Advances in Group Processes: Volume 39

Role-taking is a basic social process underpinning much of the structural social psychology paradigm – a paradigm built on empirical studies of human interaction. Yet today, our social worlds are occupied by bots, voice assistants, decision aids, and other machinic entities collectively referred to as artificial intelligence (AI). The integration of AI into daily life presents both challenges and opportunities for social psychologists. Through a vignette study, the authors investigate role-taking and gender in human-AI relations.

Participants read a first-person narrative attributed to either a human or AI, with varied gender presentation based on a feminine or masculine first name. Participants then infer the narrator's thoughts and feelings and report on their own emotions, producing indicators of cognitive and affective role-taking. The authors supplement results with qualitative analysis from two open-ended survey questions.

Participants score higher on role-taking measures when the narrator is human versus AI. However, gender dynamics differ between human and AI conditions. When the text is attributed to a human, masculinized narrators elicit stronger role-taking responses than their feminized counterparts, and women participants score higher on role-taking measures than men. This aligns with prior research on gender, status, and role-taking variation. When the text is attributed to an AI, results deviate from established findings and in some cases, reverse.

This first study of human-AI role-taking tests the scope of key theoretical tenets and sets a foundation for addressing group processes in a newly emergent form.

The purpose is to theorize and empirically estimate the impact of the gendered nature of the offender-victim dyad and crime type on time to arrest.

Predictions regarding the impact of gendered offender-victim dyads and crime type on time to arrest are constructed by extending role congruity theory and tested using data from the FBI's National Incident-Based Reporting System across five crime types using dyadic-based event history methods.

The authors find strong empirical support that role expectations derived from the gender composition of offender-victim dyads and the masculinity of the crime type affect time to clearance.

This research is the first to theorize and empirically test the relative impact of role congruency and the relational nature of the offender-victim dyad in the adjudication process. Furthermore, the research shows that the construction of “normal crime” can be enhanced by applying a gendered and relational approach, based on social psychological theory, which is predictive of crime clearance.

Future research is required to validate the results for crimes where law enforcement has less discretion and are feminine typed.

The results imply that by accounting for the expectations generated by gender roles when applied to offender-victim dyads a casual mechanism is established that better organizes previously inconsistent results with respect to the impact of gender on time to clearance. Thus, the authors' utilization of role congruity theory of gender provides a more consistent explanation for inequalities in time to clearance that may be fruitful for evaluating other steps in the adjudication process.

The authors examine how different exchange patterns affect structurally disadvantaged actors' interactional justice evaluations and group identification in situations characterized by reciprocal and negotiated exchange.

Laboratory experiment.

Although results replicate prior work finding that disadvantaged individuals view their exchange partners as less fair when exchanging via negotiation rather than reciprocation, they also show the value of considering the pattern of exchange. Indeed, both the form of exchange and the pattern of exchange prompt exchange behaviors that shape how disadvantaged actors view the exchange experience, such that much of the direct effect of the form of exchange is offset by indirect paths, especially when the disadvantaged actor remains committed to their more advantaged partner. These fairness evaluations matter because as the authors show, they affect perceptions of group identification.

Future work should more explicitly consider how emotions as well as different levels of inequality might modify the processes described.

This chapter highlights the need to consider both the form of exchange and the relative stability of exchange when considering the fairness perceptions and group identification of disadvantaged individuals.

To use a behavioral measure of legitimacy to study how differences in negotiating style and status affect the legitimacy of persons in high-power network positions. Predictions include (1) that powerful network actors who negotiate using a pro-group style will maintain legitimacy better than will those who negotiate selfishly and (2) those higher in status will be granted more legitimacy both before and after exchange than powerful actors lower in status.

An experimental study in which participants were connected in networks to powerful partners who were portrayed as consistently high or low on several status characteristics. Both before and after exchange, participants evaluated partners on a number of dimensions and made decisions on whether to vote to join a coalition to take the partner's power away, a direct behavioral indicator of legitimacy.

High-power partners lost legitimacy over the course of exchange irrespective of whether they negotiated in pro-group or selfish ways, and irrespective of whether they were high or low in status. This effect was pronounced for partners who negotiated selfishly. Although partner status predicted legitimacy prior to exchange, legitimacy evaluations after exchange appeared entirely driven by the partner's negotiating style (how the power was used) and not by status.

The project introduces a new behavioral measure of legitimacy that correlated highly with self-report items and should be of value in future research. The study also indicates promising directions for future research that might disentangle effects of power and status on legitimacy, along with adjudicating among explanations for why this study did not find status effects on legitimacy.

The purpose of this chapter is to contextualize mixed findings in the self-complexity literature. Self-complexity (when individuals' identity meanings do not overlap) theoretically buffers negative outcomes associated with distressing experiences, like identity non-verification (perceptions that others see individuals differently than individuals see themselves). However, research on self-complexity overlooks the social context in which these processes occur. Here, the author argues that multiplexity (when individuals are connected in two or more role relationships) provides meaningful context that influences whether self-complexity functions as a buffer. This leads to two competing arguments: (1) multiplexity enhances the self-complexity buffering effect and (2) multiplexity undermines the self-complexity buffering effect.

Using egocentric network survey data from 314 US adults, the author ran four regression models (two for the friend identity, two for the coworker identity). These models test whether self-complexity moderates the effect of non-verification on distress when individuals do (or do not) have multiplex ties in which they are close friends with their coworkers.

Multiplex ties amplify the buffering effect of self-complexity when non-verification occurs in the coworker identity, but multiplex ties weaken the buffering effect of self-complexity when non-verification occurs in the friend identity.

This work supports the idea that the benefits associated with self-complexity are conditional on social context as well as identity context. Future research should consider these realities when examining how identity and buffering processes relate to wellbeing.

This work integrates social network analysis and identity theory to reconcile mixed findings in the literature on self-complexity.

Second-order expectations refer to an actor's beliefs about what a co-actor believes with respect to their relative abilities on a shared task. The authors describe and compare three alternative programs of research that explain the effects of second-order expectations on behavioral inequalities in task groups. The authors’ overall goal is to work toward improving the precision and generality of theories of second-order expectations.

The authors conduct a thorough review of theory and research on each of the three alternative models of second-order expectations. In so doing, they highlight areas of convergence and divergence in terms of theory, method, and empirical support. They also suggest research designs that can help clarify the effects of second-order expectations in task groups and adjudicate among the models.

New empirical studies are needed that attempt to replicate findings across the three approaches to modeling second-order expectations. In addition, the three approaches need to be directly compared at the same time using a shared experimental design and the same participant population.

This is the first effort to systematically and critically compare and contrast three competing models of second-order expectations in structural social psychology. The authors offer a number of original, specific recommendations for future research.

The authors review experimental procedures employed in studies of status and organization processes in task groups. Early studies employed face-to-face interaction in groups in both naturalistic and laboratory settings. The creation of an abstract interaction experiment that varied initial conditions in place for actors increased validity in theory tests. Recent studies have employed face-to-face groups to increase the range of situations to improve both internal and external validity. The authors include a brief history of how experiments have increased the range of explanatory application. The authors illustrate questions related to external validity with examples from research the authors and others have conducted.

Properties of experimental design are described. The authors employ aspects of the standard experiment in the expectation states research program to highlight designs used in studying task groups. The authors discuss features of face-to-face discussion studies designed to test aspects of expectation states theories. Errors in execution of data collection and principles of good design are explicated.

The authors identify procedures that may affect the quality of data in open interaction situations. Means to reduce errors and improve data quality are identified. The authors list issues that may affect data quality and suggest several ways to improve design and reporting of face-to-face studies.

The authors' effort to create a design road map for face-to-face studies will improve the quality of data collection and analysis. This will broaden the scope of research situations and improve the range of external validity of studies of status and organization of task groups.

This chapter seeks to investigate the ways individualistic versus collectivistic values moderate neural responses to social exclusion among African American and White respondents. The author hypothesized that the vmPFC – a key brain region for emotion regulation – would correspond to collectivistic value moderation and the dlPFC – the cognitive control center of the brain – would be associated with individualistic value moderation.

This study used a virtual ball tossing game (Cyberball), where 17 African American and 11 White participants were excluded or included with ball tosses, while inside an fMRI scanner. Before the start of each round the participants were primed with individualism, collectivism or a comparison condition.

Results showed that (1) African Americans showed stronger neural responses to exclusion and (2) offered support for the hypothesis that the dlPFC showed greater activation in African Americans (compared to Whites) when they were primed with individualism values during exclusion. There was no support for the collectivism hypothesis.

Research limitations included a relatively small sample size (N = 28), a comparison of only two racial groups and that the partners in the game were virtual (pre-programmed by the experimenter).

This research offers an empirical framework for sociologists seeking to apply social theories into neurological studies.

Identifying effective coping strategies for historically oppressed racial groups.

The chapter is original for demonstrating the moderating effects of values on neural responses to exclusion for the first time and by offering a novel neurosociological framework.