Two Threats To Representation

I consider two threats to the idea that on-line intelligent behaviour (the production offluid and adaptable responses to ongoing sensory input) must or should be explainedby appeal to neurally located representations. The first of these threats occurs whenextra-neural factors account for the kind of behavioural richness and flexibility normallyassociated with representation-based control. I show how this anti-representationalchallenge can be met, if we apply the thought that, to be a representational system,an action-oriented neural system must not only be the source of at least some of theobserved behavioural richness and flexibility, it must also feature two architecturaltraits, namely arbitrariness and homuncularity. Unfortunately, however, this solutionopens the door to our second threat to representation. The homuncularity conditionwill not be met by any system in which the causal contribution of each component ismassively context-sensitive and variable over time. I end by discussing the empiricalbet that biological nervous systems will not exhibit this style of causation.