Chapter 9 Consciousness , Decision-Making and Neural Computation

Edmund T. Rolls
2011-01-01
Abstract:Computational processes that are closely related to conscious processing and reasoning are described. Evidence is reviewed that there are two routes to action, the explicit, conscious, one involving reasoning, and an implicit, unconscious route for well-learned actions to obtain goals. Then a higher order syntactic thought (HOST) computational theory of consciousness is described. It is argued that the adaptive value of higher order syntactic thoughts is to solve the credit assignment problem that arises if a multi-step syntactic plan needs to be corrected. It is then suggested that it feels like something to be an organism that can think about its own linguistic and semantically based thoughts. It is suggested that qualia, raw sensory and emotional feels, arise secondarily to having evolved such a HOST processing system, and that sensory and emotional processing feels like something because it would be unparsimonious for it to enter the planning, HOST system and not feel like something. Neurally plausible models of decision-making are described, which are based on noise-driven and therefore probabilistic integrate-and-fire attractor neural networks, and it is proposed that networks of this type are involved when decisions are made between the explicit and implicit routes to action. This raises interesting issues about free will. It has been argued that the confidence one has in one’s decisions provides an objective measure of awareness, but it is shown that two coupled attractor networks can account for decisions based on confidence estimates from previous decisions. In analyses of the implementation of consciousness, it is shown that the threshold for access to the consciousness system is higher than that for producing behavioural responses. The adaptive value of this may be that the systems in the brain that implement the type of information processing involved in conscious thoughts are not interrupted by small signals that could be noise in sensory pathways. Then oscillations are argued to not be a necessary part of the implementation of consciousness in the brain. E.T. Rolls ( ) Oxford Centre for Computational Neuroscience, Oxford, UK e-mail: Edmund.Rolls@oxcns.org V. Cutsuridis et al. (eds.), Perception-Action Cycle: Models, Architectures, and Hardware, Springer Series in Cognitive and Neural Systems 1, DOI 10.1007/978-1-4419-1452-1 9, c Springer Science+Business Media, LLC 2011 287
What problem does this paper attempt to address?