Discrimination learning with shock for right and wrong responses in the same subjects.

Abstract

Recently Muenzinger (5) and Wischner (11) have discussed certain differences in the results of their respective experiments on the effect of shock upon the learning of a discrimination habit by the white rat. The present writers do not question the results obtained by either investigator, but call attention to one noteworthy feature: In the studies of both, shock was employed for right choices only, or for wrong choices only. In view of Muenzinger's suggestion (4) that shock anywhere after choice facilitates learning, and in view of the literature concerning secondary reinforcement in discrimination learning situations, it seems desirable to discover whether shock will facilitate such learning when the shock cannot serve as a cue to the correctness or incorrectness of the responses, i.e., when shock accompanies both right and wrong responses in the same subjects. If the shocked group learns faster than the nonshocked group, Muenzinger's suggestion is upheld, and the results are consistent with the literature on secondary reinforcement. METHOD SubjectsThe subjects were 20 experimentally naive albino rats from the colony maintained by the Psychology Department at Bowling Green State University. There were 10 males and 10 females, ranging in age from 60 to 100 days at the start of the experiment. With the aid of a table of random numbers, 5 males and 5 females were assigned to each condition.

DISCRIMINATED RESPONSE AND INCENTIVE PROCESSES IN OPERANT CONDITIONING: A TWO‐FACTOR MODEL OF STIMULUS CONTROL1

“…However, these experiments must be interpreted cautiously with respect to the present analysis. Well over one-third of the response-produced shocks in tone as well as in light were presented simultaneously with food, making it likely that the mild shocks acquired some positive properties through their food-signalling function (see Azrin and Holz, 1966;Freeburne and Taylor, 1952;Logan and Wagner, 1965). [This shock-food correlation is revealed by examination of the training cumulative records presented by Wiltz (1972) when the reinforcement presentations indicated by slash marks of the response pen are lined up with response-produced shocks recorded by the event pen.…”

Masochism: An empirical analysis.

“…The initial finding (Muenzinger, 1934), that a shock which was twice the threshold value for young rats (Muenzinger & Mize, 1933) would actually facilitate the learning of a light-dark discrimination beyond the influence of reward alone, was confirmed in a number of subsequent investigations (Muenzinger & Baxter, 1957;Muenzinger, Bernstone, & Richards, 1938;Muenzinger, Brown, Crow, & Powloski, 1952;Muenzinger & Powlowski, 1951;Muenzinger & Wood, 1935). Likewise, Drew (1938) and Freeburne and Taylor (1952) found that shock enhanced discrimination learning in rats beyond the effects of reward alone. Rosenblum and Harlow (1963) report a similar effect in connection with a different species operating in a different procedure.…”

Tests of two hypotheses of shock-right facilitation.

“…In proposing their distinctive-cue hypothesis, Fowler and Wischner (1969) stressed difficulty of discrimination as a crucial variable; i.e., shockright facilitation increased with difficulty. It is worth noting that Freeburne and Taylor (1952), who found indiscriminate shock facilitating, used a 5-sec. delay of reward and that 6 of their 20 -?s failed to solve the problem in 500 trials.…”

“…The most decisive test of Muenzinger's (1934) theory is to compare a nonshocked SEWARD, ROSKIN, KOSSLYN, GREATHOUSE, AND WEXLER group with one shocked after all responses, correct or incorrect. This test has been made four times before, twice with positive results (Freeburne & Taylor, 1952;Muenzinger & Wood, 1935) and twice with negative (Fowler & Wischner, 1969;Prince, 1956). Muenzinger and Wood used the correction method; the other three experiments used noncorrection.…”

“…The possibility that shock closely followed by reward might become a secondary reinforcer had already been recognized. Freeburne and Taylor (1952) found this assumption insufficient to explain their results (see later); Prince (1956), however, found it useful in explaining his. Fowler and Wischner (1965), directly testing the hypothesis, got no evidence of acquired reinforcement in the extinction performance of rats previously shocked for correct responses.…”