Retroperception: Problems for Enactivism?

Yes, it’s consciousness again. Or rather, perceptual experience. This time, some musings about what the newly-minted cognitivist phenomenon of retroperception (see this Mindhacks post for a summary) might mean for enactivist theories of perceptual experience.

Before we begin, let us take note of a critique levelled against cognitivism by the enactivist Thompson (2007) - that the representationalist, information processing account has replaced the mind/body problem with a new mind/mind problem. This is to say, how and why does hidden cognitive processing, which goes on ‘in the dark’, yield conscious experience, and how can we take seriously a body of thought which so often seems to end up pushing our experience of the world towards an epiphenomenal status? Contrary to representationalist accounts, enactivism denies that human beings experience an ‘external’ world indirectly via some kind of 'internally constructed' model and suggests instead that, through our sensorimotor systems, we have a form of direct access that constitutes the ‘bringing forth’ of a subjective, experiential world. Therefore, like enactivism's forerunner phenomenology, the description of our experiential world could be seen as enactivism’s prime concern. This must include explaining the perceptual quirks that have been comprehensively documented by cognitive psychologists. In their influential 2001 article, Noe and O’Reagan do a good job of explaining various peculiar facets of visual experience within their own theoretical framework. 

What, then, would Thompson or Noe and O’Reagan make of the newly documented phenomenon of retroperception? According to this paper by Claire Sergent et al, experimental participants perceived a pattern when their attention was directed to the area as much as 400 milliseconds after the pattern had in fact disappeared. The implications of this are that participants consciously perceived what was no longer there, and from this, that we (within certain very limited temporal parameters) routinely perceive what is no longer there. As Mindhacks puts it, ‘This suggests that consciousness isn’t ‘filtered’ sensory information, but an active ‘conclusion’ drawn from information distributed across senses, space and time.’ This is reflected by the paper’s own theoretical assumptions, namely a Baars-style cognitivist account of consciousness as a system for sharing information across agencies within the brain. 

If Sergent’s experiment is valid, how can an enactivist account for conscious experience if that experience is not the product of what the organism encounters ‘right now’ but rather a synthesis of what the organism has experienced across the last few hundred milliseconds? If class questions are anything to go by, the Achilles’ heel of enactive theories of perception would seem to be in accounting for visual experiences which occur in the absence of ‘the thing itself’ – for example dreams, vivid memories, extreme hallucinations such as those that occur in Charles Bonnett Syndrome. 

It could be said that these are aberrant or unusual phenomena. It can always be argued that the ‘main thing’ is ordinary, wakeful, well-adjusted conscious experience, and enactivism accounts well for most of that (while also having a cursory stab at the other stuff too - I’m thinking here of Noe and O’Reagan’s argument that the transient, shifting nature of dreamscapes indicates the anchoring necessity of the world as its own best model. But still, how can there be anything at all in dreams, even if what is there does fail to achieve stability? Is it simply the harp randomly plucking its own sensorimotor strings?). However, if retroperception (and thus the temporally ‘synthesised’ nature even of everyday conscious experience, and presence within it of the 'no-longer-there' ) is confirmed by further testing, what explanation could an enactivist offer? Would retroperception not indicate that the cognitivist ‘mind/mind’ problem is a real one and not simply a theoretical artefact, and that there is some kind of elaborate construction going on behind the scenes?

Perhaps the place to begin looking for an answer would be to ask what it is Sergent takes to be ‘conscious perception’. Mindhacks might be too quick to use that magic word ‘consciousness’in the description above. From the paper, there is no indication that participants ‘experienced’ the stimulus when attention was post-cued to it; rather, similar to blindsight, they more successfully reported the nature of the stimulus when given the attentional post-cues without actually reporting to experience the stimulus. However, surely this still poses a problem for enactivists: if an organism’s perceptions are directly experienced rather than providing fodder for some internal model, how can there also be sensory detection without experience?