DECEPTION ACROSS DIFFERENT MODES OF COMMUNICATION Monica Whitty; Tom Buchanan; Adam Joinson; & Alex Meredith

Truth/Lies Paradox Greater self-disclosure Potential for greater deception

What is deception? Bok (1989) “any intentionally deceptive message which is stated. Such statements are most often made verbally or in writing, but can of course also be conveyed via smoke signals, Morse code, sign language, and the like” (p.13).

Digital deception “...the intentional control of information in a technologically mediated message to create a false belief in the receiver of the message” (Hancock, 2007; p. 290).

Identity deception Alex and Joan Nowheremom The case of Alex and Joan is often cited as a ‘classic’ case of identity deception. In this case Alex created a persona (called ‘Joan’) who became a confidant of many of the women on a discussion board, some of whom Alex had sexual relations with. Joan avoided FtF meetings by disclosing that she was disabled.   Of course, Joan was Alex, which caused considerable outrage in the community. A number of other cases are reasonably well documented. For instance, ‘Nowheremom’ (a female created by a male online community member, who he subsequently dated and then killed off in a car accident).

Whitty (2002) Item Men Women 17-20yrs 21-55yrs Lied age 63 60 66 53 Lied gender 28 18 23 22 Lied occupation 56 42 50 47 Lied education 40 25 31 35 Lied income 44 37 Only respondents who use chat rooms were invited to complete the survey. In total, 320 surveys were returned, with 160 women and 160 men completing the survey. The sample ranged from 17 to 55 years with a mean age of 21.3 years (SD=6.13). They were all Australian residents.

Message based deception

Detecting Online Deception Zhou et al. (2004) Hancock et al. (2005) According to research by Zhou et al. (2004) liars use more words, and these words are more informal and expressive, compared to people telling the truth. They also made more typographic errors.   This finding was replicated by Hancock, Curry, Goorha, and Woodworth (2005), who also found increased word count during deception in Instant Messaging. Interestingly, the people being lied to asked more questions than those being told the truth, perhaps suggesting that they knew, even subconsciously, that they were being lied to.

Predicting where people lie Social Distance Theory Media Richness Theory Feature Based Theory

FtF Phone IM Email Media Features Synchronous X Recordless Distributed Lying predications Feature-based 2 1 3 Media-Richness 4 Social Distance

Questions? What about the target? The type of lie? What about other online spaces? Spontaneous versus planned?

Whitty & Carville (2008) Self-serving Other oriented Target of the lie

You are having a face-to-face conversation with someone that you are ‘close to’ when they invite you to an event. You can think of something else you would rather spend your time doing so you tell them that you can’t make it to the event, even though you can.

You receive an email from a person you do not know well You receive an email from a person you do not know well. Within the email they ask you if you think they look attractive. You do not think that they are attractive but you do not want to hurt their feelings so you email them back and tell them that they are attractive.

Type of Lie Target FtF Phone Email Self-serving: Close 3 2 1 Self-serving: Not well-known Other-oriented: Close Other-oriented: Not well-known Lying Predictions Feature Based Media Richness Social Distance

Type of Lie Target FtF Phone Email Self-serving: Close 3 2 1 Self-serving: Not well-known Other-oriented: Close Other-oriented: Not well-known Lying Predictions Feature Based Media Richness Social Distance

Research Questions RQ1: Which mode of communication are individuals more likely to tell lies? RQ2: Which mode of communication are individuals more likely to tell planned lies? RQ3: Does the proportion of self and other directed lies differ across media?

Modes of Communication Face-to-face Telephone Social networking sites Instant Messaging E-mail Text messaging

Participants 100 undergraduate students 76 participants completed their diaries in full 18 to 32 years, mean age of 19.45 (SD= 1.836) Women (68; 89%) vs Men (8; 11%)

Pillai’s Trace = .653, F (5,39) = 14.654, p <.0005). H1: More lies will be told by the telephone than FTF and IM, followed by SNS, email and SMS. Pillai’s Trace = .653, F (5,39) = 14.654, p <.0005). Phone > FtF IM > SMS text messaging   To control for individual differences in media use patterns and honesty, each of the lie index variables was then treated as one observation in a one-way repeated-measures ANOVA. The assumption of sphericity assumption was violated, so multivariate tests Lies were told significantly more often in some media than others. Post-hoc comparisons (Least Significant Difference tests) indicated that people are most likely to lie by telephone, with FtF coming in second. Both are significantly more likely to be used for lying than the other media, where there are no significant differences other than people being less likely to lie by SMS than by IM. Thus, H1 is only partially supported. However, very few lies were actually told using email, instant messaging or social network sites, which makes interpretation difficult.

FtF (M=3.50 SD=2.30) > SMS (M=4.44, SD=2.11) (p=.007) H2: Planned lies will be more likely to occur in asynchronous media (e.g., email, SMS, SNS) than in synchronous media (e.g., ftf, phone, IM). This proportion will differ across media. (F(2,541)=5.11, p=.006) FtF (M=3.50 SD=2.30) > SMS (M=4.44, SD=2.11) (p=.007) To test this hypothesis, level of planning for lies was compared across media using a one-way ANOVA. The media examined were restricted to FTF (329 lies), telephone (147 lies) and SMS (68 lies) due to the low number of lies in other media. There was a significant effect of medium (F(2,541)=5.11, p=.006). Post-hoc tests (Tukey HSD) indicated that FtF (M=3.50 SD=2.30) and SMS (M=4.44, SD=2.11) differed significantly (p=.007). Telephone, which had a mean score between the two (M=3.90, SD=2.47) did not differ significantly from either. This result is partially consistent with the hypothesis, that lies told in the asynchronous medium SMS) had a higher level of planning than one of the synchronous media (FtF). It is worth noting that the levels of planning were relatively low in all media (planning was rated on a 1-9 scale anchored at “completely spontaneous” and “carefully planned”)  

H3: The proportion of self and other directed lies will differ across media. (chi-square(6, n=543) = 8.50, p=.20) No significance

Perceived portability FtF Phone Digital technologies SMS Media Features Synchronous x Recordless Distributed Asynchronous Perceived portability Spontaneous lies 2 1 3 Planned lies The results from this study reveal more about which theoretical model best predicts lying across different modes of communication. The ‘Feature based model’ still seems to be the preferred model over the ‘social distance theory’ and ‘media richness theory’. However, we would like to suggest that the feature based model needs to acknowledge IM as a ‘near synchronous’ mode of communication rather than synchronous, which then changes what this model would predict. It may well be that other features are important to consider; however, our study found little difference between lies told using digital technologies to suggest any other features with regards to lies overall. In addition to this change to the model we suggest that the model ought to distinguish between spontaneous and planned lies