Event-related Potentials Elicited By Numerical Induction In Children And Adults

Inductive reasoning is the process of inferring a general rule by inspection of specific instances, which is regarded as a critical constituent of human intelligence. To be more specific, inductive reasoning could be defined as a process which producing a general rule (the conclusion) based on the observation of a number of specific stimuli (the premise). It is used in generating hypothesis, formulation theories, discovering relationships. What's more, it is essential for scientific discovery.Since inductive reasoning is a rapid automatic cognitive process, it is necessary for us to explore its temporal characteristics with Event-related Potentials (ERP), which has a high temporal resolution. Numerical series completion task is a kind of typical inductive task, which primarily contains the processes of rule generation and extrapolation. Rule generation can be divided into two sub-stages including rule discovery and rule confirmation; while extrapolation is the process of rule application. Until now only a few researches have explored the neural correlates and temporal characteristics of numerical inductive reasoning in adults, neural mechanism of children in this field is totally a blank. Thus the aim of this study was to explore neural correlates of numerical induction in children, and comparing the results of children with adults to find the developmental changes. The whole study was including two experiments, one was to explore the rule generation stage, and the other one was to explore extrapolation stage.Experiment I investigated the rule generation process of numerical series completion task, including induction task and calculation task. On one hand, we compared the ERP results of three numbers in induction task, and found that the temporal characteristics of numerical induction in children and adults were almost the same:there was no difference in early components before 300ms in the three numbers, which indicated that children and adults has the similar early visual process on all numbers; There was a main effect in P3 (300-500ms) which was proved to be the index of rule generation, showing the beginning of inductive reasoning process. The larger amplitude of P3 on NO.3 suggested that no matter children or adults, had inferred a certain rule after the presentation of NO.3.and was going to apply it to the test numbers.On the other hand, comparing ERP results of three numbers in induction and calculation condition respectively, we found that the dissociation of induction from calculation was almost the same in children and adults too:dissociative process began from the process of rule generation, which means 700-900ms after presentation of NO.2 and 300-500ms after presentation of NO.3. However, children showed difference of N2 on NO.3, but adults didn't. The difference of N2 revealed that there was a conflict when children induced the certain rule form three numbers. This conflict happened when children induced a rule from three premise numbers and it was not the same as the rule generated from the first two numbers. Since children had a comparatively low cognitive level, the conflict was bigger than adults. In addition, it was easy to find from the topographies that ERP components were mainly distributed on prefrontal cortex in children, but mainly on parietal and occipital cortex in adults. As prefrontal is related to working memory and intention, this difference maybe indicated that to finish the same task, children need more working memory and intentional resources than adults.Experimentâ…¡investigated extrapolation stage. Incongruent numbers elicited larger amplitude of N2 in children and adults, which indicated an expectation violation or conflict detection when a wrong number was showed. However, children and adults had different results on P2 component: adults had no difference on P2, while children showed significant difference on P2 in congruent and incongruent condition. This result suggested that comparing with adults; children detected the difference of congruent and incongruent numbers in an earlier perceptual process. In other words, children were more sensitive to conflict stimuli. This is because children are in the process of intelligence development, which means to finish the same task children need more cognitive resources than adults; more cognitive resources naturally bring bigger reaction to conflict, as a result children showed difference of ERP component in an earlier time window.In conclusion, this study explored the process of numerical induction completely through two experiments. As to rule generation stage, children and adults showed almost the same results, indicating that the two groups have a similar temporal course of numerical induction. The only difference between children and adults lies on N2 component of NO.3, which showed children were in the process of in intellectual development, and has a lower cognitive level than adults. As to extrapolation stage, children detected the difference of congruent and incongruent numbers in an earlier perceptual process, suggesting comparing with adults, children were more sensitive to conflict. This can be regarded as the unique characteristic of children in extrapolation of numerical induction. In a nutshell, in the process of numerical induction, children showed the same temporal characteristics as adults, and the differences between two groups proved that children are in the process of intellectual development, and lacking related practice and experience.