The Typicality Effect And Diagnostic Effect On Category Learning

Category learning is the most fundamental cognitive activities of human,and the ability of category learning is basic ability.There are two ways of category learning: classification learning and inference learning.The former refers to telling all the information about the category members,and you should judge which label belongs to the stimulus,while the later is about telling the category label and part of the stimulus,and you are supposed to make a decision of what the lack of information would be like.The stimuli were sine wave gradients(gabor patches)that varied in their spatial frequency and spatial orientation or family resemblance material which refers to a feature of the family members are somewhat similar but different were usually used in perceptual category learning.But in the real life,it is not the only way to present stimulus.Another way is to present them by language or character.So it was also named semantic category learning.In previous studies,there was controversy about which way is more benefit for perceptual category learning and what are the results between classification learning and inference learning.At present,there were few studies about semantic category learning.So it is not clear that what the two way of category learning have influence on semantic category learning.If they effect semantic category learning.This research discusses the learning mechanism of the perceptual category learning and semantic category learning by two experiments.In order to be more comparative,we use word to describe the materials of semantic category learning.It can ensure the difficulty and the amount of information be the same as much as possible.The experiment 1 explored the results of the two learning ways between perceptual category learning and semantic category learning.It said that there were different leaning mechanisms.The experiment further explores the learning mechanism.It inspected the typicality effect and diagnostic effect by typical evaluation and single feature test.The studies may have conclusions as bellow: inference learning is better than classification learning in perceptual category learning,but worse in semantic category learning.There were typicality effect and diagnostic effect.But in the perceptual category learning,diagnostic feature has no influence on inference,but in have influence in the semantic category learning.