point scale (1, dislike a great deal; 5, like a great deal), since prior work suggests that lying rates and deception content may be affected by this social dynamic (Bell & DePaulo, 1996). Message coding Messages reported as deceptive were coded by two independent raters and assigned to one of three categories: (a) self-presentation lies, (b) butler lies to manage availability, or (c) other. Raters were given instructions about the differences between these lie types and examples in each category. For example, self-presentation lies help the dater appear attractive and can include any self-enhancing message where the dater is attempting to be witty, promiscuous, interesting, or appear like a good match for the other person (e.g., “I really like to hike and bike and read a lot. I am always outside. You?”; Ellison et al., 2006). Butler lies, or false messages to manage social availability, included lies related to entering, exiting, or arranging an interaction (e.g., “OK. I’ll call in a bit. I need to do some stuff”). The “other” category included short agreements or disagreements (e.g., “yes,” “sounds good,” “nope”), which were marked as deceptive, but did not contain enough information to be rated as self-presentation or butler content. The raters achieved good agreement for these messages (κ = .62; discrepancies resolved by consensus after discussion). Results The frequency of participant lies is presented on the upper left side of Figure 1 and the estimated frequency of partner deceptions is on the upper right side of Figure 1 (see Table 1 as well, for descriptive statistics of lying rates). These data do not fit normal distributions; with many low values and few high values, they resemble a power function (Serota & Levine, 2015). Figure 1 View largeDownload slide Frequency distribution of lying rates across participant messages (striped bars) and perceived partner messages (solid bars) for Study 1 (top two figures) and Study 2 (bottom two figures). Figure 1 View largeDownload slide Frequency distribution of lying rates across participant messages (striped bars) and perceived partner messages (solid bars) for Study 1 (top two figures) and Study 2 (bottom two figures). Table 1 Descriptive Statistics for Participant Lying Rates and Liking, and Number of Messages by Deceptiveness Study 1 Lying Liking M(SD) 3.33 lies (4.31 lies) 3.92 (0.72) First quartile 0 lies 3.00 Median 2 lies 4.00 Third quartile 5 lies 4.00 Study 2 Lying Liking M(SD) 1.20 lies (2.81 lies) 4.93 (1.18) First quartile 0 lies 4.00 Median 0 lies 5.00 Third quartile 1 lie 6.00 Study 1: Deceptiveness Dating app Texting χ2(3) p Somewhat deceptive 108 100 1.25 .74 Moderately deceptive 56 46 Largely deceptive 10 11 Extremely deceptive 8 11 Study 2: Deceptiveness Beginning End χ2(3) p Somewhat deceptive 42 13 4.88 .18 Moderately deceptive 22 18 Largely deceptive 14 8 Extremely deceptive 9 5 Study 1 Lying Liking M(SD) 3.33 lies (4.31 lies) 3.92 (0.72) First quartile 0 lies 3.00 Median 2 lies 4.00 Third quartile 5 lies 4.00 Study 2 Lying Liking M(SD) 1.20 lies (2.81 lies) 4.93 (1.18) First quartile 0 lies 4.00 Median 0 lies 5.00 Third quartile 1 lie 6.00 Study 1: Deceptiveness Dating app Texting χ2(3) p Somewhat deceptive 108 100 1.25 .74 Moderately deceptive 56 46 Largely deceptive 10 11 Extremely deceptive 8 11 Study 2: Deceptiveness Beginning End χ2(3) p Somewhat deceptive 42 13 4.88 .18 Moderately deceptive 22 18 Largely deceptive 14 8 Extremely deceptive 9 5 Note: Liking was measured on a 5-point scale in Study 1 and a 7-point scale in Study 2. Table 1 Descriptive Statistics for Participant Lying Rates and Liking, and Number of Messages by Deceptiveness Study 1 Lying Liking M(SD) 3.33 lies (4.31 lies) 3.92 (0.72) First quartile 0 lies 3.00 Median 2 lies 4.00 Third quartile 5 lies 4.00 Study 2 Lying Liking M(SD) 1.20 lies (2.81 lies) 4.93 (1.18) First quartile 0 lies 4.00 Median 0 lies 5.00 Third quartile 1 lie 6.00 Study 1: Deceptiveness Dating app Texting χ2(3) p Somewhat deceptive 108 100 1.25 .74 Moderately deceptive 56 46 Largely deceptive 10 11 Extremely deceptive 8 11 Study 2: Deceptiveness Beginning End χ2(3) p Somewhat deceptive 42 13 4.88 .18 Moderately deceptive 22 18 Largely deceptive 14 8 Extremely deceptive 9 5 Study 1 Lying Liking M(SD) 3.33 lies (4.31 lies) 3.92 (0.72) First quartile 0 lies 3.00 Median 2 lies 4.00 Third quartile 5 lies 4.00 Study 2 Lying Liking M(SD) 1.20 lies (2.81 lies) 4.93 (1.18) First quartile 0 lies 4.00 Median 0 lies 5.00 Third quartile 1 lie 6.00 Study 1: Deceptiveness Dating app Texting χ2(3) p Somewhat deceptive 108 100 1.25 .74 Moderately deceptive 56 46 Largely deceptive 10 11 Extremely deceptive 8 11 Study 2: Deceptiveness Beginning End χ2(3) p Somewhat deceptive 42 13 4.88 .18 Moderately deceptive 22 18 Largely deceptive 14 8 Extremely deceptive 9 5 Note: Liking was measured on a 5-point scale in Study 1 and a 7-point scale in Study 2. We observed a consistent number of lies across deceptiveness rating (e.g., somewhat deceptive, moderately deceptive, largely deceptive, extremely deceptive) and systems (e.g., mobile dating app, texting; χ2[3] = 1.25, p = .74; φc = .06; see Table 1). Therefore, we combined the messages from these two systems into one collection of messages for each participant. The number of lies did not differ across male, female, or other participants (F < 1, p = .87). Content analysis of deceptive messages Self-presentation and availability management goals played an important role in the deceptions of the discovery phase. That is, nearly two-thirds of the participant deceptions (231/350; 66%) involved self-presentation lies and lies to manage availability. Self-presentation deceptions Close to one-third of participant lies were self-presentation deceptions (131/350; 37.4%). An exploratory content analysis of the deceptive messages from the discovery phase confirmed that participants often lied to amplify their attractiveness. Daters often tried to appear more attractive by falsely stating preferences to match the other dater on interests and opinions. For example, “God, I love that show. I could literally keep watching it over and over” (Participant #91, rated as somewhat deceptive). In this case, the participant was likely trying to show a common interest. This could provide the foundation for future conversations and make the participant appear more attractive. Other self-presentation lies were similar. For example, “Damn you just keep hyping the book up lol bible is my favorite tho” (P#23, somewhat deceptive) was likely a false message about the participant’s favorite book. This deception might make the participant appear more attractive, interesting, and relatable to the partner. Presumably, the dater was trying to suggest that he or she is religious and the Bible is important to him or her, something that may also be important to the partner. Finally, another participant stated, “Haha all I want is to walk into a grocery store and buy the entire shelf of Bold Rock” (P#45, moderately deceptive), exaggerating the desire to buy an entire shelf of hard cider and making the self appear witty or interesting. Availability management deceptions Approximately 30% of participant deceptions were lies to manage availability (100/350; 28.6%), and participants often lied about their availability for activities, their internal states, and time. In one example, the speaker tried to avoid meeting the date with a lie, but he or she still tried to preserve the relationship. Hey I’m so so sorry, but I don’t think I’m going to be able to make it today. My sister just called and I guess she’s on her way here now. I’d be up for a raincheck if you wanted, though. Sorry again. (P#3, extremely deceptive) Other participants suggested that they could not meet because of work and the face-to-face activity would have to be postponed, “Lol alright. Well I have my finals on Wednesday and then I’m leaving on a vacation Thursday. So a couple weeks at least” (P#105, largely deceptive). In another case, a participant likely lied about prior activities to avoid meeting. In the message, “Probably not tonight man, i’m exhausted and have had kind of an emotional day” (P#30, somewhat deceptive), the participant hedged and said that he or she “probably” could not meet with the partner. Many people also used butler lies to conceal their internal beliefs or emotional states in the discovery phase. These lies concealed the participant’s feelings and kept information hidden from the partner. One example, “i wish i could go” (P#1, somewhat deceptive), falsely represented the participant’s feelings about going to an event. This strategy suggests that mobile daters may initially provide limited information to strangers. These deceptions are prosocial, though, as they can save face for both communicators and preserve the relationship in the event that the daters meet face-to-face. Finally, participants made several references to time in their butler lies, as people either overestimated or underestimated time when arranging an in-person meeting, or misinformed their partner with specific time references. For example, “I’m 30 minutes out. Be there as soon as i can” (P#104, somewhat deceptive), falsely stated the proximity of the participant to the partner. This lie drew on the ambiguities of mobile communication (the sender’s location) to provide a false belief that the participant would arrive at a specific time. Another example, “cool i should be available at 5” (P#34, somewhat deceptive), suggests that the participant lied about when he or she would be free to chat or meet. The dater also equivocates with the word “should,” which leaves some room for his or her availability to change. If the participant became available or unavailable, he or she would not feel obligated to the partner because a meeting was not fixed. False consensus effect for deception To examine whether a participant’s deceptive behavior was related to their perceptions of deception in partner messages, we calculated a bivariate correlation for the number of participant lies and perceived lies by the partner. We report Spearman’s ρ because the data are not normally distributed and may be sensitive to extreme values (Halevy, Shalvi, & Verschuere, 2014). We observed a strong, positive relationship between participant and partner lying rates, (ρ = .70, p < .001; two-tailed), even after excluding the most prolific liars (defined as participants with at least 19 reported lies, total; ρ = .68, p < .001; remaining n = 102). This relationship remained strong even after sub-setting the data, such as after excluding participants who told zero lies (ρ = .62, p < .001; n = 69) and participants who thought they were not lied to (ρ = .64, p < .001; n = 71). These data support our prediction for H1 and are consistent with a false consensus effect for deception (see the left panel of Figure 2 for a scatter plot of participant and perceived partner lying rates). Figure 2 View largeDownload slide Study 1 (left panel) and Study 2 (right panel) scatterplots of the total number of participant and perceived partner lies. Figure 2 View largeDownload slide Study 1 (left panel) and Study 2 (right panel) scatterplots of the total number of participant and perceived partner lies. Liking and deception Liking was not associated with the number of lies from the participant (ρ = –.15, p = .13, n = 104; one participant did not report liking) nor the number of perceived partner lies (ρ = –.18, p = .068), even when prolific liars were excluded (ρ’s < –.16, p’s > .11). When we excluded participants who told zero lies, however, participant lying rates and liking were negatively associated and significant (ρ = –.27, p = .024). Why did daters who reported at least one deception report lying less to partners they liked? One possibility is that availability management played a role. Daters who liked their partner less may have used butler lies more to avoid meeting. Indeed, when we consider daters with at least one lie and separate the lies by impression management type, liking was negatively associated with lies to manage availability (ρ = –.33, p = .006), but not with the number of self-presentation lies (ρ = .13, p = .31). These results suggest that the less a dater liked his or her partner, the more likely he or she was to lie about not being available. Discussion Deceptions during the discovery phase were instrumental for achieving self-presentation and availability management goals, and these impression management strategies accounted for a large portion of the deceptions during this time of message exchange. Self-presentation lies attempted to make the dater appear interesting and dateable, evidence that is consistent with prior research suggesting that people often lie to increase their attractiveness (Feldman et al., 2002; Weiss & Feldman, 2006). We also found that daters used deception to manage their social availability. That is, daters described why they could not meet face-to-face, lied about their current activities, and falsely described their feelings about the partner. The results from Study 1 also demonstrate a relationship between the number of lies a participant reported and the number of partner messages perceived as lies. These data support H1 and are consistent with a false consensus bias for deception during the discovery phase of mobile dating. This is some of the first evidence in a dating setting to identify that perceptions of false behavior are linked to the individual’s reported false behavior. While the data suggest that deception is important for discovery phase conversations, there are several limitations of the study. Since we did not ask participants to explain their deception ratings, we cannot be confident whether the messages were an attempt to create a false belief in the partner or examples of jocularity. Indeed, prior work (e.g., Birnholtz et al., 2010; French et al., 2015) suggests that people sometimes rate messages that contain jokes as deceptive (e.g., “his head is as big as the moon”), in which the statement is not literally true, but it is also not intended to create a false belief. Second, we only gathered the last (most recent) messages exchanged between two daters, which ignores the first messages exchanged. In Study 2, we extended the prior method to address such shortcomings. First, following methods from prior studies on butler lies (Birnholtz et al., 2010; French et al., 2015), we asked participants to explain why they rated messages as deceptive so that we could exclude any jocular messages that were in fact not intended to create a false belief. Second, we collected lies from the beginning and the end of the discovery phase to determine if lying rates are different at the conversation’s onset or resolution. We also recruited other characteristics of the interpersonal relationship that might be related to a participant’s lying rate, such as liking, trust, and attractiveness of the partner, motivations for mobile dating (e.g., using a dating app to browse), and relationship outcomes (e.g., if the daters met face-to-face or not). These data can help to understand if specific interpersonal dynamics are associated with deception frequencies. Study 2 Method One hundred ten participants were recruited from Amazon Mechanical Turk and paid for a 30-minute study on mobile dating apps. Given that we did not observe any differences in deception rates between mobile dating apps and texting in Study 1, we only collected messages sent within the mobile dating apps in Study 2. One participant was removed after importing nonsensical messages (final N = 109). This study was also approved by the authors’ Institutional Review Board. Participants Males (n = 76, M = 28.7 years) and females (n = 33; M = 28.3 years) used a variety of mobile dating apps. Most participants used Tinder (n = 84), with other apps again having lower frequencies (OkCupid, n = 7; Grindr, n = 7; Bumble, n = 6; Plenty of Fish, n = 1; Facebook, n = 1; Badoo, n = 1; eHarmony, n = 1; WhatsApp, n = 1). Consistent with Study 1, we included data from OkCupid, Plenty of Fish, eHarmony, Facebook, and WhatsApp because participants correctly followed instructions to import messages from a romantic interest that they communicated with online and traditional social media are often used as dating platforms as well (Safronova, 2017). Procedure Data collection followed the procedure outlined in Study 1, except participants were now asked to import the first (beginning) 10 sent messages and the last (most recent) 10 sent messages to their partner on the mobile dating app. First, a maximum of 10 dialogue boxes were provided to participants in order to import messages from the beginning of the conversation. After participants imported the first messages, new instructions asked for sent messages from the end of the conversation, which also included a total of 10 dialogue boxes. If the interaction lasted less than 20 messages, participants provided at least 10 messages from the beginning, and the remaining messages for the end of the interaction. Participants rated the deceptiveness of each statement based on the prior definition of deception that was provided and, following the rating, they were asked to describe why a message was deceptive. That is, a dialogue box appeared asking them to provide an explanation of the lie (i.e., “Why did you rate this message as deceptive? Please provide your explanation below and tell us why it was false.”). Finally, to evaluate H1, we asked participants to report how many received messages they thought were deceptive over the course of the message exchange they provided. Message coding Messages were coded in two phases. In the first phase, each message marked as deceptive was examined for jocularity, which included messages that may have been non-literal but were not intended to create a false belief. For example, the message “Good pick! Can you teach me to cha cha?” was rated by the participant as somewhat deceptive, but the participant explained, “I was mostly just trying to be funny and had no intentions of learning to dance.” In another example, a participant said, “now ask me if I am a Double Decker bus,” which was marked as moderately deceptive, but his or her explanation stated that the message was a “joke.” Inter-rater reliability for jocularity was good (κ = .74). Of the 139 messages identified as deceptive by participants, 8 were jocular and were excluded from further analysis. In the second phase, the remaining 131 deceptive messages were coded according to the procedure from Study 1, categorizing deceptive messages as related to self-presentation or availability management. Messages that could not be identified as having self-presentation or availability management goals were assigned to the “other” deception category (n = 21 messages). Independent raters also achieved good agreement on this phase of the coding (κ = .71; discrepancies resolved by consensus after discussion). Interpersonal dynamics We asked several questions to probe how dynamics of the interpersonal relationship might relate to deception rates. We asked participants how much they (a) liked their partner (7-point scale), (b) trusted their partner (7-point scale), and (c) how attractive they judged their partner (5-point scale), with low scale points indicating negative perceptions. We also asked about the relationship outcome (e.g., if the daters met). Finally, we asked about the participant’s motivations for using mobile dating apps based on a scale adapted from van de Wiele and Tong (2014). The scale has participants rate over 20 statements that measure the importance of typical mobile dating app uses (e.g., to feel less lonely, to get an “ego boost,” to satisfy my social curiosity, etc.; on a 5-point scale where 1 = not at all important and 5 = extremely important). Consistent with van de Wiele and Tong (2014), an exploratory factor analysis extracted 6 dimensions (Kaiser-Meyer-Olkin Measure of Sampling Adequacy = .81; see Supplementary Table S1). All retained factors had eigenvalues greater than 2.07, individual varimax rotated factor loadings were greater than .58 with no cross-loadings, and the factors cumulatively accounted for 72.60% of total variance. Each scale item was standardized and then combined into an index based on the factor extraction for further analysis (all Cronbach’s α’s > .71, average Cronbach’s α = .82). Results As the bottom distributions of Figure 1 display, the number of reported and perceived lies resemble a power function (see Table 1 for descriptive statistics as well). Seven percent of the participant mobile dating messages were rated as deceptive (131/1840; 7.1%). Note, although we asked for a combined total of 20 mobile dating messages from the beginning and end of the discovery phase, not all participants sent 20 utterances. As such, the total message frequency (N = 1,840) does not reach the maximum number we could have collected (i.e., 20 messages × 109 subjects = 2,180). On average, participants provided 16.9 total messages (SD = 5.21 messages, Mdn = 20 messages). Approximately 58% of participants provided all 20 messages (64/109). The first quartile for the total number of messages provided was 14.5 messages and the third quartile was 20 messages. Only 7% of the participants reported 5 or fewer messages (8/109; 7.3%). More lies were produced in the beginnings of the interactions (87/979 messages; 8.9%) than the ends of the interactions (44/861 messages; 5.1%). For those who told at least one lie, they told more lies in the beginning messages (M = 2.23 lies per person, SE = .34 lies per person) than the end messages (M = 1.13 lies per person, SE = .34 lies per person; F[1, 38] = 16.75, p < .001). However, ratings of deceptiveness were not statistically different between the beginning and end of the conversation (χ2[3] = 4.88, p = .18; φc = .19; see Table 1), suggesting that the magnitude of lies was similar across these two periods. Therefore, consistent with Study 1, we again combined the messages from the two periods into one collection of messages for each participant. Lying frequencies were also equally prevalent for men and women (t < 1, p = .52). Content analysis of deceptive messages Over 80% (110/131) of deceptions from participants contained an impression management strategy (e.g., self-presentation lies, lies to manage availability), a lying rate that is in line with content patterns of self-presentation and self-enhancement goals in prior deception research (Weiss & Feldman, 2006). Self-presentation deceptions Consistent with the data in Study 1, many messages used deception to appear attractive. Nearly two-fifths of deceptions (51/131; 38.9%) were related to self-presentation and self-enhancement. People often communicated their interests and attraction to the date (“Listen, I really love dogs of all kinds, but I like you by virtue that you love your dog! You seem like such a good dog dad!,” P#11, moderately deceptive), and deceptive statements tried to make the participant seem like a good match for the partner. For example, “Blue is my favorite color also” (P#73, somewhat deceptive) was a participant lie, and the individual explained that green was actually his or her favorite color. This suggests that lying helped to relay a false personal interest that might draw the attention of the partner and increase his or her attractiveness. A second lie by the same person, “I wish 100 puppies were climbing all over me” (P#73, largely deceptive) was rated as false because the participant explained that he or she is allergic to dogs. To establish a connection with the partner, presumably, a false message relayed that the two daters had common interests that could be the foundation for a relationship. Availability management deceptions In Study 2, 45% of impression management deceptions (59/131) were butler lies to manage availability, where participants often lied about their activities. For example, the statement, “Not tonight, Its late and I’m so tired, have to be up early for work tomorrow.” (P#95, somewhat deceptive) helped the participant avoid an in-person meeting. The participant explained, “I was a little tired but I mostly didn’t want to meet them because it was late at night and I didn’t feel comfortable.” Other examples describe how deception can help to strategically manage availability and the relationship. For instance, the message “lol, I know it seems like I’m blowing you off but I promised to do family stuff today. Let’s grab coffee sometime this week” (P#64, largely deceptive) falsely suggests that there were events preventing an in-person meeting. The participant explained, “I was blowing him off, I did have family stuff but I also had time for coffee,” indicating deception helped to manage his or her social availability. Finally, participants sometimes referred to technology to decelerate the mobile dating relationship. For example, participants blamed the technology for unresponsiveness or fear about mobile dating, “Im sorry I can’t text currently my phone is not working” (P#87, extremely deceptive), when the explanation for deception revealed “My phone was fine. I just get too many stalkers.” These data suggest that mobile media can act as a buffer to delay communication. False consensus effect for deception Participant lying behavior was correlated with the perception of the partner’s lying (ρ = .61, p < .001; N = 109; two-tailed). Importantly, this relationship remained strong after sub-setting the data, such as after excluding the most prolific liars (ρ = .60, p < .001; remaining n = 108), excluding participants who told zero lies (ρ = .58, p < .001; n = 39), and excluding participants who thought they were not lied to (ρ = .40, p = .005; n = 49). This evidence again supports H1 and is consistent with the false consensus effect of deceptive messaging (see the right panel of Figure 2). Relationship dynamics Liking, trust, attractiveness, and relationship outcome Participant lying rates were uncorrelated with liking, trust, or attractiveness (see Supplementary Materials for bivariate correlations). Lying was also unaffected by the relationship outcome (e.g., if the daters met). Motivations for use We performed bivariate correlations between the six extracted factors and participant lying rates, and three significant relationships emerged. First, participant lying rates were positively associated with the factor related to social inclusion, attention, and approval (ρ = .21, p = .028), which consisted of five items from the scale (e.g., to get validation from others, to get an “ego boost,” to feel less lonely, to get attention from others, to get compliments from others). Second, participant lying rates were positively correlated with the factor related to sex (ρ = .26, p = .007), which consisted of four items (e.g., to find new sexual partners, to hook up with others, to satisfy my sexual curiosity, to have casual/random sex). Third, participant lying rates were also positively correlated with entertainment (ρ = .31, p < .001), which consisted of three items (e.g., to satisfy my social curiosity, to look at pictures of others, to alleviate boredom). These data suggest that discovery phase deceptions are purposeful and associated with certain motivations for using dating apps, including social approval and attention, sexual fulfillment, and browsing. General discussion We present some of the first research to evaluate deception in mobile dating apps during the discovery phase by examining actual message content exchanged between partners. Consistent with prior research (Feldman et al., 2002; Weiss & Feldman, 2006), lies were predominantly driven by self-presentation (e.g., appearing attractive, self-enhancement) and availability management strategies. Appearing attractive and dateable are important for self-presentation, and daters may use deception to appear interesting to others (see Hall, Park, Song, & Cody, 2010). Butler lies to manage availability were also used to guide and control the social interaction for the dater. These lies draw on the constraints of mobile technology to provide misinformation about activities, beliefs, and proximity to the other person. Study 2 revealed that nearly 7% of messages in the discovery phase were untrue. Lying rates, were also associated with specific dating app uses, including social approval and attention, sexual fulfillment, and entertainment by browsing. One possible explanation for this effect is suggested by Corriero and Tong (2016), who measured goals for mobile dating, plus desires for uncertainty and information-seeking in a population of Grindr users. Daters with casual sex goals on Grindr also reported high desires of uncertainty, possibly because it allows people to maintain positive evaluations of the self and the partner. That is, the less you know about a partner and the less you reveal about the self, the more you may idealize a potential date. The distributions of lies in Studies 1 and 2 are also consistent with an emerging trend in deception research: that most people are honest, report low lying rates, and there are only a few prolific liars (Halevy et al., 2014; Serota & Levine, 2015). Furthermore, other data provide evidence in support of deception theories that suggest people often lie for a reason, rather than simply because they can (Buller & Burgoon, 1996; Levine, 2014). That is, lying is purposeful, strategic (e.g., to self-enhance, manage availability), and aligned with interpersonal goals. Why were the number of lies in Study 2 (131) lower than Study 1 (350)? First, we excluded jocular and non-literal messages, which removed statements intended to be funny but not deceptive. Second, and more importantly, participants were required to explain their lies, possibly leading to more judicious and careful ratings of message veracity. We suspect that having participants explain their lies, following procedures in prior work (Birnholtz et al., 2010; French et al., 2015), facilitated a reflective process where they needed to recall the definition of deception and determine if their message constituted an intentional false belief. It is important to note, however, that the percentage of self-presentation lies (Study 1 = 37%, Study 2 = 39%) and availability management lies (Study 1 = 29%, Study 2 = 45%) across both studies are comparable. Therefore, while the overall number of lies in Studies 1 and 2 varied, the pattern of these deception goals is relatively stable. For most participants, deception was used relatively infrequently. Ariely (2012) argues that deception is not rampant because people often have a “fudge factor,” or a level of deception that they feel comfortable with to match their social and psychological goals. In our studies, a small though nontrivial amount of deception was used, and it strategically focused on self-enhancement and managing the impressions of the social relationship. This evidence is consistent with Weiss and Feldman (2006), who observed a small number of lies in their job interview study (e.g., on average, approximately two lies per interview). To obtain the job, people were mostly honest, but used some deception to appear like an attractive candidate. There are also constraints in the online dating infrastructure that limit deception as well. Warranting theory (Walther & Parks, 2002) discusses how artifacts of an offline self that bridge the online world can help online connections gauge truthful identities about others. Warrants, including pictures and friends, hold people accountable for information that they provide online and are common for mobile dating apps because they help to relieve some uncertainty or risk involved with communicating with a stranger. It is worth noting that the three goals associated with lying behavior—social inclusion and attention, casual sex, and entertainment browsing—are relatively short-term goals. That is, since there is no expectation of future relational development, lying is not seen as an impediment to the fulfillment of short-term types of goals. Indeed, lying may actually help daters fulfill those specific goals, while longer-term goals may thwart daters’ propensity to lie or at least get them to consider being more honest, which would be consistent with warranting and with anticipated future face-to-face interaction (Walther & Parks, 2002). The deception consensus effect Consistent with the false consensus effect (H1), in both studies we observed a strong, positive relationship between the number of lies reported by the participant and their perceived number of deceptive messages from the partner (Figure 2). The correlations are consistent with the idea that people are often biased in how they perceive the actions of others (Epley, 2015); the more we lie, the more we believe our partner has lied. This relationship is illustrative because to date, deception research has typically considered how to detect deception, the circumstances for deception, or the language patterns that betray false speech (for a review, see Vrij, 2008), and has been less concerned with how deception frequencies are related to perception biases. We call this pattern the “deception consensus effect,” suggesting that produced and perceived deception rates in a romantic setting are linked. Similar to the truth bias (Levine, 2014), which suggests that people infer interpersonal honesty independent of message veracity, facilitating poor deception detection, we propose that people are biased by their own behavior when considering the false actions of others. If we are honest (or deceptive), we infer that others are honest (or deceptive) as well (see Street & Richardson, 2015). Future work should identify the boundaries of this effect and evaluate if it holds outside of romantic interactions. Since this study did not manipulate the amount of lying by the participant or partner, however, it is important to consider other causal explanations for the deception consensus effect. There are at least two other possible explanations. First, if a person perceives that his or her partner has lied, he or she may feel entitled to lie to the partner; this can explain the strong link between participant and partner lying rates. Second, if there is some suspicion that lying in online dating systems is normative, the reported correlations may reflect social conventions (e.g., believing that deception is commonplace leads to perceiving that others are lying as well) rather than one’s own behavior (see Ellison et al., 2012; Ellison et al., 2006). Therefore, it is unclear from our data if the deception consensus effect is being driven by the participant’s lying behavior or the participants perceptions about other’s behaviors (e.g., the partner is lying, people in general lie in online dating). Future studies should attempt to experimentally tease these possibilities apart. Limitations and future directions The data from both studies are self-reported and retrospective, as participants reviewed sent messages and rated their deceptiveness. Therefore, we did not have direct or indirect access to ground truth and, instead, relied on a participant’s knowledge of a past interaction to develop our understanding of discovery phase lies. Prior work has used this approach reliably (DePaulo, Kirkendol, Kashy, Wyer, & Epstein, 1996; Hancock et al., 2009, among others), but a more precise evaluation of deception in mobile communication may ask participants to rate their messages from an ongoing connection. A separate method may have people download another app to ask for deception ratings immediately after a message is sent. We observed that a number of deceptive messages were unable to be classified into our scheme, but were still considered lies by the participant. Content coding of these messages revealed that they were largely agreements, assents, or dissents, but future work should evaluate the genuine nature of their deceptiveness. Perhaps conducting interviews with participants would be helpful to understand the value of these messages. Similarly, jocular messages were removed from Study 2 because they did not meet our deception criterion, though these data may also serve impression management functions (e.g., to appear funny or witty). Future work should examine how jocular messages contribute to goal pursuit in dating with deception. Finally, the findings in Study 2 are limited to messages from the beginnings and ends of conversations and provide an incomplete picture of the entire discovery phase. We limited our request to these messages to keep our participants’ effort burden reasonable, but future research should attempt to retrieve all messages, including the messages in the middle of a conversation. Conclusion This two-study paper evaluated the frequency of deception in mobile dating conversations during the discovery phase. Consistent with deception theory, we observed that most people were generally honest, but lies were commonly told to advance self-presentation and availability management goals. The number of lies reported by the participant were also correlated with the number of perceived deceptions from the partner. We encourage future work to explore how lying relates to other interpersonal dynamics of mobile dating interactions as the pervasiveness of these relationships continue. Supplementary Material Supplementary material are available at Journal of Communication online. References Ariely , D. ( 2012 ). The (honest) truth about dishonesty: How we lie to everyone—especially ourselves . New York, NY : HarperCollins . Bazarova , N. N. , Taft , J. G. , Choi , Y. H. , & Cosley , D. ( 2013 ). Managing impressions and relationships on Facebook: Self-presentational and relational concerns revealed through the analysis of language style . Journal of Language and Social Psychology , 32 , 121 – 141 . doi:10.1177/0261927X12456384 Google Scholar CrossRef Search ADS Bell , K. L. , & DePaulo , B. M. ( 1996 ). Liking and lying . Basic and Applied Social Psychology , 18 , 243 – 266 . doi:10.1207/s15324834basp1803_1 Google Scholar CrossRef Search ADS Birnholtz , J. , Guillory , J. , Hancock , J. , & Bazarova , N. ( 2010 , February). “on my way”: Deceptive texting and interpersonal awareness narratives. Paper presented at the Proceedings of the ACM conference on Computer-Supported Cooperative Work (CSCW), Savannah, GA. Blackwell , C. , Birnholtz , J. , & Abbott , C. ( 2015 ). Seeing and being seen: Co-situation and impression formation using Grindr, a location-aware gay dating app . New Media & Society , 17 , 1117 – 1136 . doi:10.1177/1461444814521595 Google Scholar CrossRef Search ADS Bok , S. ( 1999 ). Lying: Moral choice in public and private life . New York, NY : Vintage Books . Bond , C. F. , & DePaulo , B. M. ( 2006 ). Accuracy of deception judgments . Personality and Social Psychology Review , 10 , 214 – 234 . doi:10.1207/s15327957pspr1003_2 Google Scholar CrossRef Search ADS PubMed Buller , D. B. , & Burgoon , J. K. ( 1996 ). Interpersonal deception theory . Communication Theory , 6 , 203 – 242 . doi:10.1111/j.1468-2885.1996.tb00127.x Google Scholar CrossRef Search ADS Corriero , E. F. , & Tong , S. T. ( 2016 ). Managing uncertainty in mobile dating applications: Goals, concerns of use, and information seeking in Grindr . Mobile Media & Communication , 4 , 121 – 141 . doi:10.1177/2050157915614872 Google Scholar CrossRef Search ADS DePaulo , B. M. ( 1992 ). Nonverbal behavior and self-presentation . Psychological Bulletin , 111 , 203 – 243 . doi:10.1037/0033-2909.111.2.203 Google Scholar CrossRef Search ADS PubMed DePaulo , B. , Kirkendol , S. , Kashy , D. , Wyer , M. , & Epstein , J. ( 1996 ). Lying in everyday life . Journal of Personality and Social Psychology , 70 , 979 – 995 . doi:10.1037/0022-3514.70.5.979 Google Scholar CrossRef Search ADS PubMed Ellison , N. B. , Hancock , J. T. , & Toma , C. L. ( 2012 ). Profile as promise: A framework for conceptualizing veracity in online dating self-presentations . New Media & Society , 14 , 45 – 62 . doi:10.1177/1461444811410395 Google Scholar CrossRef Search ADS Ellison , N. , Heino , R. , & Gibbs , J. ( 2006 ). Managing impressions online: Self-presentation processes in the online dating environment . Journal of Computer-Mediated Communication , 11 , 415 – 441 . doi:10.1111/j.1083-6101.2006.00020.x Google Scholar CrossRef Search ADS Epley , N. ( 2015 ). Mindwise: Why we misunderstand what others think, believe, feel, and want . New York, NY : Vintage . Feldman , R. S. , Forrest , J. A. , & Happ , B. R. ( 2002 ). Self-presentation and verbal deception: Do self-presenters lie more? Basic and Applied Social Psychology , 24 , 163 – 170 . doi:10.1207/S15324834BASP2402_8 Google Scholar CrossRef Search ADS French , M. , Smith , M. E. , Birnholtz , J. , & Hancock , J. T. ( 2015 , April). Is this how we (all) do it?: Butler lies and ambiguity through a broader lens. Paper presented at the Proceedings of the 33rd Annual ACM Conference on Human Factors in Computing Systems, Seoul, Republic of Korea. Halevy , R. , Shalvi , S. , & Verschuere , B. ( 2014 ). Being honest about dishonesty: Correlating self-reports and actual lying . Human Communication Researchγ , 40 , 54 – 72 . doi:10.1111/hcre.12019 Google Scholar CrossRef Search ADS Hall , J. A. , Park , N. , Song , H. , & Cody , M. J. ( 2010 ). Strategic misrepresentation in online dating: The effects of gender, self-monitoring, and personality traits . Journal of Social and Personal Relationships , 27 , 117 – 135 . doi:10.1177/0265407509349633 Google Scholar CrossRef Search ADS Hancock , J. T. , Birnholtz , J. P. , Bazarova , N. N. , Guillory , J. , Amos , B. , & Perlin , J. ( 2009 , April). Butler lies: Awareness, deception and design. Paper presented at the Proceedings of the ACM Conference on Human Factors in Computing Systems (CHI), Boston, MA. Hancock , J. T. , & Toma , C. L. ( 2009 ). Putting your best face forward: The accuracy of online dating photographs . Journal of Communication , 59 , 367 – 386 . doi:10.1111/j.1460-2466.2009.01420.x Google Scholar CrossRef Search ADS Heino , R. D. , Ellison , N. B. , & Gibbs , J. L. ( 2010 ). Relationshopping: Investigating the market metaphor in online dating . Journal of Social and Personal Relationships , 27 , 427 – 447 . doi:10.1177/0265407510361614 Google Scholar CrossRef Search ADS Kashy , D. A. , & DePaulo , B. M. ( 1996 ). Who lies? Journal of Personality and Social Psychology , 70 , 1037 – 1051 . doi:10.1037/0022-3514.70.5.1037 Google Scholar CrossRef Search ADS PubMed Klein , L. R. ( 1998 ). Evaluating the potential of interactive media through a new lens: Search versus experience goods . Journal of Business Research , 41 , 195 – 203 . doi:10.1016/S0148-2963(97)00062-3 Google Scholar CrossRef Search ADS Levine , T. R. ( 2014 ). Truth-Default Theory (TDT): A theory of human deception and deception detection . Journal of Language and Social Psychology , 33 , 378 – 392 . doi:10.1177/0261927X14535916 Google Scholar CrossRef Search ADS Levine , T. R. , Ali , M. V. , Dean , M. , Abdulla , R. A. , & Garcia-Ruano , K. ( 2016 ). Toward a pan-cultural typology of deception motives . Journal of Intercultural Communication Research , 45 , 1 – 12 . doi:10.1080/17475759.2015.1137079 Google Scholar CrossRef Search ADS Levine , T. R. , & McCornack , S. A. ( 1992 ). Linking love and lies: A formal test of the McCornack and Parks Model of Deception Detection . Journal of Social and Personal Relationships , 9 , 143 – 154 . doi:10.1177/0265407592091008 Google Scholar CrossRef Search ADS Markowitz , D. M. , Hancock , J. T. , & Tong , S. ( 2018 ). Interpersonal dynamics in online dating: Profiles, matching, and discovery. In Z. Papacharissi (Ed.), A networked self and love (pp. 50 – 61 ). New York, NY : Routledge . McCornack , S. A. , & Parks , M. R. ( 1986 ). Deception detection and relationship development: The other side of trust . Annals of the International Communication Association , 9 , 377 – 389 . doi:10.1080/23808985.1986.11678616 Google Scholar CrossRef Search ADS Ranzini , G. , & Lutz , C. ( 2017 ). Love at first swipe? Explaining Tinder self-presentation and motives . Mobile Media & Communication , 5 , 80 – 101 . doi:10.1177/2050157916664559 Google Scholar CrossRef Search ADS Ross , L. , Greene , D. , & House , P. ( 1977 ). The “false consensus effect”: An egocentric bias in social perception and attribution processes . Journal of Experimental Social Psychology , 13 , 279 – 301 . doi:10.1016/0022-1031(77)90049-X Google Scholar CrossRef Search ADS Safronova , V. ( 2017 , December 21). Instagram is now a dating platform, too. Here’s how it works. New York Times. Retrieved from https://www.nytimes.com/2017/12/21/style/instagram-thirst-traps-dating-breakups.html Serota , K. B. , & Levine , T. R. ( 2015 ). A few prolific liars: Variation in the prevalence of lying . Journal of Language and Social Psychology , 34 , 138 – 157 . doi:10.1177/0261927X14528804 Google Scholar CrossRef Search ADS Smith , A. , & Anderson , M. ( 2016 , February 29). 5 facts about online dating. Pew Research Center. Retrieved from http://www.pewresearch.org/fact-tank/2016/02/29/5-facts-about-online-dating/ Street , C. N. H. ( 2015 ). ALIED: Humans as adaptive lie detectors . Journal of Applied Research in Memory and Cognition , 4 , 335 – 343 . doi:10.1016/j.jarmac.2015.06.002 Google Scholar CrossRef Search ADS Street , C. N. H. , & Richardson , D. C. ( 2015 ). Lies, damn lies, and expectations: How base rates inform lie-truth judgments . Applied Cognitive Psychology , 29 , 149 – 155 . doi:10.1002/acp.3085 Google Scholar CrossRef Search ADS Toma , C. L. ( 2015 ). Online dating. In C. Berger , & M. Roloff (Eds.), The international encyclopedia of interpersonal communication (pp. 1 – 5 ). Hoboken, NJ : John Wiley & Sons, Inc . Toma , C. , & Hancock , J. T. ( 2012 ). What lies beneath: The linguistic traces of deception in online dating profiles . Journal of Communication , 62 , 78 – 97 . doi:10.1111/j.1460-2466.2011.01619.x Google Scholar CrossRef Search ADS Toma , C. L. , Hancock , J. T. , & Ellison , N. B. ( 2008 ). Separating fact from fiction: An examination of deceptive self-presentation in online dating profiles . Personality and Social Psychology Bulletin , 34 , 1023 – 1036 . doi:10.1177/0146167208318067 Google Scholar CrossRef Search ADS PubMed Vrij , A. ( 2008 ). Detecting lies and deceit: Pitfalls and opportunities ( 2nd ed. ). West Sussex, England : John Wiley & Sons Ltd . Walczyk , J. J. , Harris , L. L. , Duck , T. K. , & Mulay , D. ( 2014 ). A social-cognitive framework for understanding serious lies: Activation-decision-construction-action theory . New Ideas in Psychology , 34 , 22 – 36 . doi:10.1016/j.newideapsych.2014.03.001 Google Scholar CrossRef Search ADS Walther , J. B. , & Parks , M. R. ( 2002 ). Cues filtered out, cues filtered in: Computer-mediated communication and relationships. In M. L. Knapp , & J. A. Daly (Eds.), Handbook of interpersonal communication ( 3rd ed. , pp. 529 – 563 ). Thousand Oaks, CA : Sage . Weiss , B. , & Feldman , R. S. ( 2006 ). Looking good and lying to do it: Deception as an impression management strategy in job interviews . Journal of Applied Social Psychology , 36 , 1070 – 1086 . doi:10.1111/j.0021-9029.2006.00055.x Google Scholar CrossRef Search ADS van de Wiele , C. , & Tong , S. T. ( 2014 , September). Breaking boundaries: The uses & gratifications of Grindr. Paper presented at the Proceedings of the 2014 ACM International Joint Conference on Pervasive and Ubiquitous Computing, Seattle, WA. © The Author(s) 2018. Published by Oxford University Press on behalf of International Communication Association. 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)
TI - Deception in Mobile Dating Conversations
JF - Journal of Communication
DO - 10.1093/joc/jqy019
DA - 2018-04-26
UR - https://www.deepdyve.com/lp/oxford-university-press/deception-in-mobile-dating-conversations-jdh4tX7UHu
SP - 1
EP - 569
VL - Advance Article
IS - 3
DP - DeepDyve
ER -