Infusing Perception with Imagination
Derek H. Brown
Imagination is involved in most if not all of our perceptions.1 One could sustain this thesis by arguing that most perceptions are cognitively penetrable and that this cognitive influence involves the imagination.2 Alternatively, perhaps at a more basic (i.e., less cognitive) level perception often involves ‘elements’ which are present to us in experience but not ‘given’ to us by the world. Such elements, on this view, are created by our perceptual systems using its capacity for imagination. What is phenomenally present as absent [PPA] is one candidate for such an imaginative contribution.3 My interest is in a different one, the equally robust capacity for our vision systems to turn the thoroughly ambiguous information the world provides us with into determinate perceptual contents.
I examine some ambiguous stimuli (e.g., a wire cube, the floating man illusion) to construct a model by which one can describe both the ‘given’ ambiguous information and its relation to the determinate contents our states have when we perceive such stimuli.4 I then argue that the transition from this partially determinate ‘given’ information to the more fully determinate perceptual contents is plausibly achieved by a creative mental process that is aptly called ‘imagination’. This imaginative contribution can in turn explain some processes occurring at early stages of shape and colour perceptions, in particular those yielding perceptual constancies.5 Given that such processes are involved in most if not all of our perceptions, it follows that so too is imagination, and in a basic, non-trivial way.6
Finally, I contrast this contribution of imagination with the relevant PPA contributions. Using the floating man illusion as a guide [see attached], the thought is as follows. PPA is involved in both perceptions of this stimulus as a floating man and ones of it as a man standing behind a wet spot, for both perceptions involve, for example, experiencing the ‘man elements’ to be a man as opposed to merely the surface of a man. However, this imaginative contribution is not what is needed to explain the disambiguation performed by our vision systems. Instead, such an explanation requires, for example, a mechanism whereby the vision system can interpret the information it receives regarding the distance between the man’s feet and the dark patch to have a vertical component, and alternatively to have a purely horizontal one. In the former case the dark patch is under the man’s feet, in the latter it is in front of them. We can describe this mechanism as consisting of the vision system making a creative judgement about how these informational bits are actually related to one another. Importantly, such mechanisms can in theory be applied by a vision system which interprets the scene to be composed of mere surfaces: the difference between seeing the dark patch as being under the ‘feet elements’ or in front of them does not hinge on whether these elements are seen as mere surfaces or as actual three-dimensional objects. Therefore this disambiguating mechanism is plausibly not an utilization of our vision system’s capacity for PAA contributions, but instead of an equally-basic but distinct capacity to interpret or represent the relations between what is phenomenally present. And its creative characteristic makes it plausibly an imaginative contribution.
1 This idea has been defended by Kant, Merleau Ponty and P. F. Strawson among others. I. Kant, Critique of Pure Reason, A120. M. Merleau Ponty, The Phenomenology of Perception. P. Strawson (1970), Imagination and Perception, in L. Foster & J. Swanson, eds., Experience and Theory, pp. 31-54: Univ. Mass. Press. Here are two recent examples. M. Pendlebury (1996), The role of imagination in Perception, South African Journal of Philosophy, 15 (4), 133-47. K. Lennon (2010), Re-enchanting the world: The role of imagination and perception, Philosophy, 85, doi: 10.1017/S0031819110000239.
2 I think this is a fair reading of Strawson (1970). More recently, although Macpherson doesn’t argue that most perceptions are cognitively penetrable, she does argue that some are, and suggests that this influence is achieved by utilizing the imagination. One could thus use her contribution as a base camp from which to argue for the ubiquity of cognitive penetration and the more general conclusion that most perceptions involve imagination. See her (forthcoming), Cognitive penetration of colour experience: Rethinking the issue in light of an indirect mechanism. Philosophy and Phenomenological Research.
3 E.g., the fact that we often see partially occluded objects to be ‘whole’ as opposed to ‘cut-up’ at the points of occlusion; the fact that we often see what is perceived as consisting of whole three-dimensional objects instead of as merely the surfaces that are presented to us by the world, etc.
4 My guide here is Gupta’s thesis of the functionality of the given. See his (2006), Empiricism and Experience: Oxford University Press. Also see Pendlebury (1996).
5 It is reasonable to hold that perceptual constancies are experienced by normal humans from fairly early stages of development, and by many non-human animals. Such experiences are thus suitably encapsulated from cognitive penetration to be deemed part of ‘early vision’ in Pylyshyn’s sense. However, note that my personal view is that even these are cognitively penetrable by adult humans.
6 I will likely not have time to address how this sense of ‘imagine’ fits with, e.g., the sense utilized when we ignore our occurrent perceptions and try to ‘imagine’ something, although Macpherson’s (forthcoming) ‘indirect mechanism’ is an intriguing attempt to bridge this gap.