The objective of the project is to study the use of the relational cues """"""""same"""""""" and """"""""different"""""""" in avians. The principal task that will be used combines matching-to-sample and oddity-from-sample procedures into a single match/mismatch task (Edwards, Miller, & Zentall, 1985). This task requires birds to match sample and comparison stimuli on some trials and to mismatch those same samples and comparisons on other trials, depending on the value of a spatially distinct, higher-order conditional cue (match/mismatch instructions). Initial experiments will examine the factors that may enhance control by the instructional cues. Subsequent experiments will use transfer manipulations to examine instructional (i.e., discriminative) control of generalized match/mismatch behavior. The manipulation of instructional stimuli in training and transfer tests will provide the appropriate control conditions for assessing same/different relational responding, controls that are lacking in existing studies(see discussions by Mackintosh, 1984; Wilson, Mackintosh, & Boakes, 1985). The long term purpose of the proposed research is to bridge the empirical and theoretical gap between relational phenomena in non-primate species and relational phenomena in humans, and thereby strengthen the animal model of this particular higher-order cognitive process.