Tests of the ratio rule in categorization
Article
English
OPEN
Wills, A.J.
;
Reimers, Stian
;
Stewart, Neil
;
Suret, Mark
;
McLaren, I.P.L
(2000)
Many theories of learning and memory (e.g. connectionist, associative, rational, exemplarbased) produce psychological magnitude terms as output (i.e. numbers\ud representing the momentary level of some subjective property). Many theories assume that these numbers may be translated into choice probabilities via the Ratio Rule, a.k.a. the Choice Axiom (Luce, 1959) or the ConstantRatio Rule (Clarke, 1957). We present two categorization experiments employing artificial, visual, prototypestructured stimuli constructed from twelve symbols positioned on a grid. The Ratio Rule is shown to be\ud incorrect for these experiments, given the assumption that the magnitude terms for each category are univariate functions of the number of categoryappropriate symbols\ud contained in the presented stimulus. A connectionist winnertakeall model of categorical decision (Wills & McLaren, 1997) is shown to account for our data given the same\ud assumption. The central feature underlying the success of this model is the assumption that categorical decisions are based on a Thurstonian choice process (Thurstone, 1927,\ud Case V) whose noise distribution is not double exponential in form.

Metrics
0
views in local repository
41
downloads in local repository
The information is available from the following content providers: