Minds, modules, and meta-analysis.

Wellman and colleagues' meta-analysis of performance on the false-belief task is methodologically useful, but it does not lead to any theoretical progress concerning the nature of the mechanisms that underlie the existence and development of "theory of mind." In particular, the results of this meta-analysis are perfectly compatible with "early competence" accounts that posit a specific, innate, and possibly modular basis for theory of mind. The arguments presented by Wellman and colleagues against such views stem not from their meta-analytic data, but from mistaken assumptions regarding the requirements of such theories (e.g., that there exist manipulations that improve performance only, or to a greater degree, in young children). Contrary to what Wellman and colleagues claim, their meta-analysis, while consistent with conceptual change, does not lend any new support for such theories.