A Meta-Analysis of Functional Magnetic Resonance Imaging Studies on In-Group and Out-Group Categorization

Systematic variations in neural activation patterns during processing of group categorization is a facet often overlooked in extant meta-analyses that typically treated all group memberships and task types as homogenous entities. Recognizing the necessity to differentiate between various group memberships and task types due to them eliciting distinct neural activity, our study addressed this gap by conducting a more fine-grained exploration.

A comprehensive meta-analytic approach using the activation likelihood estimation (ALE) following PRISMA guidelines was employed. Inclusion criteria were task-fMRI studies, whole-brain analyses in a standard stereotaxic space, involved healthy adults, and published in English. Meta-analyses were performed using GingerALE.

Overall in-group > out-group (IG>OG) processing included 66 studies (108 contrasts, 518 foci), and ALE analysis revealed 22 significant clusters, with the largest (3328 mm³) located in the left insula, inferior frontal gyrus, and uncus. Overall OG > IG processing included 35 studies (61 contrasts, 371 foci), and ALE analysis revealed 25 significant clusters, with the largest (1728 mm³) located in the right presupplementary motor area. Group membership classification found 38 ethnicity studies on in-group bias (IG>OG; 59 contrasts, 279 foci), and ALE analysis revealed 27 significant clusters, with the largest (1616 mm³) located in the right superior occipital gyrus, right middle occipital gyrus, right middle temporal gyrus, and right fusiform gyrus. fMRI task types found 30 studies in empathy processing on in-group bias (IG>OG; 50 contrasts, 260 foci), and ALE analysis revealed 23 significant clusters, with the largest (2480 mm³) located in the right postcentral gyrus, and right posterior cingulate gyrus.

Our unrestricted search approach with ALE methodology extends beyond previous categorically constrained meta-analyses by capturing group dynamics not limited to predefined categories. The distinct activation patterns for ethnicity-based categorization and empathy processing tasks indicate multiple neural pathways rather than universal mechanisms for group processing. These context-dependent patterns suggest interventions must target specific group-task combinations rather than assume uniform neural processes.