Illusory Correlations and Group-Based Expectations: The Influence of Increased Learning

Research has demonstrated that people who embrace different ideological orientations often show differences at the level of basic cognitive processes. For instance, conservatives (vs. liberals) display an automatic selective attention for negative (vs. positive) stimuli, and tend to more easily form illusory correlations between negative information and minority groups. In the present work, we further explored this latter effect by examining whether it only involves the formation of explicit attitudes or it extends to implicit attitudes. To this end, following the typical illusory correlation paradigm, participants were presented with members of two numerically different groups (majority and minority) each performing either a positive or negative behaviour. Negative behaviors were relatively infrequent, and the proportion of positive and negative behaviors within each group was the same. Next, explicit and implicit (i.e., IAT-measured) attitudes were assessed. Results showed that conservatives (vs. liberals) displayed stronger explicit as well as implicit illusory correlations effects, forming more negative attitudes toward the minority (vs. majority) group at both the explicit and implicit level.

In 2 studies, the effects of mood on the formation of distinctiveness-based illusory correlations were examined. After exposure to stimuli inducing positive, neutral, or negative mood, Ss read information about behaviors performed by members of 2 groups in an illusory correlation paradigm. In both experiments, only Ss in a neutral mood formed illusory correlations. In addition, Experiment 2 assessed Ss' processing latencies as a means of investigating differential attention to distinctive behaviors. Only Ss in a neutral mood differentially attended to the minority group's infrequent behaviors. Induced mood apparently interfered with the processing necessary to differentially encode distinctive stimuli, undermining the illusory correlation effect.

The acquisition of a negative evaluation of a fictitious minority social group in spite of the absence of any objective correlation between group membership and negative behaviours was described by Hamilton and Gifford (1976) as an instance of an illusory correlation. We studied the acquisition and attenuation through time of this correlation learning effect. In two experiments we asked for participants' judgements of two fictitious groups using an online version of a group membership belief paradigm. We tested how judgements of the two groups changed as a function of the amount of training they received. Results suggest that the perception of the illusory correlation effect is initially absent, emerges with intermediate amounts of absolute experience, but diminishes and is eliminated with increased experience. This illusory correlation effect can be considered to reflect incomplete learning rather than a bias due to information loss in judgements or distinctiveness.

The voting behavior of racial and ethnic minorities is a topic that attracts much speculation, with some claiming that racial and ethnic minorities do vote for Populist Radical Right Parties (PRRPs) and some claiming they do not. In the European Union, where saving data on individual’s race and ethnicity is prohibited, it is very difficult to contribute to these conversations with real facts. Do ethnic minorities and majorities tend to vote for PRRP and what explains their (lack of) support? Thanks to a novel yet costly sampling method, I surveyed racial/ethnic minority and majority voters in France, Germany and the Netherlands and asked them about their propensity to vote for Rassemblement National (RN) in France, Alternative für Deutschland (AfD) in Germany, and Partij voor de Vrijheid (PVV) in the Netherlands. I compare racial/ethnic minority groups, including Muslims, with majority groups and test the mechanisms that might predict their support for PRRPs. My findings indicate Muslims are among the least likely to vote for PRRPs, though the difference with voters without a migration background is only significant in the Netherlands. When testing what explains the propensity to vote for PRRPs, I find that indicators of in-group favoritism usually explain support to larger extent than out-group hate. Though anti-immigration attitudes predict PRRP voting in all three countries, in-group favoritism explanations explain PRRP voting to a slightly stronger extent. In France and Germany, the ethnocentrism scale predicts voting for RN/AfD more than immigration attitudes do. In the Netherlands, feeling accepted as belonging in the Netherlands explains voting for the PVV the most. Amongst Muslim French, German and Dutch voters, in-group favoritism, or the lack thereof, explains voting for PRRPs as well. French Muslims who feel more attached to France are more likely to vote for RN. German Muslims who do not believe in religious freedom for Muslims are more likely to vote for AfD. This also applies to Dutch Muslims, though immigration attitudes also predict voting for the PVV: the more a Dutch Muslim is against immigration, the more likely they are to vote PVV. Generally, this study makes a case for expanding the standard predictors of PRRP voting towards more indicators of in-group favoritism for the majority in-group, while for Muslims PRRP voting is more driven by policy attitudes. Feeling close or distant towards ethnic in- or out-groups does not predict PRRP voting in any of the cases. These findings contribute to our understanding of PRRP voting in Europe. Keywords: Populism, Muslims, race, ethnicity, voting behavior, France, Germany, Netherlands, RN, AfD, PVV.

This study examined the proposition that blacks and whites make dispositional attributions for an in-group's positive behaviors and an out-group's negative behaviors. The study also examined whether this positive in-group bias was caused by dislike of the out-group or belief in a stereotype. Thus, blacks and whites made attributions to black and white others who succeeded or failed on tasks for which there was either no stereotype or a more negative stereotype of whites than of blacks. An out-group other's failure on both tasks was attributed to lack of ability more than was an in-group other's failure. This finding suggests that the in-group bias is caused by dislike of the out-group. Furthermore, in success conditions subjects' attributions to the in-group or out-group other did not differ. It was suggested that these attributions may result from a combination of an in-group bias and a polarized appraisal.

Group identity is known to exert a powerful socio‐psychological influence on behaviour, but has largely been explored as a uni‐dimensional phenomenon. Theoretically, an increase in the number of identity dimensions considered should make it harder to draw clear boundaries between people. Therefore, we ask whether broadening awareness of identity to multiple dimensions could help alleviate ingroup biases. We explore this in an online experiment, focusing on two different political dimensions of identity. In the control arm, we induced (narrow) awareness of a single identity dimension; in the treatment arm, we induced (broad) awareness of both dimensions. Subsequently, we used a third‐party allocation task to measure ingroup bias on each dimension. Crucially, revealed identity information was held constant across arms during the decision. We found that broader awareness did not decrease ingroup bias and in one setting increased it. We provide supplementary analysis to examine possible mechanisms for our counter‐intuitive findings.

We compared implicit and explicit ingroup bias across four minority groups who ranged in status from high (Jews and Asians) to medium (overweight people) to low (poor people). Minorities relatively high in status showed more implicit ingroup bias than minorities relatively low in status. In fact, overweight and poor people showed automatic preference for the dominant outgroup (i.e., own group devaluation). The relationship between ingroup bias and perceived status was reliably positive whether based on category membership (r = .54) or minority members' own perceptions of their status (r = .36), but only at the implicit level. By contrast, explicit attitudes were negligibly correlated with status differences. In addition, dominant group members showed stronger possession of implicit ingroup bias than did minorities, but particularly as their relative status increased. In concert, the findings support system justification theory (Jost & Banaji, 1994) by showing that minorities with the least status are the most susceptible to automatic ingroup devaluation. They also suggest that motives to legitimize the status quo are predominantly nonconscious, for minorities and dominants alike.

Categorization is not enough: Intergroup discrimination in negative outcome allocation

An association between biased impression updating and relationship facilitation: A behavioral and fMRI investigation

IntroductionThis meta-analysis investigated (1) whether ethnic minority and majority members have a neural inter-group bias toward each other, and (2) whether various ethnic groups (i.e., White, Black, and Asian) are processed in the brain differently by the other respective ethnicities.MethodsA systematic coordinate-based meta-analysis of functional magnetic resonance imaging (fMRI) studies was conducted using Web of Science, PubMed, and PsycINFO (altogether 50 datasets, n = 1211, 50.1% female).ResultsWe found that ethnic minority members did not show any signs of neural inter-group bias (e.g., no majority-group derogation). Ethnic majority members, in turn, expressed biased responses toward minority (vs. majority) members in frontal, parietal, temporal, and occipital regions that are known to be involved in e.g., facial processing, attention, and perspective-taking. We also found differences in neural response patterns toward different ethnic groups (White, Black, and Asian); broadest biases in neural response patterns were evident toward Black individuals (in non-Black individuals). Heterogeneity was mostly minor or low.Discussion:Overall, the findings increase understanding of neural processes involved in ethnicity perception and cognition as well as ethnic prejudices and discrimination. This meta-analysis provides explanations for previous behavioral reports on ethnic discrimination toward minority groups.

Minority Affirmation in Intergroup Relations: Ingroup Favoritism as Self-Affirmation

The present research investigates how emotional displays shape reactions to ingroup and outgroup members when people are reminded of death. We hypothesized that under mortality salience, emotions that signal social distance promote worldview defense (i.e., increased ingroup favoritism and outgroup derogation), whereas emotions that signal affiliation promote affiliation need (i.e., reduced ingroup favoritism and outgroup derogation). In three studies, participants viewed emotional displays of ingroup and/or outgroup members after a mortality salience or control manipulation. Results revealed that under mortality salience, anger increased ingroup favoritism and outgroup derogation (Study 1), enhanced perceived overlap with the ingroup (Study 3), and increased positive facial behavior to ingroup displays-measured via the Facial Action Coding System (Studies 1 and 2) and electromyography of the zygomaticus major muscle (Study 3). In contrast, happiness decreased ingroup favoritism and outgroup derogation (Study 2), and increased positive facial behavior towards outgroup members (Study 3). The findings suggest that, in times of threat, emotional displays can determine whether people move away from unfamiliar others or try to form as many friendly relations as possible.

The aim of this study was to explore the mediating effects of adult attachment and marital satisfaction on the relationship between childhood maternal attachment and parenting behavior. A total of 372 mothers of preschoolers completed questionnaires on childhood maternal attachment, adult attachment, marital satisfaction, and parenting behavior. Adult attachment was analyzed by distinguishing between attachment anxiety and attachment avoidance. Parenting behavior was categorized into positive and negative behaviors. For the main analyses, Preacher and Hayes’s PROCESS macro program was used to examine serial mediating effects. The results revealed that attachment anxiety mediated the effects of mothers’ childhood attachment experiences on both positive and negative parenting behavior. However, attachment avoidance only mediated the effects on positive parenting behavior. Marital satisfaction mediated the effects of mothers’ childhood attachment experiences on both positive and negative parenting behavior. Notably, the sequential mediating effects of attachment anxiety and marital satisfaction were not significant for either positive or negative parenting behavior. By contrast, the sequential mediating effects of attachment avoidance and marital satisfaction were significant for both positive and negative parenting behavior. These findings elucidate the predictive factors for parenting behavior within a process model framework, providing valuable insights for parental education and counseling aimed at enhancing mothers' parenting practices.

Previous research into intergroup attribution has addressed mainly the behavior of groups to which members are ascribed (e.g. gender, race). The attribution processes of groups of which membership is achieved (e.g. friendship groups) is less well understood, and the current study sought to address this. Fifty-five undergraduate participants were asked to explain the positive and negative behavior of a member of the in-group and a member of the out-group. As predicted, the participants attributed an in-group member's positive behavior more, and their negative behavior less, to internal, global, and specific causes than they did the corresponding behavior of an out-group member. There was also evidence that the participants employed a strategy of out-group derogation in their attributions: they made a higher intemality rating for an out-group member's negative behavior than they did for that person's positive behavior. It is proposed that the current study's use of achieved groups maximized participants' levels of group identification, and that this in turn motivated behavioral strategies aimed at protecting that identity.

Trust is essential for social interactions, cooperation and social order. Research has shown that social status and common group memberships are important determinants of receiving and reciprocating trust. However, social status and group membership can coincide or diverge–with potentially different effects. Our study contributes to the existing literature on the role of status and group membership by testing two separate trust-generating mechanisms against each other. We examine if individuals tend to place trust in high-status groups (irrespective of their own group membership) or, rather, if they tend to trust others with whom they share a common group membership. We assume that status group membership is signalled by cultural (musical) taste. This operationalization follows the theoretical reasoning of Bourdieu who argues that it is, above all, musical taste that classifies persons of different status. By demonstrating their “legitimate” cultural taste, upper-class members distinguish themselves from the middle and lower classes and signal their social status, thereby creating awe, respect and an air of trustworthiness. We report evidence from online experiments with incentivized trust games, which enable us to separate the two trust-generating mechanisms. We find no evidence that persons with “legitimate” tastes are generally trusted more. Instead, our results clearly demonstrate ingroup favouritism towards persons with a similar taste. Participants place more trust in members of their own group and expect them to be more trustworthy. In other words: members of taste-based groups trust each other more than members of different-taste-based groups. Interestingly, this group-based trust is not always justified inasmuch as received trust is not necessarily reciprocated more strongly by own group members. This suggests that ingroup favouritism is, at least in part, driven by false beliefs.