Direct estimation of multidimensional perceptual distributions: assessing hue and form.

The procedures developed to assess the perceptual and decisional processes associated with detection in multidimensional space all require specialized statistical skills and analysis programs. The present article describes a regression model, designed to assess dimensional interactions, that is both computationally simpler and more accessible than those procedures. The papervalidates the regression model by comparing the perceptual space associated with the detection of hue and form mapped by the regression model with that mapped by Kadlec and Townsend's (1992a, 1992b) macro- and microanalyses. The results of both analyses showed that hue strongly influences the perception of form but that form only weakly influences the perception of hue. The parallel results of the two analyses suggest that the regression model is a valid alternative to multidimensional signal detection theory analysis.