The Influence Of Material Properties And Cognitive Styles On Feature Induction And Prediction

Inductive reasoning entails using existing knowledge or observations to make predictions about novel cases. At presen the arguments of the feature prediction are mainly between category-based induction and feature association induction.The current study attempted to examine the possible influence of material property, cognitive style and representative typicality of examples on feature induction and prediction. The material property includes graphics, Chinese vocabulary and alphanumeric code; the cognitive style includes field independence and field dependence; the representative typicality of examples includes high representative, middle representative and low representative. Participants included 181 undergraduate students who completed two tasks. Experiment 1 is to determine target feature category, For Experiment 2, the category is uncertain.The results revealed that:(1) In both experiment, the male's reaction time (RTs) is significantly longer than female's. (2) In both experiment, the RTs on the Chinese vocabulary and alphanumeric code is considerably longer than on the graphics. (3) When it comes to the feature induction task, the confidence of high representative is higher than middle representative, and the confidence of middle representative is higher than low representative. (4) In addition, participants' confidence proportion is notably different from the proportion with category-based induction under the condition of feature prediction with category determined. Plus, participants' confidence proportion is markedly different from the proportion with category-based and single feature association induction in the task with category uncertain.Conclusions:(1) There are material property, cognitive style, gender and subject difference on feature induction and prediction. (2) There is representative typicality difference on feature induction, as the confidence is higher when the representative typicality is higher. (3) college students tend to base feature prediction with category determined on feature association induction, and base feature prediction with category uncertain on comprehensive feature association induction.