Joëlle Proust (IJN, ENS, Paris)
Presentation of the workshop
Asher Koriat (University of Haifa)
"Subjective confidence, self-consistency, and social consensus"
What is the basis of subjective confidence in our knowledge and judgments? Why are confidence judgments generally accurate in discriminating between correct and wrong responses? According to the self-consistency model, the process underlying subjective confidence in one’s beliefs has much in common with that underlying statistical inference about the outside world. Participants retrieve a small sample of cues from their memory, and their confidence reflects an assessment of the likelihood that the decision reached is true of the entire population of cues. Because the population of cues associated with an item is largely shared across participants with the same experience, subjective confidence taps the collective wisdom, and hence tends to be higher for consensual responses than for nonconsensual responses. The confidence/accuracy correlation is generally positive because collective wisdom tends to lean towards the correct answer. However, it is negative when collective wisdom is biased in favor of the wrong answer. The theory has implications for several issues including choice replicability, social conformity, group decisions, and the wisdom of crowds.
Fabrice Clément (UNINE, Neuchâtel)
"The impact of consensus in beliefs formation: Some developmental and cross-cultural elements for a global reflection".
Consensus is an important way to evaluate the epistemic validity of statements, notably when subjects are not able to check by themselves the validity of their content. During the Dividnorm project, we had the opportunity to test the “epistemic weight” of consensus for preschoolers. The main results will be summarized and we will present recent results comparing how children from France and Japan react when they had the opportunity to observe a consensus that was far less reliable than a dissenter. When they had to choose between an unreliable consensus and a reliable dissenter, we will show that Japanese and French children did not take exactly the same decision. These results tend to show that epistemic norms are influenced by the culture children grow in.
Shoji Itakura (University of Kyoto)
"Nonverbal theory of mind: Evidence in Japanese children".
There is an argument regarding whether the development of false belief reasoning is universal across cultures. Many studies have supported the universal development of false belief understanding. Callaghan et al. (2005) reported that children in five cultures could not pass the false belief task until they were between four and five years of age, and that the development trend was similar across cultures. Several studies, however, had shown that Japanese children might have a one to two year delay in false belief understanding (see Naito and Koyama, 2006). We investigated whether this delay is genuinely due to the children’s difficulty with false belief reasoning, or whether the verbal questioning technique underestimates the competence of the participants. We gave 4- and 5-year old Japanese children a verbal and nonverbal false belief task. The results showed that the children performed significantly better in the nonverbal task than in the verbal task. In addition, 5-year-olds performed significantly above chance in the nonverbal task, but not in the verbal task. These results suggest that Japanese children show difficulty with false belief tasks because verbal tasks may underestimate their competence.
Sunae Kim (Ludwig Maximilian University of Munich)
"Cross cultural studies on young children's metacognition".
In this talk, I will first present 3- and 4-year-old German children's sensitivity to their own ignorance as revealed in their informing decisions and uncertainty gestures in contrast to the overestimation of their knowledge in their verbal judgments. In the latter half of the talk, I will present data on cross-cultural differences and similarities with respect to metacognition and mind reading. Theoretically it is argued that mindreading, children's false belief understanding more specifically, is related to children's explicit metacognitive abilities. A few recent studies report empirical evidence for this. We intended to extend the prior research and further systematically investigate the relationship between mindreading and metacognition via a cross cultural investigation. To this aim, we tested 4-year-old German and Japanese children's implicit and explicit forms of metacognitive abilities as well as false belief understanding. We found that Japanese and German children equally displayed a gap between implicit and explicit forms of metacognitive abilities. By contrast, they differed in their mindreading abilities. The cross cultural findings shed light on the relationship between metacognition and mindreading.
Chris Frith (University College London)
"The social relevance of explicit metacognition for perception and action".
If metacognitive representations are defined as representations of the properties of cognitive processes, then we should distinguish between implicit and explicit metacognition. At the sub-personal (implicit) level, behaviour is affected by many metacognitive properties, such as precision of sensory signals, without awareness. However, some of these properties become available at the personal (explicit) level. Examples include, perceptual fluency, action selection fluency, and mental effort. These are properties of cognitive processes that are experienced subjectively and can be reported to others.
There is, however, considerable leeway in how we report and interpret these experiences. Our behavioural response to metacognitive signals depends upon how we interpret them and this interpretation can be influenced by others and by our cultural environment more generally. For example, we use our experience of action to justify our behaviour and learn to feel responsibility and regret. We can also translate our sensory experience into reports of confidence that can create advantages either for the group or for the individual. Rather than being a problem, the malleability of our explicit metacognition has a critical role in enabling sharing of subjective experience and in the development of cultural norms.