Abstract
The weather prediction (WP) task is a probabilistic category learning task that was designed to be implicit/procedural. In line with this claim, early results showed that patients with amnesia perform comparably to healthy participants. On the other hand, later research on healthy adult participants drew attention to the fact that the WP task is not necessarily implicit. There have been results showing that participants can access structural information acquired during the task. Participants also report that their responses are based on memories and rule knowledge. The contradictory results may be reconciled by assuming that while explicit learning occurs on the WP task in case of adults, in children the learning process is implicit. The present study aims at testing this hypothesis. Primary school children completed the WP task; the experimental group performed the original task, whereas in the control group cues and outcomes were associated on a random basis, hence their version of the WP task lacked a predictive structure. After each item, participants were asked whether they relied on guessing, intuition, “I think I know the answer” type of knowledge, memories of previous items, or knowledge of rules. Self-insight reports of the experimental group were compared to a control group. Results showed that children learn similarly to adults: they mostly (but not completely) rely on explicit, and not on implicit processes.
Highlights
SELF-INSIGHT IN PROBABILISTIC CATEGORIZATION – NOT IMPLICIT IN CHILDREN EITHER Traditionally cognitive psychology interprets human memory as a fractionated system
Weak cues were learned later and were considered less important. These results show that healthy participants rely on explicit mechanisms during the weather prediction (WP) task
DATA ANALYSIS Accuracy in the WP task is the ratio in which participants make the correct prediction based on the combined probability
Summary
SELF-INSIGHT IN PROBABILISTIC CATEGORIZATION – NOT IMPLICIT IN CHILDREN EITHER Traditionally cognitive psychology interprets human memory as a fractionated system. PD patients show the opposite pattern as the Amnesia group: their performance on the debriefing task is identical to that of healthy participants, while their categorization is not better than chance, especially in the early phases of the task (Knowlton et al, 1996) These results suggest that early in the task participants rely on their procedural systems, while the later phases tap declarative memory. Previous studies of the WP task found that while participants show consistent strategy use, they are mostly unable to report it, and even if they do report strategy use, it is very unlikely they report the one they had employed (Gluck et al, 2002) This result does not necessarily reflect implicit learning as self-reports only reveal strategy use, no structural information is collected about acquired representation (in the current case: about the strength of association between cues and outcomes). These results show that healthy participants rely on explicit mechanisms during the WP task
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
