In recent studies, bimodal Visual-Auditory (VA) spatial localization precision as shown to exceed those of the most precise unimodal condition i.e. vision, in accordance with Bayesian integration models such as the Maximum Likelihood Estimate. The novelty of the present study is:
a)To provide for the first time two-dimensional data about bimodal AV integration, for which results suggest the need for multidimensional data analysis (and potentially vectorial) to account for the non-independency of the responses along the two Cartesian dimensions.
b)To take into account the idiosyncratic differences in precision between A and V as a function of Eccentricity and Direction to predict accuracy assuming validity of the Modality precision (MP) model
Results confirmed near optimal integration, localization precision for the AV targets being greater than that of the more precise modality V, while accuracy for the bimodal stimulus tended to be a compromise between the values of individual modalities in favor of vision. The lack of complete integration is discussed in terms of methodological reduction.