ESTIMATION METHOD OF PERCEIVED SEMANTIC CONTENT BY ANALYSIS OF BRAIN ACTIVITY

摘要

PROBLEM TO BE SOLVED: To propose an estimation method that estimates a perceived semantic content by measuring and analyzing brain activity.SOLUTION: An estimation method is characterized by: (1) inputting the language description of a perceptual content induced in a subject by training stimulus and the output of cranial nerve activity detection means detecting brain activity into data processing means; (2) associating the semantic space representation of the training stimulus with the output of the cranial nerve activity detection means in a stored semantic space and storing the result in training result information storage means; (3) inputting the output of the cranial nerve activity detection means detecting the brain activity induced by new stimulus in the data processing means, and obtaining the probability distribution on the semantic space representing the perceived semantic content from the association concerning the output of the cranial nerve activity detection means of the brain activity by the new stimulus; and (4) estimating the content of high probability as the perceived semantic content from the probability distribution. The association may be performed for every subject. The estimation of the probability uses a coordinate on the semantic space of a given single word and the likelihood calculated from the probability distribution as an index.SELECTED DRAWING: Figure 1