Aim of the project
It may seem that group fragmentation and polarization are only ever detrimental to epistemic cooperation, and that an open exchange of information among group members is always beneficial to reaching the group’s epistemic goals. However, conceptual and computational analysis reveals that the development of adequate opinions may sometimes require group members to temporarily withhold information from each other (e.g., Zollman 2007). Along similar lines, we might think that peer pressure is only ever detrimental to group opinions: crowd wisdom obtains when crowd members make up their mind independently. But there are numerous examples of group members who keep each other in check through peer pressure. In short, there may be a certain rationality to peer pressure and group fragmentation.
The aim of this project is to establish how the social mechanisms that drive peer pressure and group fragmentation impact on the rationality of information exchange. We combine conceptual work on opinion formation processes in groups, as studied in social epistemology, with a social psychological, and hence empirical, understanding of these influencing mechanisms. We study under what conditions information sharing or withholding, based on such mechanisms, is epistemically rational, i.e., serves the epistemic goals of the group and its members.
In social epistemology there is a considerable body of normatively oriented work on opinion formation in groups. This is parallelled by an equally large body of more descriptive work on information sharing in social psychology. The project will make use of both.
In social epistemology, social deliberation is often approached using mathematical models which contain simple representations of how individuals influence each other’s opinions (Hegselmann and Krause 2002, Douven and Riegler 2009, Bala and Goyal 1998, Zollman 2007, O’Connor and Weatherall 2017, Lehrer and Wagner 1981). The dynamics of interaction is explored in order to characterise the conditions for reaching either consensus or polarisation of the group. Work in social epistemology has what philosophers call a ‘normative’ aim. A normative endeavour in this sense, attempts not only to describe opinion formation in groups, but primarily attempts to characterise how it should work. Thus, it will often focus on modeling how an adequate opinion for most individuals, or for the group as a whole, emerges when these agents follow classical communicative rationality principles in their deliberations. This focus means that in these models individual agents are often treated as interchangeable, assumptions about the mechanisms involved in social influence are often rather schematic, and not much attention is paid to the underlying psychology.
In social psychology, on the other hand, we naturally find extensive empirical study of the nature of social influence. A key insight from this research is that social identity has a rather profound influence on what a person believes or communicates. Even the self is partially constituted by a group identity, and so belief formation in the self is deeply impacted by the social context. People are likely to trust members of their group as sources of information especially when that group is relevant to the topic of influence, when they identify strongly with the group, and when group identity is threatened (Spears 2021). They will then also mistrust outgroup sources, making cooperation and consensus across group boundaries increasingly unlikely (cf. Spears, 2021; Spears, Jetten & Scheepers, 2002; Spears, Lea & Lee, 1990). While such mechanisms are important drivers of real-life information exchange, they are rarely, if ever, represented in the social epistemological models alluded to in the foregoing.
In order to clarify how social mechanisms impact on opinion formation processes, the models must be revised and made psychologically more realistic, i.e., they must be connected to the empirical situations that are supposed to be about. We plan to make this connection by examining closely the kinds of assumptions about social influence that are currently at work in specific models from social epistemology, and to see if we can revise these assumptions in the light of insights from social psychology, or else determine what the limitations of such models are. For this we will rely on psychological literature about both the intra- and inter-group exchange of information alluded to above.
Opinion dynamics in social epistemology is standardly studied with reference to certain models of what counts as rational. In the project we aim to redesign our notion of rationality by taking into account psychological insights on social influence. The first step will be to review these insights and adapt the afore-mentioned deliberation models, in order to capture the relevant social influence mechanisms. The following mechanisms will be considered.
- Exclusion based on social identity, whether positively or negatively evaluated, will hamper information sharing within heterogeneous groups, e.g., when cultural and institutional divides fragment a research community. But this may impact positively on the community’s truth-finding ability.
- Positively evaluated inclusion based on social identity lines can help interest groups to form coherent posititions within a larger social deliberation. This can positively influence emancipation movements.
- Negatively evaluated inclusion, i.e., forced opinion alignment within a group, is often thought of as introducing epistemically detrimental “group think”. But peer pressure can also help individuals to improve on their contributions to a social deliberation.
The second step will be to assess the merits and defects of the above mechanisms. We suggest that these specific mechanisms, which are normally thought of as distorting rational information exchange, might in fact support it. This realization will make us look differently at social mechanisms that are until now considered solely as a threat, and as a source of irrationality.
For social epistemology, and for philosophy more generally, such a reevaluation will be a major advance: until now the understanding of social influencing by social epistemologists is very partial, and social mechanisms are mostly viewed as harmful distortion. Besides that, the project will connect to and hopefully impact on social psychology itself. At the input end, we rely on existing literature describing the social mechanisms that are at play in information exchange. And at the output end, our hope is that the revisions to our notion of rationality will offer social psychologists a different perspective on their findings. Rather than reading these findings negatively, like social epistemologists mostly do, and taking them as a reason to design deliberative practices in which the driving mechanisms are suppressed and counteracted, social psychologists might read them more positively, as contributing to successful information exchange. Accordingly, research into these mechanisms might lead us to design social practices and institutions in which the existence of these mechanisms is acknowledged, and in which their advantages can come to the fore.
- Bala, V. and Goyal, S. (1998) Learning from neighbors, Review of Economic Studies 65 (3), pp. 595-621.
- Douven, I. and Riegler, A. (2010) Extending the Hegselmann-Krause model I, Logic Journal of IGPL 18(2), pp. 323-335.
- Habermas, J. (1981). The Theory of Communicative Action.
- Hegselmann R. and Krause, U. (2002) Opinion dynamics and bounded confidence models, analysis, and simulation, Journal of Artificial Societies and Social Simulation, 5 (3).
- Kitcher, P. (1990) ‘The division of cognitive labor’, Philosophy of Science 87, pp. 5-22.
- De Langhe, R. (2014) ‘A Unified Model of the Division of Cognitive Labor’, Philosophy of Science 81, 3, pp. 444-459.
- Lehrer, K. and C. Wagner (1981). Rational Consensus in Science and Society: A Philosophical and Mathematical Study, Springer Verlag.
- O’Connor, C. and Weatherall, J. (2017) Scientific polarization. European Journal for Philosophy of Science 8 (3)pp. 855-875.
- Spears, R. (in press/2021). Social influence and group identity. Annual Review of Psychology.
- Spears, R., Jetten, J., & Scheepers, D. (2002). Distinctiveness and the definition of collective self: A tripartite model. In A. Tesser, J.V. Wood, & D.A. Stapel (eds.). “Self and motivation: Emerging psychological perspectives” (pp. 147-171). Lexington: APA.
- Spears, R., Lea, M., & Lee, S. (1990). De-individuation and group polarization in computer-mediated communication. British Journal of Social Psychology, 29, 121-134.
- Zollman, K. (2007) The communication structure of epistemic communities. Philosophy of Science 74 (5), pp. 574-587.
prof. dr. Russell Spears (Social Psychology)
prof. dr Jan-Willem Romeijn (Philosophy)
dr. Leah Henderson (Philosophy)
Faculty of Philosophy of the University of Groningen