The role of meta-cognitive certainty on the relationship between identity fusion and endorsement of extreme pro-group behavior

ABSTRACT

Identity fusion is a powerful feeling of connectedness to one’s group. The current research explores whether measuring certainty in identity fusion improves its ability to predict extreme pro-group outcomes. Across three studies, participants reported their level of identity fusion with their country and their certainty in responses to the scale (predictor variables). Responses to a trolley dilemma) (Studies 1 and 3) and willingness to fight and die for one’s group (Studies 1 and 2) were the dependent measures. As expected, certainty moderated the effects of identity fusion on self-sacrifice, with greater consistency between them obtained for those with high (vs. low) certainty. In conclusion, taking certainty into consideration can be useful to predict the association between identity fusion and self-sacrifice.

KEYWORDS:

• Meta-cognition
• certainty
• confidence
• identity fusion
• self-sacrifice
• individual differences

Disclosure statement

No potential conflict of interest was reported by the authors.

Correction Statement

This article has been republished with minor changes. These changes do not impact the academic content of the article. .

Notes

1. The databases and/or materials used in this research are available upon request. We report all measures, manipulations, and exclusions in these three studies. Also, we report all studies conducted in this line of research.

2. To test whether the two-way interaction between identity fusion and certainty was statistically different between responses to the willingness to fight and die scale and the continuous measure of the trolley dilemma, we ran a new linear regression analysis. Two additional variables were created in this new test. The first variable was a within-subjects factor called “Dependent variable” and resulted from the combination of willingness to fight and die and continuous measure of the trolley dilemma (both were standardized first). This means each participant had two rows: one with their response to the willingness to fight and die scale, and another one with the continuous measure of the trolley dilemma. The second variable was a between-subjects factor called “Type of measure” and had two possible values (−1 = fight and die measure, 1 = continuous trolley dilemma measure). Our linear regression analysis was then run on fight and die and continuous trolley dilemma (the “Dependent variable”), with Type of measure (effect coding: −1 = fight and die measure, 1 = continuous trolley dilemma measure), Identity fusion (centred), Certainty (centred) and their interactions as our independent variables. Results show that the two-way interaction reported for both DV´s is not significantly moderated by type of measure, B = −.007, t(587) = −0.27, p = .786. This analysis strategy was also followed to compare a single-item measure of certainty with a three-item measure of certainty within Study 3 (endnote 3) and to compare the continuous and the dichotomous responses to the trolley dilemma in Study 3 (endnote 4).

3. For interested readers, we also compare the ability of the single item to produce moderation compared to the ability of the new three-item certainty measure. To test whether the two-way interaction between identity fusion and certainty was significantly different when certainty was composed by one vs. three items in the new study, we treated certainty measure (one vs. three items) as a within-subjects variable in the linear regression analysis. We ran similar regressions with the continuous and the dichotomous responses to the trolley dilemma. Results show that the two-way interaction reported for both DV´s is not significantly moderated by type of measure. Specifically, the comparison in the new study showed that the difference in effect size on the continuous response to the dilemma when using a three-item measure (B = .236, t (958) = 4.079, p < .001) compared to using just a single-item measure (B = .141, t (958) = 3.38, p < .001) did not differ significantly (B = .047, t (958) = 1.33, p = .183). Similarly, the comparison in the new study showed that the difference in effect size on the dichotomous response to the dilemma when using a three-item measure (B = .298, z = 2.81, p = .005) compared to using just a single-item measure (B = .203, z = 2.57, p = .010) did not differ significantly (B = −.047, z = −0.72, p = .472).

4. After standardizing both measures of the dilemma, we included type of measure as a within-subjects factor. It did not significantly moderate the two-way interaction between identity fusion and confidence found for both measures of the dilemma, B = −.019, t(958) = −0.67, p = .502. This suggests that, even though the pattern of results may show slight differences between dependent variables, both measures of the dilemma yielded similar results.

5. A finding one might expect is that the correlation between willingness to fight and die and trolley dilemma responses was higher for people who are certain than for people who are not. However, we did not find this effect. In this particular case, confidence refers to the metacognitive certainty about people’s responses to the identity fusion questionnaire and not certainty about their willingness to fight and die. Changing the construct for which people are certain about is consequential in this case.

Additional information

Funding

This work was supported by the Ministerio de Ciencia e Innovación [PSI2015-67754-P,PSI2017-83303-C2-1-P,RTI2018-093550-B-I00];