How Images of the Brain can Constrain Cognitive Theory: the Case of Numerical Cognition

We have been invited by Max Coltheart to play a game. At first sight the game looks easy and the rules are quite clear. First, we have to identify a conflict between two competing psychological theories and second, we have to find at least one neuroimaging experiment that allows selecting one of the two theories as the good one. The game is not about neuroimaging per se since the use of neuroimaging for localisation is, explicitly, not discussed. Rather, the question pertains to the relevance of neuroimaging for cognitive theories. Curiously, Coltheart indicates that his question is not a theoretical one and he emphasizes that he is not interested in the “in-principle question” but in an “in-practice so-far” question. However, this restriction seems mainly rhetorical, since Coltheart underlines that “given the enormous volume of published recent empirical work in this area, if it turns out that none of this work can be used to distinguish between competing psychological theories, the in-principle question of whether cognitive neuroimaging data can ever serve this function will deserve much more attention than it has so far been given”. In other words, what Coltheart suggests is that if after so much work the in-practice question will result in a negative answer then one has to come back to the in-principle question to examine whether this failure isn’t due to some intrinsic theoretical obstacle.