About this srcd poster session

---

Panel information
Panel 20. Social Cognition

Abstract

---

Young children revise their beliefs in response to new evidence (Kimura & Gopnik, 2019), often replacing a previous belief with one suggested by that evidence. Can this be considered rational belief revision? Some philosophers argue that replacing a belief does not qualify as rationality, as it lacks reflection. They contend children can only be deemed rational when they can reflect on their beliefs and the reasons supporting them, specifically, responding to "why" questions, typically around the "age of reason" (6 years) (Boyle, 2018). However, responding to counterevidence that undermines the evidence supporting a belief indicates a level of reflection, as it requires the identification and assessment of that evidence (Melis & Monsó, 2023).
Research on children’s belief revision in response to undermining counterevidence is limited. Schleihauf et al. (2022) found that 4- to 5-year-old children were more likely to revise their beliefs when exposed to undermining evidence that challenged their initial belief, compared to positive evidence that supported it. This pattern suggests that children reflect on the reasons behind their beliefs and the supporting evidence, indicating that rational belief revision occurs earlier than the "age of reason." This raises the possibility that reflective belief revision and rationality may be within reach of minimally verbal or non-verbal populations. To explore this, we conducted two preregistered studies examining reflective belief revision in 2- to 6-year-old children in response to verbal and non-verbal undermining counterevidence.
In Study 1, we aimed to replicate Schleihauf et al. (2022) with 3- to 6-year-old’s (N = 35). Using the original materials, we presented children with initial evidence and asked them to state their belief and reasoning. They then received either positive or undermining evidence, followed by evidence suggesting an alternative belief, and were asked to restate their belief. Replicating previous results, 4- to 6-year-olds were significantly more likely to revise their belief when exposed to undermining evidence (Figure 1a). Data collection with 3-year-olds is ongoing (due for completion November 2024); preliminary findings suggest a similar pattern of belief revision to that of older children.
In Study 2, we examined belief revision in adults (N = 23) and 2- to 6-year-old’s (N = 84) using a minimally verbal adaptation of the same task, conducted with eye-tracking. The task structure was maintained, with verbal testimony replaced by visual and auditory evidence. Full analyses are pending (data collection completed September 2024), though preliminary results indicate adults, like children in Study 1, revised their beliefs significantly more often when exposed to undermining evidence compared to positive evidence (Figure 2). Children showed an interaction between the evidence type and age. While 3- to 6-year-olds exhibited the same pattern as adults, albeit to a lesser degree, 2-year-olds displayed an opposing pattern suggesting that minimally verbal 2-year-olds were not reflectively revising their beliefs.
Together, these results confirm that children as young as 3 can reflectively revise their beliefs in response to verbal and non-verbal undermining counterevidence, suggesting they possess the capacity for rational belief revision from this age.

Author information

Author	Role
Kirsten H. Blakey, University of Toronto	Presenting author
Brina Recelj, Kings College London	Non-presenting author
Chloe Dow, University of Stirling	Non-presenting author
Zsófia Virányi, Messerli Research Institute, University of Veterinary Medicine Vienna, Medical University of Vienna, University of Vienna	Non-presenting author
Giacomo Melis, University of Stirling	Non-presenting author
Eva Rafetseder, University of Stirling	Non-presenting author

⇦ Back to session