Date of Award
Master of Arts in Psychology
This experiment was designed to investigate the mechanisms underlying perceptual set effect. It was hypothesized that a change in synaptic resistance due to previous experience is the operational mechanism underlying perceptual set effect. An associative memory model was used to simulate a simple neural circuit. Experience of the model was manipulated by exposure to a set of unambiguous stimuli. A classification task involving ambiguous stimuli was used to measure perceptual set effect.
Two independent variables were manipulated. The first involved the structure of the stimuli as measured by the Hamming Distance. The second involved the learning rate variable of the model and was used to assess the effects of manipulating key variables affecting the model's behavior.
The experiment indicated that perceptual set effect could be successfully induced in the model through previous exposure to unambiguous stimuli. In addition, two other interesting behaviors were noted. First, the model showed remarkably different behaviors at different values of learning rate. At high learning rates, the model tended to classify an ambiguous stimulus in accordance with an internally-constructed representation of the unambiguous stimulus. Second, the model experienced a hysteresis effect at high levels of learning rate.
These findings indicate that associative memory systems may be useful tools in the investigation of certain human perceptual phenomena.
Skrmetti, Thomas J., "Inducing Perceptual Set Effects in Associative Memory Systems" (1989). Open Access Master's Theses. Paper 2299.