Development Of Category Induction

Inductive reasoning is essential in our daily lives as a cognitive ability. Inductive reasoning is defined as the process of inferring the general object also has the same attributes by observing and analyzing the attributes of special objects. It is the basis of establishing rules, forming concepts, generating hypotheses and selecting hypotheses. People understand the world and construct knowledge in this way. Induction constitutes a central aspect of intellectual functioning. People have been thinking about induction since ancient times. Eastern and western philosophers also probed into inductive reasoning long ago. While philosophers are interested primarily in the epistemic relationship between premises and conclusion, psychologists are concerned with the cognitive processes or mechanisms involved in drawing the inference.From initial behavior study to neural mechanisms research that carried on in recent decades, it laid a preliminary theoretical foundation for inductive reasoning. In fact, early behavior studies did not investigate the inductive reasoning process itself, and not involve the formation mechanisms. And the previous image study used PET or fMRI technique to invested brain localization. Due to their low time resolution, it is not adequate to explore the dynamic time course of induction process. The temporal resolution of ERPs is in the range of milliseconds, can provide more information on the relative timing of the cognitive component underlying inductive reasoning process. In addition, the subject of the previous neural mechanisms study was mostly with brain lesion patients and normal adults, but few studies for children and adolescence. As described above, in the present study we used ERPs to examine the temporal course and electrophysiological correlates of category induction process with different age groups. In addition, with the physical brain continues to mature, especially the maturation of prefrontal cortex, changes in cognitive function will follow. Therefore, the present study also investigated whether there are development differences. The materials and procedures adapted from the study of Li et al (2009). In order to eliminate the effects of knowledge background and language comprehension, we designed the task that composed of geometric figure. Each figure were represented a piece of biscuit. Participants were presented simultaneously with three biscuits that varied in shapes and colors, and then they were asked to learn what kind of biscuit was eatable. The biscuit was eatable or not by detecting with an instrument. In this task, we defined condition as the induction in which the triplets of biscuits have a common attribute (i.e., can make an inductive generalizations). Whereas the condition in which no attribute shared by the triplets of biscuits is a non-induction (i.e., can not make an inductive generalizations). In a later study, in order to effectively examine the development trends of category induction, the same experiment materials and paradigm were used in the study 2 and 3. The three studies to investigated time course and electrophysiological mechanisms of category induction process for late childhood, early adolescence and early adulthood, respectively. Also, the development pattern was explored.The results showed that there were some common features of scalp potentials between children and adolescence, similar ERP waveforms and components was evoked in front of the scalp, the early components N1 and P2 were elicited in the fronto-central electrode, for the amplitude and latency of N1 and P2, the main effect of condition were not significant. About 300ms after onset of stimulus, there was an amplitude peak obvious N2, main effects of conditions was significant. Following the N2, there was a late positive component with a longer duration. The interaction of condition by laterality was significant for late positive component. Tests for simple effects indicated that the effects of condition were dominantly occurred in left front electrode rather than right. But there were not significant difference between left hemisphere and right hemisphere for adults.The three studies have several implications. Firstly, the behavioral data showed shorter reaction times in the induction than in the non-induction condition for three age groups, reflecting faster detection of the attribute were shared by the two stimuli. And the reaction time was decreased with the increases of age, indicating that category induction process undergoes mature changes. Secondly, the amplitude and latency of the ERP components showed an age-related development changes, that is, amplitude and latency were significantly shortened with the increasing age. In particular, N2 latency showed a reverse development changes. Thirdly, the ERP data indicated that the more negative N2 amplitude was observed in non-induction condition for three age groups, which may reflect mental conflict detection. The late positive component reflects the process of hypothesis generation and the memory context updating. Fourth, brain topographic map revealed the activation in the left prefrontal lobe for children and adolescence, but non-lateralization for adults, reflecting there was an age-related development differences on the core process stage of the category induction. Finally, category induction can be divided into the following four stages:the first was attention stage, in which is mainly included the allocation of early attention resources and perceptual analysis, reflected by N1 and P2 component; then was comparison stage, at this stage subject compared stimulus precisely, reflected by N2 component; the third stage was hypothesis generation, the corresponding component of hypothesis generation and updating of memory context was the late positive complex. The final stage was hypothesis testing stage that after the probe stimulus appeared.