Computational models of social cognition should incorporate social relationships as core primitives
- 28 May 2026
- journal article
- open peer-commentary
- Published by Cambridge University Press (CUP) in Behavioral and Brain Sciences
- Vol. 49, e29
Abstract
Relational primitives can facilitate rapid inductive generalization beyond individual mental-state reasoning. Computational models of social cognition should incorporate social relationships as core primitives, enabling richer explanations for how minimal social observations lead to flexible, efficient, and culturally adaptive social predictions.Keywords
This publication has 14 references indexed in Scilit:
- Planning with Theory of MindTrends in Cognitive Sciences, 2022
- Toward a computational theory of social groups: A finite set of cognitive primitives for representing any and all social groups in the context of conflictBehavioral and Brain Sciences, 2021
- Rational quantitative attribution of beliefs, desires and percepts in human mentalizingNature Human Behaviour, 2017
- Pragmatic Language Interpretation as Probabilistic InferenceTrends in Cognitive Sciences, 2016
- The Naïve Utility Calculus: Computational Principles Underlying Commonsense PsychologyTrends in Cognitive Sciences, 2016
- We read minds to shape relationshipsPhilosophical Psychology, 2014
- The mentalistic basis of core social cognition: experiments in preverbal infants and a computational modelDevelopmental Science, 2013
- A tutorial on Bayesian nonparametric modelsJournal of Mathematical Psychology, 2012
- Action understanding as inverse planningCognition, 2009
- Learning overhypotheses with hierarchical Bayesian modelsDevelopmental Science, 2007