A new theoretical approach to stimulus identification is proposed through a
probabilistic multidimensional model based on the maximum information entr
opy principle. The approach enables us to derive the multidimensional scali
ng (MDS) choice model, without appealing to Luce's choice rule and without
defining a similarity function. It also clarifies the relationship between
the MDS choice model and the optimal version of the identification model ba
sed on Ashby's general recognition theory; it is shown theoretically that t
he identification model derived from the new approach includes these two mo
dels as special cases. Finally, as an application of our approach, a model
of similarity judgement is proposed and compared with Ashby's extended simi
larity model. (C) 2001 Academic press.