Extended training and multiple shifts: Percentage of reward

Abstract

This is one paper of a series investigating the effects of extended training and multiple succesSve reward shifts. Four groups were run in a straight alley. The control groups received either 75% or 25% reinforcement throughout training. Each experimental group was given 40 acquisition trials at either 75% or 25% reinforcement followed by three shifts of 24 trials each. All groups were given 24 extinction trials. There were initial significant differences betMen the 75% and 25% groups. These differences disappeared after about 40 trials. The only significant shift effect was a NCE on the third shift. Ext;inction results were a function of the initial acquisition series. Together with earlier studies, these results indicate that extended training effects are often different from early stages of training.

Contrast effects as a function of delay and shifts in magnitude of water reward in thirsty rats

“…The failure to obtain contrast effects in the nondelay condition is consistent with sucrose studies in which initial shifts in magnitude of reward did not result in a PCE (Shanab et al, 1976) or a NCE (Flaherty et al, 1973;Shanab et al, 1976). Although a NCE is usually readily obtained in food studies under nondelay conditions (Dunham , 1968), some studies have reported negative results for NCE (Weinstock, 1971), as well as for PCE (McCain, Lobb, & Newberry, 1976).…”

“…The failure to obtain a NCE in the delay condition and the absence of a PCE or NCE in the nondelay condition are obviously inconsistent with this prediction. The present results receive support from other studies in which a second shift in magnitude of food reward produced neither a significant PCEnor a significant NCE (McCain,Lobb, & Newberry, 1976 ;Weinstock, 1971). These findings , however, are to be contrasted with those of other studies that obtained contrast effects during a second shift with both food reward (Benefield et al, 1974) and sucrose reward (Shanab et al, 1976).…”

“…Bower, 1961), however, a reliable PCE has been obtained (Mellgren , 1972;Shanab, Sanders, & Premack, 1969). Repeated-shifts procedures, in which subjects are shifted back to a reward level they have previously experienced, have also produced PCE and NCE (Benefield, Oscos , & Ehrenfreund , 1974;McCain, Lobb , Almand, & Leek, 1976;McCain, Lobb , & Newberry, 1976 in which two or more independent variables are simultaneously shifted, have also resulted in PCE (Shanab & Biller, 1972;Shanab & Cavallaro, 1975).…”

Positive and negative contrast effects as a function of shifts in percentage of reward

“…The superiority of the larger percentage in preshift is similar to the McCain et al (1976) finding that 75% reward generated faster responding than 25% reward, at least early in training. Although it is not possible to determine, either from the present data or from the McCain et al results, whether differential percentage would result in differential asymptotic performance, these data do indicate that percentage will influence rate of approach to asymptote.…”

“…The literature concerning the effects of upshifts in percentage is somewhat ambiguous. While Leung and Jensen (1968) observed positive contrast following upshifts in percentage, several authors have failed to observe such effects (e.g., Lehr, 1974, when magnitude was held constant; McCain, Lobb, & Newberry, 1976;Roop & Nation, 1976). Thus, the present experiment compared a group that received an upshift in reward percentage (from 33% to 100%) to a group that received the higher percentage throughout training.…”