Professor Harald Lachnit, School of Psychology, University of Marburg, Germany will give a special school seminar on Tuesday, July 4th, from 4:00 - 5:00 in Rm 302, Psychology Building entitled: Empirical evaluation of the predictions of five associative theories with three discrimination problems in three paradigms (two species). A post seminar will be held in the staff club.
We compared the predictions of elemental (unique cue, modified unique cue, replaced elements model) and configural associative theories (Pearce's theory and a modification of it) for three modified patterning problems initially suggested by Redhead and Pearce (1995). We used predictive learning and eyelid conditioning in humans as well as conditioning of the proboscis extension reflex in honeybees. Both configural theories, the replaced elements model, and the unique cue approach failed to account for our empirical observations. Only a modified unique cue approach successfully accounted for all of our data by assuming that the salience of stimuli decreases with an increasing number of stimuli in a compound. The success of this idea asks for explicitly taking into account attentional factors in associative learning theory.