A note on interresponse time distributions during generalization testing

Abstract

Following variable interval training a pigeon was reinforced for responding to S+ (1099 cps tone) but was not reinforced for responding during S-(no tone). When discrimination was stable test stimuli were introduced at preselected (formerly S-) periods. Interresponse time (IRT) distributions were obtained during presentation of stimuli. Minimal changes occurred in the IRT distributions for test stimuli relative to those for S+. Problem Jenkins & Harrison (1960) studied the effects of differential and nondifferential training on auditory generalization in the pigeon finding that steeper gradients were obtained following differential training. Guttman & Kalish (1956) found in studying wavelength gradients that as the test stimulus values deviated more from the training stimulus pigeons generally responded more slowly. What does responding more slowly mean? In steady state generalization where more reinforcements are received at 550 millimicrons than at other wavelengths "short IRT peaks" seem insensitive to stimulus variation. Fewer intermediate IRT's occur at or near 550 millimicrons than elsewhere (Blough, 1963). The purpose of this study consisted in obtaining separate IRT distributions for responses emitted to test stimuli so that response patterning as well as rate changes could be investigated. Use of a probe technique in generalization testing results in at least two advantages: (1) Interaction of test tones via serial effects on responding may be evaluated. (2) Total number of reinforcements possible during S+ presentation is not attenuated making interaction effects of extinction less likely (Skinner, 1950). Method The S was one experimentally naive white Carneaux pigeon maintained at 80% of its free feeding weight. A standard one cubic foot experimental chamber containing a one key response panel was used. Chamber and response key illumination was provided by white and green 7-watt lamps, respectively. Access to grain for 4.5 sec. was accompanied by magazine light onset and extinction of the chamber and key lamps. Standard programming equipment was used. Pure tones were delivered through a ceiling mounted speaker at 80 db amplification. Twenty-five sessions of variable interval VI 30-sec. and VI 1-min. schedules of reinforcement in the presence of a 1099 cps tone preceded discrimination training. From session 26 responding was reinforced on a VI 1 schedule in the presence of 1099 cps tone (S+) and not reinforced in the absence of tone (S-). Working periods ,

Postdiscrimination generalization as a function of interresponse time

“…Blough (1963) concludes that "the gradient is produced mainly by a shift upward of the longer IRTs which leave the time of the short peaks relatively unaffected. It appears, as might be expected, that stimulus control is exerted over the responses following long IRTs: the thing that controls the occurrence of a Sewell and Kendall (1965) present similar data for a single pigeon tested along the dimension of auditory frequency. Again, it appeared that external stimulus control is exerted only over responses following long IRTs.…”

“…1969) and Sewell and Kendall (1965), pigeons were trained to respond -in the presence of a single stimulus value before testing for generalization. In the present experiment.…”

RESPONSE PATTERNING DURING STIMULUS GENERALIZATION IN THE RAT1

“…Both Blough (1963) and Sewell and Kendall (1965) have reported essentially the same finding, i.e., the response pattern remains remarkably uniform with most responses occurring in the same narrowly defined IRT category despite large differences in rate of responding in the presence of the various generalization test stimuli.…”

“…In analyzing IRTs in generalization with pigeons, Blough (1963) The studies by Blough (1963) andSewell andKendall (1965) employed the key-pecking response in pigeons, single stimulus training, and a visual stimulus which was then varied in extinction. To assess the generality of their findings, the present study utilized the barpressing response in rats, auditory rather than visual stimuli in a discrimination training procedure, and generalization testing under the condition of continued reinforcement of responses in the presence of the positive stimUlUS (SD).…”

LOCAL PATTERNS OF RESPONDING MAINTAINED BY CONCURRENT AND MULTIPLE SCHEDULES1

“…Interresponse times in the neighborhood of the modal IRT define a local pattern of responding (Blough, 1963;Crites, Harris, Rosenquist, and Thomas, 1967;Sewell and Kendall, 1965). Interresponse times longer than 4 sec are not plotted in Figure 2, as these longer IRTs presumably are not descriptive of the local pattern of responding.…”