Facebook Therapy? Why Do People Share Self-Relevant Content Online? Eva Buechel. University of Miami. Jonah Berger. University of Pennsylvania

Transcription

1 Facebook Therapy 1 Facebook Therapy? Why Do People Share Self-Relevant Content Online? Eva Buechel University of Miami Jonah Berger University of Pennsylvania Word Count: 1000 Contact Information: Jonah Berger, University of Pennsylvania, 3730 Walnut St. Philadelphia, PA 19104; jberger@wharton.upenn.edu. Electronic copy available at:

2 Facebook Therapy 2 Abstract Why do people use online social networking sites (e.g., Facebook) and what are the implications of this behavior for well-being? While emotionally unstable individuals experience emotions more intensely, they are less adept at regulation. Consequently, we suggest they may rely on others more to help them deal with their emotions. Further, given they tend to be socially apprehensive they may be particularly likely to rely on online expression because it is less threatening. Consistent with this perspective, we find that emotional unstable individuals are more likely to post self-relevant information online and write about their emotions when doing so a tendency not observed offline. Further, such emotional writing, paired with the potential to receive social support helps them repair well-being after negative experiences. These results shed light on a motivator for, and benefit of, online social networking, while also demonstrating how the social sharing of emotion can boost well-being. Electronic copy available at:

3 Facebook Therapy 3 The internet is pervasive, yet its impact on well-being is not well understood. Researchers (Kraut, et al., 1998) and cultural critics (Yoffe, 2009) have argued that it reduces face-to-face interaction, leaving people depressed and lonely. Further, sites based on social interaction (i.e. Facebook) are thought to merely be havens for people with poor selfimage and narcissists demanding the world s attention, (DiSalvo, 2010, p. 53; Buffardi & Campbell, 2008). In contrast, we suggest that certain online behaviors (i.e., sharing self-relevant content) may actually be beneficial, aiding emotion regulation. People often post short messages about their thoughts, feelings, or actions (e.g., tweets or Facebook status updates). Though one might argue these updates are driven by vanity, or extraverts maintaining existing social ties (Gosling, et al., 2011), we suggest that such behaviors may also have therapeutic value, boosting emotionally unstable individuals well-being after negative emotional experiences. Emotionally unstable individuals experience emotions more intensely (Barr, Kahn, & Schneider, 2008) and negatively (Costa & McCrae, 1980), and are less adept at cognitive regulation (Gross & John, 2003). Consequently, they may be more likely to rely on others to help them deal with their emotions. Further, given they tend to be socially apprehensive (Luminet, et al. 2000) they may be particularly likely to rely on online expression because it is less threatening (i.e., non-face-to-face setting and general nature of posting). But while research has documented long-term cognitive benefits of expressive writing (Pennebaker, 1997; 1999), the social benefits of sharing have not been systematically isolated (Frattaroli, 2006). Given that sharing emotion encourages social bonding (Peters & Kashima, 2007), however, it may buffer negative emotions in the short-term by encouraging emotional support (Rimé, 2007; 2009). We test this possibility. In particular, we suggest that emotionally

4 Facebook Therapy 4 unstable individuals are more likely to rely on, and benefit from, sharing self-relevant content with others. We examine whether emotionally unstable individuals (1) post more status updates and share their emotions when doing so and (2) whether this type of sharing boosts their wellbeing after negative emotional experiences by increasing perceived social support. Study 1: Frequency of Sharing Participants (N = 81) were either asked (1) how often they update their Facebook status and share feelings and emotions with others through such updates or (2) with others in person. They also completed a Big Five Personality Inventory (Gosling, Rentfrow, & Swann, 2003). Results and Discussion Multiple regression examined how Big Five personality traits related to status update frequency and emotion expression, both online and offline. Emotionally unstable participants reported updating their status more frequently, = -.50, t(40) = , p <.05. Further, while they did not report sharing feeling and emotions more with others in person, = -.08, p >.60, emotionally unstable people did report sharing feeling and emotions more than others through online updates, =.47, t(40) = 2.49, p <.05 (correlations significantly different, p <.05). 1 Study 2: Impact of Sharing Participants (N = 174) completed two ostensibly unrelated studies. First, everyone had a negative affective experience, getting false feedback on an anagram performance task (Forgas, 1 Additional data show that emotional unstable people s status updates are rated by observers as containing more emotion expression.

5 Facebook Therapy ). Participants also provided a close other s address to ensure that close others were similarly activated for across conditions. Next, they completed a writing study. Some participants wrote about a control topic (office products). The other three conditions wrote about their current emotions, either (1) in private while thinking about the close other whose they provided, or (2) to be shared with the close other who they were told would not be able to respond or (3) who might respond. This allowed us to test whether merely thinking about or sharing with a close other is sufficient, or whether potential response is necessary to boost well-being. Finally, participants reported their current well-being, perceived social support (Metzler, 2003), and emotional stability (McCrae & Costa, 2004). Results and Discussion In addition to a main effect of emotional stability on well-being, = -.62, t(167) = -4.16, p <.001, regression analysis revealed a significant emotional stability x condition interaction for the response possible and control condition, =.26, t(167) = 2.32, p =.02, see Figure 1. The other two conditions did not differ from control, ps >.1. Slope analysis (Aiken & West, 1991) revealed that for low emotionally stable individuals (-1SD), writing only increased well-being when it was to be shared with a close other who might respond, =.30, t(167) = 2.54, p =.01 (other conditions ps >.4). There was no corresponding effect of condition among highly emotionally stable individuals (+1SD, p >.18). Looked at another way, while there was a negative correlation between emotional stability and well-being in the other conditions (r s < -.28, p s <.06), writing with potential response helped reduce the well-being gap (r = -.18, p >.20).

6 Facebook Therapy 6 Finally, consistent with our theorizing, this boosted well-being from sharing with potential response was mediated by perceived social support (Supplementary Materials). General Discussion These results provide insight into a motivator for, and benefit of, online social networking. Emotional unstable individuals are more likely to post self-relevant information and write about their emotions when doing so. Further, such emotional writing, paired with the potential to receive social support as on social networking sites helps them repair well-being after negative experiences. Importantly, these benefits come not merely from writing (which can provide long-term cognitive benefits), writing about emotions (i.e. venting), or sharing emotion. Further, they were not driven by the content written: There was no condition x emotional stability interaction on the valence, length, or emotionality of what people wrote, and most participants did not even write about the negative feedback (see Supplementary Materials). Instead, the notion that a close other would read what they had written and potentially respond boosted well-being by increasing perceived social support. The findings increase understanding of online behavior and have important implications for emotion regulation and clinical psychology.

7 Facebook Therapy 7 Figure 1 Writing to a close other with the ability to receive a response increases well-being among low emotionally stable individuals. 50 HighEmotionalStability LowEmotionalStabiliity 40 Well-being Control OnlyWriting WritingToOther WritingtoOther+ PotentialResponse

8 Facebook Therapy 8 References Aiken, L., & West, S. G. (1991). Multiple regression: Testing and interpreting interactions. Newbury Park, CA: Sage. Barr, L., Kahn, J., & Schneider, J. (2008). Individual differences in emotion expression: Hierarchical structure and relations with psychological distress. Journal of Social and Clinical Psychology, 27 (10), Buffardi, L.E., & Campbell, W. K. (2008). Narcissism and social networking sites. Personality and Social Psychology Bulletin, 34, Costa, P. T., & McCrae, R. R. (1980). Influence of extraversion and neuroticism on subjective well-being: Happy and unhappy people. Journal of Personality and Social Psychology, 23, DiSalvo, D. ( 2010). Are Social Networks Messing With Your Head? Scientific American Mind, 20 (7), Forgas, J. F. (1991). Affective Influences on Partner Choice: Role of Mood in Social Decisions. Journal of Personality and Social Psychology, 61 (5), Frattaroli, J. (1996). Experimental disclosure and its moderators: A meta-analysis. Psychological Bulletin, 132 (6), Gosling, S., Augustine, A. A., Vazire, S., Holtzman, N. S., Gaddis, S. (forthcoming). Manifestations of personality in online social networks: Self-reported Facebook-related behaviors and observable profile information. Cyberpsychology, Behavior, and Social Networking. Gosling, S. D., Rentfrow, P. J., & Swann, W. B., Jr. (2003). A very brief measure of the Big-Five personality domains. Journal of Research in Personality, 37, Gross, J. J., & John, O. P. (2003). Individual differences in two emotion regulation processes: Implications for affect, relationships, and well-being. Journal of Personality and Social Psychology, 85, Kraut, R., Lundmark, V., Patterson, M., Kiesler, S., Mukopadhyay, T., & Scherlis, W. (1998). Internet paradox: A social technology that reduces social involvement and psychological well-being? American Psychologist, 53(9), Luminet, O., Zech, E., Rime, B., & Wagner, H. (2000). Predicting cognitive and social consequences of emotional episodes: The contribution of emotional intensity, the five factor model, and alexithymia. Journal of Research in Personality, 34, McCrae, R. R., & Costa, P. T. (2004). A contemplated revision of the NEO Five-Factor Inventory. Personality and Individual Differences, 36 (3), Meltzer, H. (2003). Development of a common instrument for mental health. In: Nosikov & Gudex (eds). EUROHIS: Developing Common Instruments for Health Surveys. Amsterdam: IOS Press. Pennebaker, J. W. (1997). Writing about emotional experiences as therapeutic process. Psychological Science, 8 (3), Pennebaker, J.W. (1999). The effects of traumatic disclosure on physical and mental health: The values of writing and talking about upsetting events. International Journal of Emergency Mental Health, 1, Peters, P., & Kashima, Y. (2007). From social talk to social action: shaping the social triad with emotion sharing. Journal of Personality and Social Psychology, 93,

9 Facebook Therapy 9 Rimé, B. (2007). The social sharing of emotion as an interface between individual and collective processes in the construction of emotional climates. Journal of Social Issues, 63, Rimé, B. (2009). Emotion elicits the social sharing of emotion: Theory and empirical review. Emotion Review, 1 (1), Yoffe, E. (2009, August 12th). Seeking: How the brain hard-wires us to love Google, Twitter, and texting. And why that's dangerous. Slate. Retrieved from

10 Facebook Therapy 10 Supplementary Materials Study 1 Participants either reported (1) how often they update their Facebook status (1 = Multiple Times A Day; 6 = Never) and how often they share feelings and emotions with others online through status updates (1= Not at all, 7 = Very much) or (2) how often they share feelings and emotions with others in person (1= Not at all, 7 = Very much). The Ten Item Personality Inventory (Gosling, Rentfrow & Swann, 2003) measured the Big Five personality traits (Extraversion, Emotional Stability, Conscientiousness, Agreeableness, and Openness to Experience) with two items for each dimension. Study 2 Negative affect was induced by telling participants that they had performed below average on a verbal ability test. They were given five minutes to solve 33 multiple-choice anagrams. They were then given their actual score (6-26), but told that the average performance was correct answers. Well-being was measured using participants current feeling (1= Bad, 100 = Good), happiness (1= Sad, 100 = Happy), and relaxation state (1= Tense, 100 = Relaxed), adapted from Williams, Cheung and Choi (2000). Perceived social support was measured using the Oslo Social Support Scale (Metzler, 2003), which asked participants about their number of close confidants, (1 = None, 4= 6+), perceived sense of concern or interest from other people (1 = A lot, 5 = None), and the ease with which they could get help if needed (1 = Very Easy, 5 = Very Difficult). Emotional stability was measured with the 12-item neuroticism subscale from the NEO Five Factory Inventory (McCrae & Costa, 2004). There was no effect of condition on emotional stability, p >.1. Mediation Analysis (Preacher and Hayes, 2004) revealed that the interaction between writing manipulation and neuroticism for the control condition and the response possible condition predicted well-being after the writing task, b = 8.12, t(86) = 2.17, p =.03, and perceived social support, b =.15, t(86) = 2.29, p =.02. When the perceived social support was included in the model, the interaction no longer predicted well-being after the writing task, b = 6.20, t(86) = 1.64, p >.1, but the perceived social support did, b = 12.41, t(86) = 2.10, p =.04. The mediating effect of social support was significantly different from zero (b = 1.91) and the 95% confidence interval (Lower =.04 and Upper = 5.87) did not contain zero, supporting mediation. The well-being effects cannot be explained by differences in content. Participants expressions were coded based on valence (-3 = Very Negative, 3 = Very Positive), length (1 = Very Short, 7 = Very Long), and number of emotional words (1 = None, 7 = Very Many), but there was no condition x emotional stability interaction on any of these measures (Fs < 1.3, ps >.3). Further, most participants (86%) did not even mention the negative feedback manipulation, and this percentage did not differ based on the condition x emotional stability interaction, b > , Wald-Statistic < 2, p >.2.