Pierre Jacob Talk

Pierre Jacob

Monday, December the 4th


IHPST, 13 Rue du Four, 75006 Paris

Title: Challenging the two-systems model of mindreading.


Mindreading is the fundamental human social cognitive ability to ascribe psychological states to others for the purpose of predicting and explaining their behavior. The fundamental question addressed by the two-systems model of mindreading advocated by the psychologist Ian Apperly and the philosopher Steve Butterffill is: is there a middle ground between full-blown mind reading and what is called “behavior reading” (by Daniel Povinelli and colleagues) and “sub-mentalizing” (by Cecilia Heyes)? The two-systems model response rests on a distinction between minimal (or implicit) and full-blown (or explicit) mindreading, which is taken to reflect the trade-off between flexibility and automaticity. Minimal mindreading is supposed to be efficient, automatic and to emerge early in human ontogenetic development. Full-blown mindreading is supposed to be flexible and not automatic, less efficient and to develop later. One of the fundamental tasks of the model is to account for the discrepant developmental findings: while preschoolers have been shown to fail explicit false-belief tasks (Wimmer & Perner, 1983; Wellman et al., 2001) preverbal infants have been shown to expect agents to act in accordance with the contents of their beliefs (whether true or false) in implicit tasks (Onishi & Baillargeon, 2005; Baillargeon et al., 2010).

The model makes two basic related claims: (1) the output of the full-blown explicit system is the representation of others’ genuine beliefs (or beliefs as such); the output of the minimal implicit system is the representation of others’ so-called ‘registrations’, not beliefs. (2) Only the full-blown explicit system, not the minimal implicit system, can handle the aspectuality of genuine beliefs.

I will raise three challenges for this model. First, I will challenge the claim that the model can resolve the developmental puzzle. Secondly, I will challenge the claim that the representation of the aspectuality of beliefs falls outside the scope of minimal mindreading. Finally, I will argue that the examination of the contrast between Level-1 and Level-2 visual perspective taking undermines the sharp dichotomy between automatic and flexible cognitive processes.