This paper presents the optimum decision rule for an m-interval oddity
task in which m-1 intervals contain the same signal and one is differ
ent or odd. The optimum decision rule depends on the degree of correla
tion among observations. The present approach unifies the different st
rategies that occur with ''roved'' or ''fixed'' experiments (Macmillan
& Creelman, 1991, p. 147). It is shown that the commonly used decisio
n rule for an m-interval oddity task corresponds to the special case o
f highly correlated observations. However, as is also true for the sam
e-different paradigm, there exists a different optimum decision rule w
hen the observations are independent. The relation between the probabi
lity of a correct response and d' is derived for the three-interval od
dity task. Tables are presented of this relation for the three-, four-
, and five-interval oddity task. Finally, an experimental method is pr
oposed that allows one to determine the decision rule used by the obse
rver in an oddity experiment.