In modern times, there is an increasing tendency to use "us versus them" rhetoric in the political realm. Motivated reasoning literature offers a solution as to how group conflict contributes to outgroup bias. People may develop negative feelings towards outgroups in order to reduce cognitive dissonance and feel better about their judgments. Nevertheless, this is not the situation for everyone. My research aims to understand how peoples’ predisposed tendencies may affect the level of their out-group bias. Drawing from the distinction between shame (a negative sense of identity) and guilt (a response to a specific behavior) in psychology, I hypothesized that self-conscious emotions would moderate the relationship between information about in-group transgressions and out-group bias. To test this, I conducted a survey experiment with three different groups: Men/Women, White/Black people, and Democrats/Republicans. Participants were randomly assigned to watch videos depicting misbehavior from their respective group. Results showed that the proneness to self-conscious emotions did moderate the relationship to some extent, although the connection between shame and guilt proneness was more complex than anticipated. Criticizing one's in-group generally caused people to experience cognitive dissonance and reinforced out-group bias, particularly among those who were highly prone to both shame and guilt. The three-way interaction between treatment, shame proneness, and guilt proneness varied across and within the different identity categories, suggesting that there is no single theory that can entirely explain the degree of out-group bias. / Political Science

Individer har en systematisk tendens att favorisera medlemmar i ens egen grupp och diskriminera de som tillhör en annan grupp, trots minimala grunder för gruppindelning. Denna studie syftar till att undersöka om social interaktion mellan individer i grupper påverkar denna särbehandlingstendens. I sådana fall, kan detta komma att yttra sig genom bland annat skillnader i kunskapsutbyte och spridning av information mellan de anställda på en arbetsplats. Grupper delades in slumpmässigt och särbehandlingen mättes via poängfördelning. Datainsamlingen skedde via både experiment (n = 41) och enkät (n = 135) samt med tre olika nivåer av social interaktion. Undersökningsdeltagarna fördelade mer poäng till sin egen grupp, än till den andra gruppen. Däremot påverkade varken grad av social interaktion eller typ av metod poängfördelningen. Vid samtliga tillfällen var det mellan en tredjedel till strax över hälften av undersökningsdeltagarna som särbehandlade. Slutsatsen är att även vid en slumpmässig gruppindelning så kommer en särbehandlingstendens att ske.