The function(s) of consciousness revisited: insights from a modular/constitutive model using vision as a test case

This paper extends an earlier analysis of the functions attributable to consciousness, applying it to a modular/constitutive model that limits the role conscious sensations are allowed to play in integrative processes operating at scale. The focus is on conscious vision because of the challenge it poses for any theory of consciousness, of how to account for a 2-dimensional perceptual display. For a modular/constitutive model, what is required is that awareness of visual stimuli has the property of position-dependence, meaning that the physical location of each module contributing to the visual experience has perceptual consequences. Conscious gaze control can then act as a device for registering salient visual features in a way that has no preconscious counterpart, while contributing also to the process by which visual experience acquires noetic content. The analysis provides (1) theoretical justification for Merker's ideas on vision and gaze control as a way of understanding the core functions consciousness performs for our species, where the post-natal period during which visual skills are refined equates to the learning processes through which agency is acquired, and (2) a framework for thinking about the hard problem and the adaptive advantage conscious vision confers over all possible zombie alternatives. This would imply that conscious experience for species that rely either less than we do on vision or not at all may nevertheless incorporate a position-dependent component, which for those species either could or would depend on something other than light perception. Hence the argument is not that conscious vision was necessarily a component of consciousness at it first evolved, but that the contrasting roles position-dependent contents play in behavioral control compared with those conferring valence requires that both be present if the adaptive advantage of an emergent consciousness is to depend on its ability to assign meaning to experience. This leaves by default a model of proximate cause where it is the combined advantages of conscious attention control and memory encoding that makes consciousness adaptive, and a minimalist conception of the functions performed by consciousness if it is not a direct participant in large-scale integrative processes.