Date of Award:


Document Type:


Degree Name:

Master of Science (MS)


Animal, Dairy, and Veterinary Sciences


Clive W. Arave


Eight heifers were trained using operant conditioning to press a plate to receive a feed reward. Different wavelengths of light were presented as correct and incorrect stimuli. Positive and negative responses to the stimuli were registered electronically. Daily sessions of 17 minutes were conducted in a chamber with external light being excluded.

The duration of the stimulus was fixed at 17 seconds after which stimuli were randomly presented. Only presses on the plate when the correct stimulus was presented were reinforced with feed. A 75% correct choice was the criterion used as acceptable discrimination.

Ratios of correct to incorrect responses were computed. A stability of response was judged to occur when the median of these ratios over 5 days did not differ by more than .05 from the median of the ratios from the previous 5 sessions. Three colors i.e. green {535nm), red {610nm), and blue {450nm) were compared pairwise during 8 trials. Trial 7 was a repeat trial of green vs red and trial 8 was a comparison of green vs green.

Heifers gave random response to green vs green. Red was distinguished from blue by five of the heifers: 1, 2, 3, 4, and 5 at 76%, 91%, 78%, 88%, and 81% correct choice respectively. Blue was distinguished from green by three of the heifers: 1, 2, and 5 at 89%, 88%, and 85% correct choice respectively. Green was distinguished from red by three of the heifers: 1, 5, and 7 at 90%, 84%, and 85% correct choice respectively. These last discriminations we r e made in the repeat trial of green vs red after heifers failed to do so in the first trial of green vs red. Color discrimination and discrimination learning have been demonstrated by these results.