The Image of What Happens
Where I ended the last posting (I hope) was where the evolving organism began to develop a central nervous system. Here I want to pick up from this point to talk about how representation might operate within this developing CNS, and specifically what is represented. Where I am going with this narrative is in the direction of an account of how these primary developments and mechanisms form the basis for the more complex processes we call ‘thought’. I then want to suggest that the structure of that thought, the ‘poetics’ as I will be calling it, is grounded in the very basic processes indicated here.
In the last posting I talked about the relationship between moving and ‘thinking’, citing particularly Llinas claim that ‘what we call thinking is the evolutionary internalisation of movement’, and relating this to the notion of representations and their role in the the life of an organism. An organism may evolve to a level of complexity where it is able not only able to respond reactively to its environment but, through the construction of internal representations of that world, can be proactive. Such an organism can use representations as the basis for prediction about what might happen if it were to engage in certain actions, in a sense it can run small simulations of possible actions through its representation of the world and tailor its actions in the real world according to the results of these simulations. Its ability to construct representations might even increase to a point where it is able to represent itself as a being in that world in a stable and consistant way. Before we go there though it might be important to get a clearer sense of what these representation are and how they operate.
What is immediately obvious is that it is not the outside world (whatever that means) that is represented but the sensations or perceptions associated with that world. To use the human organism as an example, we don’t ‘see the world’, rather ‘seeing’ is what we call the interaction between the world and the visual parts of our cognition. Our eyes are not simply portholes which open out onto an unproblematically ‘visual’ world; the world is seen as it is because we have the kind of eyes we have, and to describe the world as being inherently visualisable, that it is visible because we can see it, is obviously tautological. It is also incorrect to assume that sense organs are simply passive instruments recording data in particular ways but otherwise not engaging or interacting with that data. Sensation and perception emerge not from the sensory organs themselves but from integrated functioning of the somatosensory system, which includes not only the sensory input but also the motor output. When we feel a surface as ‘rough’ not only are we not identifying a property sovereign to the surface, we are also not identifying the passive sensing of inert fingertips. Feeling a surface is experiencing the movements that the fingertips are making across that surface. It is feeling the movements of the skin as it catches and slips from one tiny indentation to another, and feeling the squeezing of the tissues as the pressure on the cells of that skin varies with changes in the texture of the surface. To take another example, seeing the edge of an object is not the transparent photographic recording of real edges in a naively real world, nor are edges an artefact of the sensitive cells in the retina on which the light falls. Seeing an edge involves the eye saccading back and forth across a few seconds of arc and this motion being coordinated with that pattern of changing light on the retina. Seeing involves this kind of combination of both sensory and motor information and it is this sensorimotor synthesis which forms the representations. In other words, cognitive representations are not pictures of the world, they are pictures of what happens in the sensorimotor system in relation to the world. This concept that perception and action are intimately related in the formation of representations was well cover by the neurologist Roger Sperry. He referred to the process which allows for the internalisation of movement as thought as ‘perception-action coupling’ (Sperry, 1952). More recently, in his book ‘Motor Cognition: what actions tell the self’ (2006) Marc Jeannerod refers to this picturing of sensorimotor sytheses as ‘action representation’, and such representations form a significant part of his description of cognitive functioning.
If I could speculate briefly on where this concept of action representation might show itself in modern human experience, it is possible that it is this kind of action representation which, when presented to consciousness, forms the basis of emotions such as fear, joy, anger, and love. These feelings were described by William James as “the perception of bodily changes as they occur” (1950) and it seems likely that these coordinated bodily changes are also action representations. This idea that emotions can be understood as the conscious experiencing of sensorimotor schema has been revivified recently because of findings in neuroscience which lend support to this kind of process. Antonio Damasio (1999) has been particularly influential in this area, memorably referring to the way in which somatosensory activity becomes the representational content of human experience as ‘the feeling of what happens’.
This concept that perception and action are intimately related in the formation of representations was well cover by the neurologist Roger Sperry. He referred to the process which allows for the internalisation of movement as thought as ‘perception-action coupling’ (Sperry, 1952). More recently, in his book ‘Motor Cognition: what actions tell the self’ (2006) Marc Jeannerod refers to this picturing of sensorimotor sytheses as ‘action representation’, and such representations form a significant part of his description of cognitive functioning.
I think it is significant to note that there is some evidence that such representations not only instantiate the cognitive connection between perceiving and actual performed action, they also persist when no actual action is carried out. As Marc Jeannerod puts it, “(t)he core of motor cognition is the concept of action representation, the covert counterpart of any goal-directed action, executed or not” (Jeannerod, 2006: 165).
In other words the representation may not appear on the surface of the body at all, existing only in the form of simulated action within the cognitive system of the organism. Complementary to this, representation may also occur in the absence of any immediate percepion; in a modern human sense we might call this ‘imagining’ or ‘remembering’. This has been described as the cognitive system operating ‘off-line’ such that the input contributing to the sensorimotor synthesis comes not from the senses but from memory. As Diane Pecher and Rolf Zwaan express it in their introduction to Grounding Cognition (2005) “offline cognition occurs when sensorimotor functions are decoupled from the immediate environment” (2005:1).
This last point, that representations can exist within cognition in the absence of either sensory input or motor output is something I will be coming back to later.
Damasio, Antonio R. (1999) The Feeling of What Happens: Body and Emotion in the Making of Consciousness. New York: Harcourt Brace.
James, W. (1950) What is an Emotion, in The Principles of Psychology, New York, Dover (first published 1890).
Jeannerod, M. (2006) Motor cognition: what actions tell the self. Oxford University Press, Oxford.
Pecher, Diane and Rolf A. Zwaan (eds.) (2005) The Role of Perception and Action in Memory, Language, and Thinking. Cambridge University Press, Cambridge.
Sperry, R.W. (1952). Neurology and the mind-body problem. American Scientist, 40, 291-312.