Abstract:

While computational models are an invaluable tool in understanding how people behave, there is a problem that they all share. Namely, since human behavior is almost entirely the consequence of some kind of constructed understanding, and we have no formal model(s) of how such understanding is created, we are always missing something. All models are wrong, and so whether this matters will depend on what we're trying to learn from modelling. So, after first outlining the argument for the above, we will discuss its consequences for various aspects of the psychological literature and the phenomena we are trying to explain.