The inutility of studying fictional-coin giving for understanding the relationship between attachment and prosocial development

Abstract

Liu, Wang, and Van Lange (2022) have published a study that “replicated earlier findings that prosociality was associated with a greater level of secure attachment and a lower level of avoidant attachment” and that “prosociality was positively related to the number of siblings, especially sisters, as discovered by Study 3 of the original study”. These earlier findings refer to a study by Van Lange et al. (1997, which we have tried (but failed) to replicate in an earlier publication (IJzerman & Denissen, 2018). In a new paper, Liu et al. (2022) replicated the design another time, using an even larger sample (total N ranging from 3,789 - 4,090), and found a significant effect for the fictional coin-giving correlation with attachment security, with effect sizes ranging from r = .03 to r = .07. However, given measurement problems that Liu et al. (2022) reported in combination with these small effect sizes, we think that it is misleading to use such effect sizes to support the proposed theoretical models, as a correlation of .07 translates into half a percent of explained variance. We conclude that the fictional coin-giving measure is not the right paradigm to understand the developmental origins of prosociality.