An Investigation on Human Multi-Perception Mechanism by Cooperatively Using Psychometrics and Data Mining Techniques

摘要

In previous studies, human mechanisms of the five senses have been investigated separately, such as the visual perceptive system, the auditory perceptive system, the tactual system and so on. Based on human perceptive characteristics, many artificial systems have been developed. However, the performance of such systems is nowhere near human ability, particularly in the cognitive technology and the artificial intelligence related areas. In order to develop the artificial systems which match human ability, it is necessary to investigate human multi-perception mechanism systematically. The key issues are how to design the psychological and physiological experiments for obtaining various data from human multi-perception mechanism, as well as how to analyze such data from multiple aspects for discovering new models of human multi-perception. In this paper, we propose a new methodology with a multi-step process, in which various psychological experiments, physiological measurements and data mining techniques are cooperatively used to investigate human multi-perception mechanism. In particular, we describe the design of psychological experiments and the measurements by the functional magnetic resonance imaging (fMRI) to obtain various data. Furthermore, data transformation and data mining techniques for such purpose are also presented.