Dennis G. Dyck

Preshift reward magnitude and positive contrast in the rat

Four groups of rats were given 24 trials of training in a straight runway with 1, 2, 4, or 8 food pellets, and then shifted to 8 pellets. There was a 20-sec delay of reinforcement on each trial, to prevent a ceiling effect on running speeds. The three shifted groups showed stable positive contrast relative to the unshifted control group. Postshift running speed was inversely related to magnitude of preshift reward.

Programmed Student Achievement: A Test of the Avoidance Hypothesis

College students operated under different intensities of the consequence of failing to evidence criterion performance (100 percent mastery) on weekly quizzes. Although there were no differences in test performance as a function of the aversiveness of the contingencies, Programmed Achievement (PA) student performance on weekly quizzes and major examinations was superior relative to control groups where no contingencies were in effect. Additionally, the Programmed Achievement (PA) students demonstrated higher performance than control students on an unannounced retention test. An anxiety scale administered to all students at the beginning and end of the semester revealed that no differential anxiety levels were produced by the contingencies. The results suggested that an avoidance interpretation of Programmed Student Achievement may be inappro priate . SUCCESS IN A college classroom is usually defined by the teacher in terms of performance on exams, with student motivational level being one of the more important factors determining exam per formance. Even though motivational variables have long been suspected to interact with learning at the college level they have largely been ignored by most college educators. This is particularly true in large general survey courses which are often structured for the convenience of the instructor. Students must evidence learning through test per formance, thus, the educator should teach the stu dent to perform at an optimum level. These moti vational variables, which are strong contributors to an individuals test performance, deserve sub stantial consideration. Whaley and Malott (3) investigated motivational variables which affect subsequent scholastic achieve ment. They required students in an introductory psychology course to retake exams until they evi denced 100 percent mastery of discrete units of material. This was not as malicious as it might first appear, as the student was responsible for only small amounts of material at a time and had virtually unlimited opportunities to retake each exam. Under these contingencies Whaley and Malott reported that 98 percent of their students received As in the course. This Programmed Student Achievement (PA) procedure defined above was hypothesized to be analogous to an avoidance task, with the student avoiding failure if he evi denced criterion performance (100 percent mastery) . The student could avoid the noxious consequence of failure only by exhibiting a high level of performance. Their assumption was that test performance reflects study behavior and that the threat of failure energizes this behavior. The Whaley and Malott results must be treated with caution since only the last test score in the series of retake examinations was recorded, insuring 100 percent mastery or failure. Knight (2 ) recording only the initial test grade in the series of retake examinations, provided further support for the assumption that student performance is enhanced when the PA contingency is in effect. It was also demonstrated that the threat of losing a letter grade was as powerful a motivator as failing the course when the criterion was not reached, and that a group of students who were not operating under the PA contingency significantly increased their performance on quizzes when shifted to the PA contingency at mid-semester. While this study supported the hypothesis that the introduction of the PA contingency results in superior performance only 47 percent of the students received a letter grade of A; substantially lower than the 98 percent reported by Whaley and Malott. Whaley and Malott, in proposing an avoidance model, assumed that the student is controlled by threats of failure. In order to avoid the punish ment of failure the student must study discrete This content downloaded from 157.55.39.238 on Sat, 02 Jul 2016 06:19:27 UTC All use subject to http://about.jstor.org/terms 58 THE JOURNAL OF EXPERIMENTAL EDUCATION units of material and evidence complete mastery of the material. If this analogy to the avoidance paradigm is applicable, then performance should vary as a function of the intensity of the consequence of failing to evidence mastery. Knight (1) in a direct test of this prediction, failed to find any dif ferences between two Programmed Student Achieve ment groups which were operating under different intensities of the consequence of failing to evidence criterion, i.e., course failure for one group and loss of a letter grade for the other. It should be pointed out, however, that the lack of difference between the PA conditions might be attributed to the fact that loss of a letter grade and failure may psychologically represent equally severe conse quences to the student. The purpose of the present experiment was to further investigate the potential of the PA contin gency as an academic motivational factor. Of theo retical interest is the question of whether or not the avoidance paradigm is an appropriate model for the PA contingency. In order to answer this question a considerably lessened consequence of failing to evidence criterion was provided, i.e., loss of a letter grade if the student failed to demonstrate criterion on three of seven weekly quizzes. Secondly, in previous research the instructor has known the design of the experiment and may have inadvertantly contributed to the superior performance demon strated by the PA students. The instructor variable was also controlled in the present investigation.

Excitatory backward conditioning in conditioned punishment and conditioned suppression in rats

Three experiments were conducted to examine the effect of assessment procedure, either conditioned punishment extinction or conditioned suppression reacquisition, on the associative outcome of backward conditioning in the rat. Each experiment consisted of three phases: (a) magazine and barpress training; (b) Pavlovian training in which groups received one of backward, explicitly unpaired, or no presentation of a conditioned stimulus (CS) and unconditioned stimulus (US); and (c) either conditioned punishment or conditioned suppression transfer of training tests. In Experiment 1, backward conditioning resulted in excitatory associative effects during the conditioned punishment test. In Experiment 2, backward conditioning produced excitatory conditioning when assessed by a conditioned suppression reacquisition test. In Experiment 3, a number of backward conditioning parameters were manipulated, and backward pairings yielded facilitative reacquisition relative to the unpaired group, but did not differ from the no-treatment control group. The pattern of results suggests that backward conditioning produces qualitatively similar outcomes when assessed by a conditioned punishment extinction or conditioned suppression reacquisition test.

Differential reward magnitude and human conditioning with social reinforcers

In two separate investigations, human Ss received social stimuli as reinforcers in a discrete-trials paradigm of instrumental conditioning. The major variable of interest was reward magnitude utilizing stimuli from attraction research. In the first experiment, a group receiving large reward performed the instrumental response faster than a group receiving small reward. In a second experiment, two groups received large and small reward, respectively, and a third group received differential magnitudes of reward (both large and small) correlated with external stimuli. The results indicated that the differential group was depressed below both the large and small groups in the first portion of the response. These data (1) demonstrate magnitude of reward effects in an instrumental learning paradigm utilizing human Ss and (2) indicate that attraction stimuli function in an analogous manner to more traditional reinforcers.

Task- and Setting-Related Cues in Immunization against Learned Helplessness

Learned helplessness theory predicts that experience with response contingent outcomes prior to exposure to noncontingent outcomes should interfere with the formation of an expectation that outcomes are response independent, and thereby prevent performance deficits which would otherwise occur. This study examined the generality of such immunization by varying the similarity of task and setting-related cues between the immunization and testing situations. Immunization training consisted of success experience on cognitive problems either similar to or different from the final anagrams test, in an experimental setting which was either the same as or different from the one in which pretreatment and testing later occurred. Interference effects were observed in the group receiving uncontrollable noise without prior immunization. The group immunized in the same setting and on a similar task as testing showed no deficits. Diminishing degrees of immunization were observed in the remaining immunization conditions as first setting similarity and then task similarity decreased. These results were interpreted within a stimulus control framework.

Transfer of escape conditioning to extinction of a food-reinforced response

Four experiments test the hypothesis that escape learning in response to shock will transfer to a similar food-reinforced response and affect resistance to appetitive extinction. In the first two experiments, subjects were given escape training in a straight alley followed by continuous food reinforcement and then extinction. Prior escape training resulted in greater resistance to extinction of the food-reinforced response as compared to several control procedures. In the third experiment, the escape response was manipulated to be compatible or incompatible with the subsequent food-reinforced response. Greater resistance to extinction was shown when the two responses were compatible. The fourth experiment confirmed and extended this finding. The relationship of the present results to Amsel’s theory of persistence was discussed.

Wolpe's Reciprocal Inhibition Principle: An Animal Analogue

Wolpes reciprocal inhibition principle represents, on a purely operational level, nothing more than a procedural description of counterconditioning. The most obvious implication of this principle is that a positive relationship, at least to some asymptotic point, should exist between the degree of counterconditioning and the subsequent reduction of avoidance behavior. An analogue study using rats (N = 70) in a passive avoidance paradigm was used to test this implication and to compare the efficacy of response prevention and extinction with varying degrees of counterconditioning. Licking (the animals were water deprived) in the former safe area served as the incompatible response. The effects of five specific conditions—Counterconditioning High, Counterconditioning Medium, Counterconditioning Low, Extinction, Response Prevention—and two control procedures, Untreated Control/Home Cage and Untreated Control/Trash Can, were assessed on each of three days following passive avoidance acquisition on Day 1. A nonshock control group was also used. The Response Prevention, Counterconditioning High, Counterconditioning Medium, and Untreated Control/Home Cage subjects evidenced a significant reduction in passive avoidance behavior relative to the Counterconditioning Low, Extinction, and Untreated Control/Trash Can subjects. It is suggested that lowered arousal may enhance cue utilization and thereby serve as a facilitator for increased exposure, which allows for the elimination of motivational as well as discriminative cues associated with fear and avoidant behavior.