ホムンクルスをノントリヴィアルな方法で取り戻すこと

抄録

At present, investigating the neural correlates of conscious percept (Crick and Koch 2003) is one of the promising scientific approach to the problem of subjective experience. The neural correlates paradigm is a direct extension of the standard model of neural information representation based on the response selectivity paradigm (Hubel and Wiesel 1962). With the advent of various imaging techniques such as fMRI and MEG, the neural correlate approach is an obvious and useful way of approaching the physical origin of consciousness. Here I argue that in order to come to a full understanding of the neural correlates of conscious percept, we need to "reclaim homunculus" in a nontrivial way. The old idea of a single central area monitoring the neural activities in various areas of the cortex is gone. However, in order to account for the "integrated parallelism" of information processing in the brain, as is evident from observations in active vision (e.g. binocular rivalry) and sensori-motor coordination (e.g. body image, mirror neurons), we do need to come to understand a neural basis of subjectivity in which all information representation in the brain should become relevant.Citing several neurophysiological evidences from my own laboratory and elsewhere, I put forward several working hypotheses regarding the neural basis of subjectivity. In particular, I discuss how the time parameters relevant in the neural coding of information across the multiple cortical regions are coordinated in a systematic way to give rise to a coherent sense of agency. I also propose several specific experiments to be conducted in order to test the suggested hypotheses. [Jpn J Physiol 54 Suppl:S49 (2004)]