Naturalization of consciousness

自然主义意识理论的建构需要考虑两个问题：感受质的自然化问题和无意识认知过程的本质的阐明问题。研究结果如下：1.意识经验具有两种特征，即内在性特征和意向性特征。质是经验的意向性特征，而不是内在的。经验具有某种意向性内容的事实，可以用它具有某种功能的事实来解释。因此，感受质作为经验的内在特征被还原为它的功能。有意识的经验之所以有意识，是因为它们有感受质。但是，如果感受质只是意向性特征，那么经验就不一定是有意识的。因为可能存在具有相同意向特征的无意识体验。质必须是有意识的意向性特征。感质是有意识的，这一事实可以用它们可以被言语化来解释。语言是意识的源泉。没有语言的生物缺乏意识。有意识的思维是经典的计算主义的，因为它是一个语言过程。另一方面，无意识思维是连接性的。这是一个转变神经元激活模式的过程。例如，对善与恶或语法性的判断在有意识的情况下是经典的计算主义的，但在直觉的情况下是连接主义的。但是，即使在有意识的推理中，日常推理也往往充满了逻辑缺口。严格地说，它不是经典的计算主义。

Two problems have been considered in order to construct a naturalistic theory of consciousness : the problem of naturalizing qualia and that of elucidating the nature of unconscious cognitive process. The results are the following.1. Conscious experiences have two kinds of features, namely, intrinsic features and intentional ones. Qualia are intentional features of experiences, not intrinsic ones. The fact that an experience has a certain intentional content is explained by the fact that it has a certain function. Therefore qualia as the intrinsic features of an experience are reduced to its function.2. Conscious experiences are conscious because they have qualia. But, if qualia are nothing but intentional features, experiences are not necessarily conscious. For there can be unconscious experiences which have the same intentional features. Qualia must be conscious intentional features. The fact that qualia are conscious can be explained by the fact that they are available to verbalization. Language is the source of consciousness. Organisms without any language lack consciousness.3. Conscious thinking is classical computationalistic because it is a linguistic process. On the other hand, unconscious thinking is connectionisitic. It is a process of transforming activation patterns of neurons. For example, judgements of good and evil or of grammaticalilty are classical computaionalistic in the case of being made consciously, but connectionistic in the case of being made intuitively. But, even in the case of conscious inference, everyday inference is usually full of logical gap. Strictly speaking, it is not classical computationalisitic.