Denial,Attention-Seeking,and Posting Online While Intoxicated: Three Key Predictors of Collegiate Sadfishing

Abstract

Sadfishing, or the exaggeration of one's emotional state online to generate sympathy, is a maladaptive behavior that can negatively affect mental health. A better understanding of the characteristics of individuals who sadfish could inform tailored interventions to decrease sadfishing and improve quality of life. However, to date, the phenomenon of sadfishing remains understudied. Thus, the current project was designed to identify some of the key psychological and behavioral characteristics that may be associated with sadfishing. Undergraduate college students (N = 374) recruited from introductory psychology courses at a large, Hispanic-serving institution completed an anonymous online survey assessing sadfishing and other online behaviors, psychological characteristics (coping, stress, resilience, and social support), and alcohol use. Both univariate and multivariate statistical analyses were conducted. Results of the binary logistic regression analysis found that students who reported using denial as a coping strategy (p = 0.005), who endorsed the attention-seeking behaviors associated with histrionic personality disorder (p = 0.021), and who used social media while intoxicated (p = 0.017) were most likely to report sadfishing. This study furthers our knowledge of the maladaptive online behavior of sadfishing and identifies several key predictors that could become targets for tailored interventions. In particular, our results highlight the importance of coping skills training for individuals who sadfish.

Introduction

The term “sadfishing” refers to exaggerating one's emotions online to gain sympathy and attention.¹ While it may attract support, this exposure also leaves sadfish vulnerable to predation^2,3 and may adversely impact their mental health.^4,5 A better understand of sadfishing could illuminate intervention strategies to decrease this behavior.

Milestones such as learning to self-soothe, comfort oneself, and maintain resilience during stress are integral to adult well-being.⁶ Adults who cannot self-soothe may depend on external feedback for validation.^6,7 Thus, sadfishing may be an external, tension-reducing behavior that is used instead of self-soothing to cope with stress, garner rewarding feedback, and enhance resilience. This aligns with findings that people use social media for coping, and that positive reinforcement mechanisms moderate the relationship between stress and problematic social media use.⁸ Therefore, coping strategies, resilience, and perceived stress are all potential sadfishing predictors.

Sadfishing might be part of a larger, ingrained pattern of behavior seen in histrionic personality disorder (HPD) and borderline personality disorder (BPD). Lengel⁹ describes manipulative exaggeration, elevated emotional lability, and attention-seeking as HPD traits. Furthermore, the pervasive interpersonal difficulties common in BPD can also be seen in affected individuals' social media behavior, even at a subclinical level.¹⁰ Individuals with HPD and/or BPD features may struggle to self-soothe and may use external validation to buoy their self-esteem.^11,12 Their maladaptive behaviors may be more likely during stress.¹³ Treatment often includes increasing distress tolerance, emotional regulation, and interpersonal effectiveness.^11,12 Clearly, symptoms of HPD and BPD are relevant to sadfishing.

Social media is itself behaviorally addictive.⁸ Also, for those who seek social status online, intensely curating their online platform through posts (potentially including sadfishing) may increase danger for other risk behaviors.¹⁴ Thus, we included overall social media use as a potential predictor of sadfishing. Furthermore, the association between alcohol and social media use is well-known,¹⁵ and even low levels of drinking may disrupt inhibitory control and increase risky behavior.¹⁶ Disinhibition also tends to increase intimate disclosure online.¹⁷ Therefore, drinking might increase sadfishing.

We predicted that, compared with individuals who do not sadfish, the sadfish would exhibit (H1) less effective coping, (H2) lower resilience, (H3) higher perceived stress, (H4) similar perceived social support, higher BPD (H5) and HPD (H6) characteristics, (H7) greater social media use, and (H8) higher alcohol consumption. Although social support was not a significant predictor in our previous work,¹⁸ we have included it here for further validation, yet we have predicted similar levels for both groups. Finally, we asked, “Of these theoretically relevant predictors, which will be most significantly associated with sadfishing?”

Materials and Methods

This study was approved by the University's Institutional Review Board, and research was performed in accordance with the ethical standards of the Declaration of Helsinki. Students provided informed consent before participation.

Participants and procedures

Students were recruited from the Introduction to Psychology course (a general core course option for all undergraduate students) at a large Hispanic-serving institution in the Southcentral United States using the SONA human subject pool and were compensated with class credit. They completed an anonymous, online, cross-sectional survey administered via Qualtrics. Participants were grouped as sadfish (n = 67) or non-sadfish (n = 307) based on their yes/no responses to “Have you ever exaggerated your emotional state online in order to get sympathy?”

Questionnaires

Demographics

Demographic questions assessed gender, age, ethnicity, and race.

Coping

The Brief COPE (α = 0.89) has 28 statements and a four-point scale.¹⁹ Subscales include self-distraction, active coping, denial, substance use, emotional support, instrumental support, behavioral disengagement, venting, positive reframing, planning, humor, acceptance, religion, and self-blame. Higher scores indicate greater strategy use. The subscales have acceptable internal consistency.¹⁹

Resilience

The Adolescent Resilience Questionnaire (Short Form; α = 0.77) has 49 statements and a five-point scale.²⁰ It has good internal consistency.^20,21 The total resilience score was positive (confidence, emotional insight, social skills, empathy/tolerance) minus negative subscales (negative cognition). Higher scores indicate greater resilience.

Stress

The Perceived Stress Scale (α = 0.83) assesses past-month feelings using 10 items and a five-point scale.²² Higher scores indicate greater stress. It has adequate reliability and is a better predictor than life-event scores.^22,23

Social support

The Multidimensional Scale of Perceived Social Support (α = 0.94) measures support from family, friends, and significant others using 12 items and a seven-point scale.^24,25 Higher scores indicate greater support. It has good internal reliability.²⁵

Personality

The Brief Histrionic Personality Scale (α = 0.88), based on the DSM-5,²⁶ has 11 statements and a four-point scale. There are two subscales: seductiveness and attention seeking. Higher scores indicate greater HPD traits. It has acceptable internal consistency.²⁶

The BPD (α = 0.91) items from the larger Personality Belief Questionnaire consisted of 14 statements on a four-point scale.²⁷ Higher scores indicate greater BPD traits. It has moderately high internal consistency.²⁷

Social media use

Participants reported their favorite social media platform and average past-week social media use (hours per day, any platform). Participants indicated on a four-point scale how often they used social media while intoxicated.²⁸ Higher scores indicate greater endorsement.

Alcohol use

Using standard drink sizes for alcoholic beverages, participants reported their average number of drinks per typical week and any lifetime binge drinking.²⁹

Data analysis

Data were examined for outliers, skewness, and kurtosis. Drinks per week was square root transformed. Univariate comparisons assessed differences in demographics and other characteristics between sadfish and non-sadfish (Hypotheses H1–H8). To do this, chi-square tests of independence compared categorical variables, and independent t tests compared continuous variables. Pairwise deletion was used for missing data. To test the final research question, a binary logistic regression model was developed to determine the key characteristics associated with sadfishing. Listwise deletion accounted for missing data. Only variables significant at the univariable level were included in the model. Due to multicollinearity, humor, seductiveness, and BPD were excluded. To determine significance, two-tailed tests with an alpha level = 0.05 were used.

A post hoc power analysis was conducted based on an independent t test, with alpha = 0.05 and a small-to-moderate effect size (d = 0.4), which exhibited sufficient obtained power (1-β) = 0.84. All analyses were conducted using SPSS version 27 (IBM Corp, Armonk, NY).

Results

Table 1 shows demographics and psychological and behavioral characteristics.

Table 1.

Participant Characteristics

	Sadfish (n = 67)	Non-Sadfish (n = 307)	Significance, p
Demographics
Age	19.4 (1.1)	19.5 (1.3)	ns
Percent female	63 (42)	76 (234)	0.02
Percent Hispanic	40 (27)	42 (128)	ns
Percent white	73 (49)	71 (218)	ns
Coping strategies
Self-distraction	5.52 (1.39)	5.45 (1.34)	ns
Active coping	5.47 (1.20)	5.36 (1.36)	ns
Denial	4.36 (1.82)	3.63 (1.52)	0.005
Substance use	4.33 (1.89)	3.70 (1.86)	0.02
Use of emotional support	5.35 (1.55)	5.19 (1.61)	ns
Use of instrumental support	5.53 (1.57)	5.11 (1.68)	ns
Behavioral disengagement	4.21 (1.39)	4.04 (1.44)	ns
Venting	4.95 (1.58)	4.67 (1.43)	ns
Positive reframing	5.39 (1.41)	5.30 (1.61)	ns
Planning	5.50 (1.66)	5.46 (1.57)	ns
Humor	6.23 (1.71)	5.58 (1.72)	0.01
Acceptance	5.79 (1.48)	5.65 (1.42)	ns
Religion	4.87 (1.95)	4.56 (1.95)	ns
Self-blame	5.75 (1.48)	5.60 (1.58)	ns
Resilience	41.61 (12.20)	42.79 (11.99)	ns
Perceived stress	21.92 (5.69)	21.58 (6.40)	ns
Perceived social support	61.30 (13.64)	62.46 (15.27)	ns
Personality
Seductiveness	14.10 (3.32)	12.20 (3.31)	<0.001
Attention-seeking	12.00 (3.38)	10.36 (3.13)	<0.001
Borderline	32.18 (8.54)	29.02 (7.72)	0.004
Social media
Hours per day	5.5 (2.5)	4.9 (2.9)	ns
While intoxicated	3.39 (1.51)	2.61 (1.32)	<0.001
Alcohol consumption
Drinks per week^a	5.9 (10.0)	3.0 (5.2)	0.001
Binge drinking	73 (49)	54 (165)	0.004

Mean (standard deviation) or percent (n).

Drinks per week (means and standard deviations) are presented here in their raw form, however, transformed data were used for statistical analyses.

n.s., not significant.

Demographics

Average age was 19.4 years (SD = 1.3, range = 18–23). Groups did not differ on age (p = 0.63). Seventy-four percent of participants were female and 26 percent were male. Gender distribution differed between groups (χ²(1) = 5.21, p = 0.02). Sixty-three percent of sadfish were female, while 76 percent of non-sadfish were female. Forty-one percent of participants were Hispanic. Groups did not differ on Hispanic ethnicity (p = 0.82). Three percent of participants were American Indian or Alaska Native, 3 percent were Asian, 8 percent were black or African American, 71 percent were white, 13 percent were multiracial, and 1 percent preferred not to answer. Groups did not differ on race (p = 0.91).

Psychological and behavioral characteristics

Sadfish (vs. non-sadfish) used more denial (t(351) = 2.92, p = 0.005), substance use (t(350) = 3.00, p = 0.02), and humor (t(343) = 2.58, p = 0.01) for coping. There were no other coping differences (ps > 0.07). Groups did not differ on resilience (p = 0.48), perceived stress (p = 0.70), or perceived social support (p = 0.58). Sadfish had more HPD (seductiveness: t(351) = 4.11, p < 0.001; attention-seeking: t(353) = 3.71, p < 0.001) and BPD (t(353) = 2.87, p = 0.004) features. Groups did not differ on their favorite platforms (p = 0.29) or hours per day spent using social media (p = 0.15).

Snapchat was ranked #1 by 36 percent of participants, followed by Instagram (28 percent), TikTok (21 percent), and Other or Could not decide (15 percent). Sadfish reported more use of social media while intoxicated (t(369) = 4.25, p < 0.001). Sadfish consumed more drinks per week (t(343) = 3.25, p < 0.001) and reported a greater history of binge drinking (ꭓ²(1) = 8.15, p = 0.004).

Multivariate analyses

Table 2 shows the correlation matrix of variables considered for the multivariate binary logistic regression model. The overall model was significant, χ²(6) = 30.865, p < 0.001, with a -2LL = 259.630 and Nagelkerke R² = 0.154 (Table 3). Denial was significantly associated with sadfishing. For each unit increase in the frequency of denial, individuals are 35.6 percent more likely to sadfish (p = 0.005). Attention-seeking was also significantly related to sadfishing. For each unit increase in attention-seeking, individuals are 12.0 percent more likely to sadfish (p = 0.021). Lastly, using social media while intoxicated was significantly associated with sadfishing. For each unit increase in the frequency of social media intoxication, there is a 33.4 percent increase in the likelihood of sadfishing (p = 0.017).

Table 2.

Pearson Correlation Matrix for Predictors of Sadfishing

	Denial	Substance use	Humor	Seductiveness	Attention-seeking	Borderline	SM while intoxicated	Drinks per week
Coping
Denial	1	0.484^a	0.213^a	0.288^a	0.154^a	0.367^a	0.146^a	0.100
Substance use		1	0.154^a	0.286^a	0.173^a	0.270^a	0.346^a	0.280^a
Humor			1	0.212^a	0.173^a	0.165^a	0.193^a	−0.067
Personality
Seductiveness				1	0.677^a	0.271^a	0.292^a	0.261^a
Attention-seeking					1	0.026	0.204^a	0.221^a
Borderline						1	0.103	0.143^a
Social media
While intoxicated							1	0.361^a
Alcohol use
Drinks per week								1

Correlation significant at the p = 0.01 level. SM, social media.

Table 3.

Binary Logistic Regression—Factors Associated with Sadfishing

	B	SE	Wald	Sig	OR	95 percent CI lower	95 percent CI upper
Male gender	0.262	0.356	0.544	0.461	1.300	0.647	2.610
Denial—coping	0.305	0.108	7.917	0.005	1.356	1.097	1.677
Substance use—coping	−0.092	0.103	0.791	0.374	0.912	0.745	1.117
Attention seeking	0.113	0.049	5.313	0.021	1.120	1.017	1.232
Social media while intoxicated	0.288	0.121	5.646	0.017	1.334	1.052	1.692
Drinks per week (transformed)	0.124	0.129	0.923	0.337	1.131	0.879	1.456
Constant	−4.826	0.761	40.233	0.000	0.008

B, beta; CI, confidence interval; OR, odds ratio; SE, standard error; Sig, p-value; Wald, chi-square value.

Discussion

Ours is among the first studies of the predictors of sadfishing. The results largely supported our hypotheses. Compared with non-sadfish, sadfish engaged in less effective coping strategies (H1), including denial and substance use, although they also used humor. Sadfish had more BPD (H4) and HPD (H5) features, such as seductiveness and attention-seeking. Sadfish reported greater alcohol use (H7), including more drinks per week, use of social media while intoxicated, and binge drinking history. Perceived social support (H8) did not differ between groups, which replicates our previous work.¹⁸ Some of our hypotheses were not supported; for example, groups did not differ on resilience (H2), perceived stress (H3), or overall social media use (H6). Finally, by using a multiple regression model, our study moves the literature beyond simple group comparisons on individual variables and identifies factors that are most significantly related to sadfishing: denial, attention-seeking, and using social media while intoxicated.

Coping

Sadfish coped by using denial, substance use, and humor. While the negative effects of substance use are well documented, and humor is often considered positive, the role of denial may be more nuanced. Historically, denial has been considered maladaptive.³⁰ When faced with an unacceptable threat or consequence, deniers cope by refusing to consciously acknowledge the significance of their circumstances.³¹ However, there are situations, such as during a traumatic event, when denial offers some degree of psychological protection and/or relief from stress until assistance becomes available.^31,32 Exploration of the effectiveness of different coping strategies is beyond our scope, however, denial might be an important clue to identify sadfish.

Personality

Multicollinearity excluded several personality features from the regression analysis. Ultimately, attention-seeking emerged as a key predictor, which aligns with the sadfishing definition of exaggerating negative emotional states online for attention.^1,8 Social media users who place a high value on this recognition may be more willing to engage in increasingly risky behavior to achieve it.³³ Thus, sadfishing might escalate over time, an issue that has not previously been investigated. Interventions to improve self-soothing and coping for individuals with these personality features might decrease sadfishing.

Social media while intoxicated

Alcohol is disinhibitory and may increase intimate disclosures.^16,17 Sadfish reported riskier drinking, and using social media while intoxicated significantly predicted sadfishing. Social media use may encourage a culture of intoxication and normalize heavy drinking through peer exchanges and algorithmically mediated interaction with alcohol advertisements.^34–36 Qualitative studies of intoxication online reveal a sense of “drunken regret” linked to an intoxication-related loss of control over self-presentation.³⁴ For sadfishing, this aspect of online intoxication might increase negative consequences. This is a topic for future studies.

Limitations and future directions

This self-report study is potentially limited by social desirability bias, and the results from our collegiate sample have imperfect generalizability. Furthermore, while the lack of a validated sadfishing instrument is a limitation, our results may guide the future development of a multifaceted questionnaire highlighting the nuances of sadfishing behaviors. Other suggestions for future research include objective longitudinal assessments, ecological momentary assessments, and a broader participant sample. Self-identified motivations for sadfishing and participants' general self-awareness of their motivations are other potential future topics.

Conclusions

Our results point to three targets for intervention in sadfishing: (a) developing healthier coping strategies, (b) managing unhealthy personality characteristics, and (c) changing social media habits by either avoiding social media while intoxicated or avoiding alcohol while using social media. Overall, sadfish would benefit from cultivating emotional regulation, stress tolerance, and self-soothing. This might lessen their dependence on external validation from social media.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

Funding Information

No funding was received for this study.

References

Coughlan

‘Sadfishing’ social media warning from school heads. BBC News. 2019. Available from: https://www.bbc.com/news/education-49883030 [Last accessed: June 1, 2023].

Mahmood

‘Sadfishing’ is playing into hands of pedophiles and bullies. Metro. 2019. Available from: https://metro.co.uk/2019/10/01/sadfishing-playing-hands-paedophiles-bullies-10838350/ [Last accessed: June 1, 2023].

Weale

Young people who seek support online being accused of ‘sadfishing’. The Guardian. 2019. Available from: https://www.theguardian.com/society/2019/oct/01/young-people-seeking-support-online-accused-sadfishing-bullying-report [Last accessed: June 1, 2023].

Malik

Cysterstalking and self-harm behavior among college students: Depression as mediator. EEO, 2021; 20(4):3078–3084.

Varghese

, Pistole

. College student cyberbullying: Self-esteem depression, loneliness, and attachment. J Coll Couns, 2017; 20:7–21.

Gallop

Failure of the capacity for self-soothing in women who have a history of abuse and self-harm. J Am Psychiatr Nurses Assoc, 2022; 8:20–26.

Winnicot

The capacity to be alone. In: The Maturational Processes and the Facilitating Environment. ( Winnicot

ed.) Hogarth Press: London; 1965; pp. 29–36.

Moretta

, Buodo

, Santucci

, et al. Problematic social media use is statistically predicted by using social media for coping motives and by positive reinforcement processes in individuals with high COVID-19-related stress levels. J Psychiatr Res, 2023; 158:104–113.

Lengel

GJ.

Histrionic personality disorder. In: The Wiley Encyclopedia of Personality and Individual Differences. ( Carducci

, Nave

, Mio

, Riggio

RE.

eds.) Wiley: Hoboken, NJ, USA; 2020; pp. 187–191.

10.

Ooi

, Michael

, Lemola

, et al. Interpersonal functioning in Borderline Personality Disorder traits: A social media perspective. Sci Rep, 2020; 10:1068.

11.

French

, Shrestha

Histrionic Personality Disorder. StatPearls Publishing: Tampa, FL, USA; 2021.

12.

Van der Kolk

, Hostetler

, Herron

, et al. Trauma and the development of borderline personality disorder. Psychiatr Clin North Am, 1994; 17(40):715–730.

13.

Cramer

Personality, personality disorders, and defense mechanisms. J Pers, 1999; 67(3):535–554.

14.

Vannucci

, Simpson

, Gagnon

, et al. Social media use and risky behaviors in adolescents: A meta-analysis. J Adolesc, 2020; 79:258–274.

15.

Alhabash

, Park

, Smith

, et al. Social media use and alcohol consumption: A 10-year systematic review. Int J Environ Res Public Health. 2022; 19(18):11796.

16.

Weafer

, Fillmore

. Low-dose alcohol effects on measures in inhibitory control, delay discounting, and risk-taking. Curr Addict Rep, 2016; 3:75–84.

17.

Joinson

AN.

Disinhibition and the internet. In: Psychology and the Internet, Second Edition. ( Gackenbach

ed.) Academic Press: Cambridge, MA, USA; 2007; pp. 75–92.

18.

Petrofes

, Howard

, Mayberry

, et al. Sad-fishing: Understanding a maladaptive social media behavior in college students. J Am Coll Health, 2022; doi: 10.1080/07448481.2022.2132110

19.

Carver

CS.

You want to measure coping but your protocol's too long: Consider the Brief COPE. Int J Behav Med, 1997; 4:92–100.

20.

Anderson

, Killian

, Hughes

, et al. The adolescent resilience questionnaire: Validation of a shortened version in U.S. youths. Front Psychol, 2020; 11:1–8.

21.

Gartland

, Bond

, Olsson

, et al. Development of a multi-dimensional measure of resilience in adolescents: The Adolescent Resilience Questionnaire. BMC Med Res Methodol, 2011; 11(134):1–10.

22.

Cohen

, Kamarck

, Mermelstein

. A global measure of perceived stress. J Health Social Behav, 1983; 24:386–396.

23.

Cohen

, Williamson

Perceived stress in a probability sample of the United States. In: The Social Psychology of Health. ( Spacapan

, Oskamp

eds.) Sage: Newbury Park, CA, USA; 1988; pp. 31–67.

24.

Zimet

, Dahlem

, Zimet

, et al. The multidimensional scale of perceived social support. J Pers Assess, 1988; 52:30–41.

25.

Dahlem

, Zimet

, Walker

. The multidimensional scale of perceived social support: A confirmation study. J Clin Psych, 1991; 47(6):756–761.

26.

Ferguson

, Negy

. Development of a brief screening questionnaire for histrionic personality symptoms. Pers Indiv Diff, 2014; 66:124–127.

27.

Butler

, Brown

, Beck

, et al. (2002). Assessment of dysfunctional beliefs in borderline personality disorder. Behav Res Ther, 2002; 40:1231–1240.

28.

Ceballos

, Howard

, Dailey

, et al. (2018). Collegiate binge drinking and social media use among Hispanics and Non-Hispanics. J Stud Alcohol Drugs, 2018:79(6):868–875.

29.

Cranford

, McCabe

, Boyd

. A new measure of binge drinking: Prevalence and correlates in a probability sample of undergraduates. Alcohol Clin Exp Res, 2006; 30:1896–1905.

30.

Marcus

ER.

Modern ego psychology. J Am Psychoanal Assoc, 1999; 47:843–871.

31.

Fang

, Albarqouni

, von Eisenhart Rothe

, et al. Is denial a maladaptive coping mechanism which prolongs pre-hospital delay in patients with ST-segment elevation myocardial infarction?. J Psychosom Res, 2016; 91:68–74.

32.

Esteve

, Valdes

, Reisco

, et al. Denial mechanisms in myocardial infarction: Their relations with psychological variables and short-term outcome. J Psychosom Res, 1992; 36:491–496.

33.

Shabahang

, Ghaemi

, Aruguete

, et al. Measuring gratification from and consequences of likes. J Media Psychol, 2022; 34(6):395–400.

34.

Barnes

, McCreanor

, Goodwin

, et al. Alcohol and social media: Drinking and drunkenness while online. Crit Public Health, 2016; 26(1):62–76.

35.

Measham

, Brain

. ‘Binge’ drinking, British alcohol policy and the new culture of intoxication. Crime Media Cult, 2005; 1:262–283.

36.

McCreanor

, Lyons

, Griffin

, et al. Youth drinking cultures, social networking and alcohol marketing: Implications for public health. Crit Public Health, 2013; 23:110–120.