Abstract:Most models indicate that category learning is a kind of associated learning driven by reducing categorization errors.According to this account,people solely learn features useful for categorization.By contrast,some researchers indicate that different from association learning,category learning is a high level cognitive learning,and people like to learn more features than classification.A dualprocessing account is advanced to describe the category learning mechanisms,which indicates that there are both an error - driven associative mechanism and a cognitive mechanism in category learning.According to the dual - processing account,the goal of associative mechanism in category learning is to learn as fewer features as possible,which is error - driven.Whereas,the goal of cognitive mechanism is to learn as more features as possible,and learning more features would be useful for future feature induction and so on.And these two mechanisms exist in category learning and affect features learning. Three experiments examined the dual - processing account in a blocking paradigm.In the blocking paradigm,participants learned one defining feature dimension,which was crucial for categorization.By accounting the blocking effect that the defining feature dimension had on other features learning,we investigated the dual - processing mechanism in category learning.In Experiment 1,all features were presented randomly to investigate the blocking effect in category learning.The results showed that even some non - defining features were blocked by pre - learning of the defining features,which supports that the associative mechanism and cognitive mechanism both exist in category learning.In Experiment 2,defining features were defined on the top of exemplars to prevent participants from learning the non -defining features that,in Experiment 1,was caused by searching for defining features.Consistent with Experiment 1, the results supported the dual - processing account.Different with the sentence described categories in Experiment 1 and 2,figure categories were adopted in Experiment 3 and similar results were found. Across all three experiments,the dual - processing account has been verified in different feature presentations and category materials. The associative mechanism predicted the blocking effect in category learning,which the cognitive mechanism predicted the learning of some non—defining features.

The Development of Category Learning Researches of Cognitive Neuroscience

Dual-system Model of Category Learning

Multiple Systems of Category Learning From the Perspective of Cognitive Neuroscience and Neuropsychology

Cognitive and Neural Mechanisms of Category-based Feature Inference

Mechanisms underlying category learning in the human ventral occipito-temporal cortex

Exploration, Distributed Attention, and Development of Category Learning

The Cognitive Mechanism of Category Learning

Dissociable changes in functional network topology underlie early category learning and development of automaticity

The Multiple Systems Theory in Category Learning

A computational model of prefrontal and striatal interactions in perceptual category learning

The Influences of Feedback on Perceptual Category Learning and Its' Cognitive and Neurophysiological Mechanisms

Category Learning Selectively Enhances Representations of Boundary-Adjacent Exemplars in Early Visual Cortex

A comment on a New Category Learning Model: SUSTAIN

Learning Shapes Spatiotemporal Brain Patterns for Flexible Categorical Decisions

Learning the exception to the rule: model-based FMRI reveals specialized representations for surprising category members

Neural dynamics underlying the acquisition of distinct auditory category structures

Brain Mechanisms of Concept Learning

Representing Object Categories by Connections: Evidence from a Mutivariate Connectivity Pattern Classification Approach.

On the interaction of prior knowledge and stimulus structure in category learning

Exploration of the Potential of Brain in Learning: Research on Cognitive Neuroscience

Category Learning Blocking and Dual-processing Account