Reading minds or reading scripts? De‐intellectualising theory of mind

Abstract

Understanding the origins of human social cognition is a central challenge in contemporary science. In recent decades, the idea of a ‘Theory of Mind’ (ToM) has emerged as the most popular way of explaining unique features of human social cognition. This default view has been progressively undermined by research on ‘implicit’ ToM, which suggests that relevant precursor abilities may already be present in preverbal human infants and great apes. However, this area of research suffers from conceptual difficulties and empirical limitations, including explanatory circularity, over‐intellectualisation, and inconsistent empirical replication. Our article breaks new ground by adapting ‘script theory’ for application to both linguistic and non‐linguistic agents. It thereby provides a new theoretical framework able to resolve the aforementioned issues, generate novel predictions, and provide a plausible account of how individuals make sense of the behaviour of others. Script theory is based on the premise that pre‐verbal infants and great apes are capable of basic forms of agency‐detection and non‐mentalistic goal understanding, allowing individuals to form event‐schemata that are then used to make sense of the behaviour of others. We show how script theory circumvents fundamental problems created by ToM‐based frameworks, explains patterns of inconsistent replication, and offers important novel predictions regarding how humans and other animals understand and predict the behaviour of others.

What is in an action? Preschool children predict that agents take previous paths and not previous goals

“…Moreover, there was no difference between animated stimuli and the presentation of a human grasping action in children. This pattern may relate to recent theoretical models that propose that action prediction is based on knowledge about the transition probabilities between actions (Thornton & Tamir, 2021), past regularities (Ruffman et al, 2012), and/or structured experiences and scripts (Taylor et al, 2023). In another study, 2-year-old children anticipated neither the goal nor the previous location systematically ( Krogh-Jespersen et al, 2018).…”

“…It happens rarely that someone grasps the same object over and over again. At the same time, a large part of human action is based on regularities and/or scripts in which one action leads to another action (Taylor et al, 2023). One does not necessarily need to know about the goal of another person and nevertheless can learn about scripted scenes and some behavioral regularities.…”

“…For example, Csibra (2003) suggested that the findings might be explained by infants associating a hand with a specific object (for a similar view see, Perner & Ruffman, 2005). From another point of view, considering the replication failures in a related domain such as Theory-of-Mind (ToM), results should be interpreted carefully (Taylor et al, 2023). Thus, it would be valuable to have further evidence from other approaches on how to interpret these findings.…”

Rethinking ostensive communication in an evolutionary, comparative, and developmental perspective

“…Even so, it may well be that much of the time, infants understand an agent's intentional behaviour nonmentalistically. And indeed, actions or action sequences of a familiar type may often elicit such basic, non-mentalistic understanding in older children and adults (Kaufmann & Clément, 2014;Malle, 2004;Taylor et al, 2023).…”

“…Humans are spontaneous mentalizers. Even if, as Malle (2004), Kaufmann andClément (2014), andTaylor et al (2023) suggest, they are often content to categorize and contextualize the routine actions of others without metarepresenting the intentions behind them, these mentalistic representations are manifest to them, and are likely to be mentally represented whenever they are relevant enough.…”