Research On The Influencing Factors Of Information-Integration Category Learning

Category learning is an essential cognitive function. Empirical evidence and theoretical reasons suggest multiple dissociable category learning systems are involved in category learning tasks, and the extent to which each system is recruited is task dependent. Much recent research in the categorization domain has focused on tasks that can be learned outside of conscious awareness, which can be characterized as implicit and involving information-integration. These tasks are thought to be procedural-learning-based, consequently recruiting basal ganglia similar to those used in motor-skills tasks. Across three experiments, we examined the effects of3variables on information-integration category learning, the delay between response and feedback, the informativeness of feedback, and the instructional condition.In Experiment1we investigated how feedback delay affected procedural, information-integration category learning. We predicted that small feedback delays of several hundred milliseconds would lead to the best information-integration category learning based on a highly regarded neurobiological model of learning in the striatum. The result showed that information-integration category learning was best with feedback delays of500ms compared to delays of0and250ms.In Experiment2information-integration category learning was compared under minimal and full feedback conditions with2categories. With minimal feedback subjects are told whether their response was correct or incorrect, but are not informed of the correct category assignment. With full feedback subjects are informed of the correctness of their response and are also informed of the correct category assignment. The result showed that full feedback facilitates the information-integration category learning.Explicit instruction sometimes impairs implicit learning of artificial grammars and of sequential regularities. In Experiment3we asked whether such negative effects of explicit instruction also occur in implicit learning of information-integration category structures. Half of the participants were informed about the pattern and were instructed to try to discover it (rule-discovery instructions), and half were instructed to memorize the exemplars (memory-based instructions). The result showed that explicit instruction impairs information-integration category learning.In conclusion, these results provide evidence for the view that information-integration category learning is subserved by an implicit procedural-learning-based system.