A Window to What We Missed: Effects of Self- versus Other-Exclusion on Social Media Users’ Fundamental Needs, Emotional Responses, and Online Coping Behaviors

ABSTRACT

The present paper investigates the psychological and behavioral responses postulated within the Temporal Need-Threat Model of a specific form of social exclusion: Not being included in other users’ social media posts. Across two experiments (N _Total = 563), we distinguished two reasons for this aversive experience: (a) Voluntarily choosing not to engage in the activity being displayed in the post (i.e. self-exclusion); (b) not being invited to this activity (i.e. other-exclusion). Compared to the inclusion condition, being excluded was – for both reasons – associated with threatened needs and impaired emotional responses. As hypothesized, these detrimental effects were more pronounced for other- than for self-exclusion. Concerning coping behavior, we found that self-excluded users more frequently behaved in a prosocial way, whereas those who were excluded by others performed more withdrawal behavior. Depending on self- versus other-exclusion, different coping behaviors were more versus less effective in restoring needs and regulating emotions.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data described in this article are openly available in the Open Science Framework at https://osf.io/ypq9a/ and https://osf.io/396gc.

Open Scholarship

This article has earned the Center for Open Science badges for Open Data, Open Materials and Preregistered. The data and materials are openly accessible at https://osf.io/ypq9a/ and https://osf.io/396gc.

Notes

1. Study 1 also involved independent (i.e., information relevance) and dependent variables going beyond the scope of the present manuscript. Here, we only focus on variables relevant for the outlined hypotheses/RQs.

2. The initial purpose of assessing baseline emotions was to include these pre-experimental scores in our analysis. However, because pre-experimental scores did not significantly vary by condition (see section “Checking the baseline experimental assumption” in OSF file “S1_markdown”) we did not include them as control variables.

3. Going beyond the scope of the present manuscript, there was also a “not relevant” condition in which the caption said: “Thinking back to my last birthday! We had such a good time at the beach/bar! Can’t believe this was 10 months ago!” Participants assigned to this condition were excluded from analysis. Interested readers can find further information and the results in OSF file “S1_results information relevance”

4. For exploratory purposes, we also used an implicit word-fragment completion task measuring the positive and negative valence of participants’ emotions. Due to space limitations and in line with the pre-registered main analyses, we only report direct measures here.

5. Note that we originally pre-registered to run multiple regressions. However, we found ANCOVAs more intuitive for understanding our results. We also pre-registered a second approach to analyze coping effectiveness: Computing difference scores for need threat and discrete emotions (i.e., t₂ – t₁) and comparing participants in the self-exclusion condition with those in the other-exclusion condition. Due to space limitations, the results are summarized in OSF file “S2_Results difference score.”

6. Due to the low frequencies of antisocial behavior after inclusion (n = 1) and self-exclusion (n = 2), we did not calculate post-hoc-tests for antisocial behavior.