Part I: Representations

Matej Kohár




Proponents of 4E (embodied, embedded, extended, enactive) approaches to cognition have often questioned the explanatory role of mental and neural representations in cognitive science. Reconceiving cognition as dynamical coupling and skillful engagement with the environment obviates the need for positing content-bearing neural states. However there is a subset of cognitive achievements known as representation-hungry problems, which cannot be straightforwardly understood in terms of skilful interaction with the environment. These problems have to do with adapting the cognitive system’s behaviour to distal or non-actual states of affairs. Talking about unicorns, or looking for a misplaced pair of glasses are examples of representation-hungry problems. Since the glasses and unicorns are not parts of the cognitive system’s immediate environment, they cannot be directly interacted with. The received view dictates that mental/neural representations which stand in for these objects are necessary.

In my project I investigate the possibility of solving representation-hungry problems without involving content-bearing neural representations. The new-mechanistic view of explanation, which has gained popularity as an account of explanation in biology and cognitive science in recent decades provides the tools for a satisfactory solution of representation-hungry problems. It is also, so I argue, incompatible withany viable representational theory of cognition. Importantly, the new-mechanistic framework does not imply that the boundaries of cognitive mechanisms line up with organismic boundaries. This opens up the possibility of including extraneural entities in cognitive mechanisms and using the tools of ecological psychology and dynamical systems theory to make the explanations more tractable.