In a previous study (Pierrel, 1958) a gradient of generalization was obtained for rats following prolonged training in an auditory intensity discrimination. In that situation, SD and SA were at opposite ends of a 40-decibel intensity continuum. When generalization stimulus intensities intermediate between SD and SA were presented, a concave function was obtained relating response rate to stimulus intensity. The maximum of this function occurred at the SD intensity and the minimum at the SA intensity.Hanson (1959), using a nonintensive continuum, obtained generalization gradients showing a maximum displaced from SD in a direction away from 5A. This would suggest the desirability of placing SD and SA away from the extremes of the continuum explored in order to get at certain other relationships. In the current study, we proposed to examine generalization rate relations within the same range of intensities employed in our previous study. However, in this case, SD and St were situated only 20 decibels apart and generalization stimuli were presented at both extremes of the intensity continuum from SD and &5, as well as intermediate between them. By means of a recent pilot study, it was established that, within this intensity range, rats can develop stable discrimination behavior to stimuli separated by a 5-decibel difference. It was therefore decided to obtain a generalization gradient employing stimuli spaced at 5-decibel steps across a 40-decibel range. METHODThe subjects for this experiment were eight male albino rats of the Sprague-Dawley strain. These animals were obtained from the Charles River Breeding Houses, Brookline, Massachusetts, and were 90 to 110 days old at the start of experimentation.Four identically constructed operant spaces were used. One of these is shown in Fig. 1. The box on the right is shown with the door open. The outer shell of the animal enclosure consisted of a 12.8-cubic foot refrigerator shell equipped with shock mounts. The inner surfaces, including the door, were lined with 2-inch Fiberglas board covered with glass cloth. The floor of the box was built up with Fiberglas blocks and a slotted partition inserted between the animal chamber and the space enclosing the pellet dispenser and the motor related to the retractable bar. This reduced the spread of noise from equipment operation to the animal. The resulting dimensions of the animal chamber were 15 inches in depth, 13.5 inches in width, and 23.5 inches in height. A University model 4401 tweeter was shockmounted through the Fiberglas onto the inner shell of the refrigerator, with its base 7.5 inches above the experimental cage. The experimental cage was 5 inches deep, 8 inches wide, and 5.75 inches high. It was constructed from 0.1875-inch-diameter stainless steel rodding, with a separation of 0.5 inch between the rods. The cage frame was made of Lucite, and the entire front of the cage was a hinged door with a water bottle held onto it by a spring. The motor on the right side of the partition was programmed to either extend the bar...

The acquisition of auditory intensity discriminations in rats trained on multiple variableinterval extinction schedules was studied as a function of some of the variables that contribute to the speed of development of differential responding and the final level attained. The effects of three variables were isolated and studied in detail: (1) the decibel difference between the discriminative stimuli (intensity difference); (2) the intensity relationship between the stimuli (relative intensity); and (3) the position of the stimuli on the intensity continuum (absolute intensity). Each of the three variables generated orderly relationships and interacted with one another to produce complex effects upon differential responding.The present investigations examined the effects of three aspects of stimulus intensity upon the development of an auditory discrimination. Comparable experimental conditions obtained in each of the studies, excepting the manipulation of the variable under investigation.After Pavlov developed the "method of contrasts", many procedures, devices, and response measures were invented to explore the stimulus control of differential responding (see Warden, Jenkins, and Warner, 1935;Woodworth and Schlosberg, 1954). It would appear that on occasion the technique itself has nIot only been a tool for research but has also defined the problems to be studied. This early literature contains generous amounts of information, much of which is difficult to interpret or impossible to integrate, and it sometimes produced irrelevant feuds arising out of the lack of standardized technique.In

No abstract