An ERP Study Of Inductive Reasoning And Its Relationship With Memory

Inductive reasoning,as a high-level cognitive process,plays a critical role in human cognitive system,and mainly contains two forms:numerical inductive reasoning and category-based inductive reasoning.At present,the neural mechanisms of inductive reasoning and the relationship of inductive reasoning with other cognitive processes such as classification,perceptual processing,memory are hot research topics,however,two major problems are still unsolved:First,the time course of inductive reasoning remains to be further clarified.Second,the relationship between inductive reasoning and memory lacked evidence from a neurological perspective..Thus,this article carried out two experiments to solve the problems,using event-related potential(ERP)technology.Experiment 1 explored the time course of checking under different time pressure conditions from the perspective of numerical inductive reasoning,using number series completion task.Valid,invalid,and anomalous series were selected according to whether they meet the sequence rules and the degree of departure from the rules.In the present study,numbers were presented in sequence,participants were asked to verify whether the sequence complies with a rule.The results of the study showed that the longest reaction time was for the invalid series,followed by the anomalous series and the shortest for the valid series.At the neural mechanism level,both invalid series and anomalous series recorded N2 reflecting cognitive conflict monitoring,P300 reflecting sequence evaluation,and LPC reflecting calculation process.The amplitudes of N2 induced by invalid series was stronger than anomalous series,and the P300 and LPC were weaker than invalid series.Time pressure only affected the LPC.There are significant differences in the amplitudes of LPC of invalid and anomalous series under time pressure but no significant differences under no time pressure.Experiment 2 was carried out from the perspective of category-based inductive reasoning and recognition,providing neural evidence for the relationship between inductive reasoning and memory.Four types color images of dogs were used as exprement stimuli:old large dogs,new large dogs,new medium dogs,and new small dogs.Except differences in task requirement,the stimuli and procedures were identical in category-based inductive reasoning and recognition.In category-based inductive reasoning task,participants were told to generalize the common property of large dogs presented in series of images during the study phase.During the test phase,participants were asked to judge whether the presenting image had the property;in the recognition task,subjects observed and remembered images during the study phase.During the test phase,participants were asked to judge whether they had seen the image presented.The results of the study showed that the rate of positive response to category-based inductive reasoning and recognition was highly correlated(r = 0.669),the positive response rate and the response criterion(d)of category-based inductive reasoning was lower than recognition.At the neural mechanism level,FN400 components reflecting familiarity and LPC components reflecting recollection were found in the recognition task,which were not found in category-based inductive reasoningThe main conclusions of this study are as follows:(1)Checking can be divided into two stages:stimulus evaluation and calculation,which are reflected by P300 and LPC respectively.(2)Time pressure has an impact on checking.It showed that checking under time pressure was more flexible.Different calculation strategies were used for different types of series.But when under no-time pressure,the same strategies were used to solve different types of series.(3)Category-based inductive reasoning and recognition supports the "deep-correspondence" view at the behavioral data level,but not at the neural mechanism level.