Effectiveness of social reinforcement as a function of contingent and noncontingent satiation

The social satiation effect is the inverse relation between the availability of a social stimulus and its subsequent efficacy in a reinforcing role. According to a suggested cognitive-interactive theory, the satiation effect is mediated by children's attributions of contingency between their own behavior and the experimenter's actions in the satiation treatments. Perceived contingencies depend, at least to an extent, on actual contingencies, and it was therefore predicted that a satiation effect would be observed only for groups presented in the satiation treatment with noncontingent social stimuli but not for groups presented with contingent stimuli. Middle-class 5- and 7-year-old children were subjected to a 10-min waiting period in which the stimulus word “Yafeh” (“good” in English) was presented 2 or 20 times, contingently or noncontingently. They were then given a 75-trial binary discrimination test: correct responses were reinforced with “Yafeh”. The hypothesis was confirmed in the analysis of variance. However, the predicted difference between the slopes of the contingency and noncontingency conditions was found clearly only in the older sample, while the younger children were more influenced by the number of social stimuli presented in the treatment (satiation) and less influenced by the method of stimulus presentation.     

Satiation for social reinforcers: The effects of contingency,vicarious experience,and ambiguous situations

The relative satiation effect, an inverse relationship between the frequency of prior social reinforcement (the word “good”) and the later effectiveness of the social reinforcer in controlling behavior, was studied. In Experiment 1, a discrimination task in which social reinforcement was given for correct responses was administered to first- to fourth-grade children (6 to 10 years of age), who had during a preexposure phase performed a preliminary task or observed another child performing. During the preexposure phase, the experimenter delivered frequent or infrequent social reinforcement that was either contingent or noncontingent. Only performers and observers who had experienced frequent noncontingent reinforcement showed the satiation effect during the discrimination task phase. The results were interpreted as inconsistent with J. L. Gewirtz' (Developmental Psychology, 1969, 1, 2–13) social drive formulation but supportive of an informational analysis in which the children are seen as responding appropriately to unambiguous evidence concerning the reliability of contingency information. In Experiment 2, seating arrangements were varied so that information concerning the direction of reinforcement was made ambiguous. Performers were less responsive during the discrimination phase after experiencing frequent noncontingent reinforcement when seated alone or opposite an observer than when seated next to an observer. The results are interpreted as indicating trust of the reliability of the contingency under ambiguous conditions.     

A negativity bias in interpersonal evaluation

Two studies were conducted to demonstrate a bias toward negativity in evaluations of persons or their work in particular social circumstances. In Study 1, subjects evaluated materials written by peers. Those working under conditions that placed them in low status relative to the audience for their evaluations, or conditions that made their intellectual position within a group insecure, showed a clear bias toward negativity in those evaluations. Only individuals who believed their audience to be of relatively low status and at the same time believed their intellectual position to be secure did not show this bias. In Study 2, subjects viewed a videotape of a stimulus person and rated him on several intellectual and social dimensions. Again, subjects believed their audience to be of either relatively high or relatively low status. As a cross dimension, they were given instructions to focus on either the intellectual or the social abilities of the stimulus person while viewing the videotape. A strong main effect of audience status was demonstrated, but only in ratings of intellectual traits; subjects who believed their audience to be of relatively high status rated the stimulus person's intellectual qualities significantly more negatively. Moreover, this effect was independent of the instructional focus subjects had been given. The negativity bias is discussed in the context of previous demonstrations of biases toward weighting negative information more heavily than positive information, as well as previous demonstrations of seemingly pervasive positivity biases in memory and judgment.     

Priming effects with food and water reinforcers in hamsters

Hamsters were tested in a square maze with either food or water in each corner compartment. A substantial priming effect (runway times increase as intertrial interval increases) was demonstrated using both food and water reinforcers. A lengthy period of free access to food and water (satiation) immediately abolished the priming effect, although runway times in general remained relatively rapid for some time. Extinction of rapid runway times proceeded more slowly under food than under water deprivation. It is concluded that priming is a widespread phenomenon associated with the presentation of a reinforcing stimulus to an appropriately deprived animal.     

An empirical investigation of self-esteem as a composite construct

Three measures of chronic self-esteem, plus four measures developed to assess situational components of self-esteem (task-specific self-esteem and social self-esteem), were administered to 238 undergraduates at a large, urban university and a community college in the southeastern United States. Evidence for a composite construct of self-esteem was found. Results also indicate task-specific self-esteem correlates as high as .42 (p < .01) with chronic self-esteem and as high as .76 (p < .01) with social self-esteem. Task-specific self-esteem was found to be a better predictor of grade point average than chronic or social self-esteem. Suggestions are made for further research on the situational components of self-esteem to extend current findings.     

Social facilitation in chickens: A different level of analysis

An assumption of the arousal (D × H) model of social facilitation effects is that in an otherwise constant stimulus situation the simple presence or absence of a conspecific has a quantitative influence on the D component, and that the habit hierarchy in the situation (H) is unaffected. However, certain evidence from avian subjects seemed inconsistent with the multiplicative effects predicted by the model, and experiments were conducted with the aim of resolving this discrepancy. In general, chickens were tested for consummatory responses in the company of companions, in isolation, or in the company of strangers. Clear findings emerged from these tests that showed that only the companion conditions were “facilitative”. Under the other test conditions there were qualitative (as opposed to merely quantitative) shifts in responses. Apparently, the different social test conditions evoked qualitatively different response hierarchies, and “facilitation” reflected this sort of shift rather than simple quantitative changes in arousal (D) across conditions. It was argued, therefore, that the appropriate level of analysis for facilitation effects for animals in a free response situation is not within the (D × H) envelope, but rather at the level of S (stimulus situation) in the S → (D × H) → R formulation. Implications for the analysis of social facilitation effects in humans were discussed.     

Observer states and traits and the imitative process: A test of a new formulation

In three studies the formulation was tested that characteristics of the observer have an effect on imitation only in situations which provide little information concerning appropriate or expected behaviors. In Experiment I, female college students of high, medium, and low need for social approval underwent either a high or low arousal manipulation. Subjects observed a videotaped model who engaged in a size judgment task. For half the subjects the model received verbal reward (high information); for the other half the model received no feedback (low information). Subsequently, the model and subject responded alternately in the task during which no feedback was given. Neither need for social approval nor manipulated arousal was found to be systematically related to imitation in either information condition. However, the hypothesized relationships were found when subjects' arousal or anxiety levels were determined by self ratings. These results were replicated in a second, similar experiment. In a third experiment, which involved an improved arousal manipulation, the hypothesized relationships were confirmed.     

Factors effecting nonreinforced imitation: the model as a source of information or social control

Three experiments were conducted with kindergarten, second-grade, and fourth-grade children (N = 208) which investigated whether modelling of unreinforced behavior exerts control over childrens' behavior by providing information concerning other types of behavior more likely to be reinforced or by creating social demands for imitation. After learning that reinforcement was available, children observed an adult model emit a reinforced response or an unreinforced response and then remain to monitor the child's subsequent behavior or leave the situation. Compared to a no-model control, all children except kindergarten girls emitted more reinforced responses after observing the model being reinforced. Only second-grade children, however, showed performance changes after observing the unreinforced model. Second-grade children also only performed what they learned when the unreinforced model was not present. Conclusions were that the unreinforced behavior of the model serves not only as a source of information but also as a cue for unreinforced imitation.     

Extraversion and neuroticism in orienting reflex dishabituation

The present study examined the effects of extraversion (E) and neuroticism (N) as individual difference dimensions and of stimulus intensity on electrodermal orienting response habituation, test response, and dishabituation. Subjects were 162 college students selected from a screening sample of 860 as representing extreme groups of high, medium, and low E crossing with high, medium, and low N. Subjects in all nine resulting groups were habituated to criterion on auditory stimuli of 60, 80, and 100 db. A novel (test) stimulus and a dishabituating standard stimulus followed criteron habituation at each intensity. Results indicated that response amplitude for initial, test, and dishabituation responses was a direct function and that habituation rate (trials-to-criterion) was an inverse function of stimulus intensity. Extraverts habituated more rapidly than introverts at high intensity and gave smaller test and dishabituation responses. Some results differed when regression coefficients defined habituation rate and when first stimulus intensities alone were examined. Correlations among initial, test, and dishabituation responses and between each of these and the two measures of habituation rate are also reported. Results were interpreted as partially supportive of Eysenck's theory.     

Facilitating referential communication among children: The listener as teacher

This study was designed to test the hypothesis that systematic variations in listener behavior can have an important influence on both speaker behavior and communicative success among children. In particular, we investigated the idea that systematic variations in listener behavior might have not only within-trial effects on the adequacy of speakers' messages and the accuracy of communication among children, but also cumulative effects on speakers' initial messages across trials. Effects of stimulus complexity were also examined. Pairs of 7- and 9-year-old children participated in a referential communication game, with the younger child serving as speaker and the older one as listener. Half of the listeners were given a plan for effective listening which emphasized the importance of asking questions if the speakers' messages were ambiguous. Replicating earlier findings, the plan manipulation was successful in encouraging listeners to ask questions when necessary. The major result was that listener questions not only had the expected trial-by-trial effect on message adequacy and communicative accuracy, but also showed a cumulative effect on speaker performance. When exposed to systematic listener feedback, speakers improved their initial messages over trials. Stimulus complexity was not a major determinant of performances. These findings suggest that provision of systematic listener feedback may be an effective method for teaching speaker skills to young children.     

Evaluation and recall of evidence: Authoritarianism and the Patty Hearst case

By using the Patricia Hearst case as the stimulus material, two experiments were conducted to examine the relationship between authoritarianism and recall of evidence. In Experiment I it was found that high authoritarians recall more prosecution evidence than defense evidence. This was not the case for the low authoritarians. Results concerning the hypothesis that high authoritarians would recall more character information than low authoritarians were equivocal. In Experiment II a trend indicating that high authoritarians draw more direct inferences from incriminating evidence than do low authoritarians or an attorney was obtained. The results are discussed in terms of previous findings, authoritarian theory, and characteristics of the Patricia Hearst case.     

Post-iconic visual storage: chunking in the reproduction of briefly displayed visual patterns

Two experiments are reported which investigate the organization of visuospatial information in post-iconic storage. In both experiments, stimuli consisting of 10 disks randomly placed in a four-by-five array were tachistoscopically presented to subjects whose task was to recreate the pattern. In Experiment 2, reproduction was constrained (on a row-by-row basis) while in Experiment 1 it was unconstrained. The results of Experiment 1 showed that subjects recalled in terms of “chunks” of spatially adjacent disks, with most “chunks” consisting of about three of four disks. Within each sequence of 10 responses the probability of correctly recalling a chunk decreased with its serial position but was relatively independent of chunk size per se (for chunks containing seven or less disks). In addition, clear topographical variations in accuracy were found, which tended to covary strongly with order of recall. In Experiment 2, the order of reproduction was prespecified (either top row down to bottom row, or bottom row up to top row) in order to induce chunking by rows. The direction of reproduction was either pre- or post-cued. The results of this study showed that subjects encode the stimulus, wherever possible, in a form which is compatible with the constraints imposed on recall order. The results for the postcued conditions provide strong support for the argument that topographical variations in accuracy are a function of variations in accuracy of encoding, and not simply a function of order of report. The results are discussed in terms of an attentional model. It is proposed that a general “anticipatory schema” (cf. Neisser, Cognition and Reality 1976) presets the distribution of attention in the visual field, preselects a set of coding heuristics, and subsequently interacts with the present stimulus pattern. Spatial discontinuities in the distribution of attention resulting from this interaction are regarded as “defining” chunks of stimulus elements.     

ESTABLISHING OPERATIONS AND REINFORCEMENT EFFECTS

Positive reinforcement procedures have had a major impact on educational programs for the developmentally disabled; nevertheless, variation in reinforcer effectiveness both within and across individuals is a common phenomenon. This study examined one class of variables—establishing operations—that might influence the effectiveness of reinforcers. Five developmentally disabled adult males participated. Responding on one of two motor tasks—switch closure or block placement—was assessed during baseline, satiation, and deprivation conditions with respect to three classes of consequences: small food items, music, and social praise. Deprivation and satiation conditions were constructed so as not to alter significantly the normal course of events in a subject's day. For example, food deprivation entailed scheduling sessions just prior to a subject's regular lunch, and social deprivation involved limiting a subject's access to social interaction for 15 minutes, during which time the subject had access to an assortment of other activities. Results showed that each stimulus class functioned as reinforcement with different degrees of effectiveness during satiation versus deprivation conditions. These results are discussed in light of previous research on enhancement of reinforcer efficacy as well as the assessment and identification of functional reinforcers, and implications are presented for future research and client habilitation.     

Responding to a positive stimulus by “satiated” pigeons

Skinner (1938) found that rats given discrimination training (Phase I) and then reinforced to “satiation” for responses in the presence of the negative stimulus (S?) (Phase II), began to respond again when the positive stimulus (S+) was reintroduced (Phase III). Experiment I replicated Skinner's finding with pigeons, alternating S+ and S? presentations during Phase III. In Experiment II, Phase II was extended, and Phase III results were similar to those of Experiment I, demonstrating that the recovery of S+ responding could not be attributed to a lax Phase II satiation criterion. In Experiment III, a uniform schedule of reinforcement was maintained throughout the three phases, and results similar to those of Experiment I were found, indicating that renewed S+ responding was not due to the shift in schedule between phases. In Experiment IV, Phase I consisted of discrimination training with two positive stimuli (S+_s and S+_n), and Phase II consisted of reinforcement for responses in the presence of S? and S+_s. During Phase III, significantly more responding was found to S+_s and S+_n than to S?, but no difference in responding was found between the two positive stimuli. In Experiment V, Phase I consisted of simple discrimination training, and during Phase II, responses in the presence of both S? and a novel stimulus (So) were reinforced. During Phase III, significantly more responding was found to S+ than to either S? or So, with no difference found between S? and So responding. Renewed responding to S+ during Phase III in the present experiments is best explained by behavioral contrast developed during Phase I.     

Trait attribution to a described action as a function of changes in salient information

The relative importance of five information variables in determining the amount of friendliness attributed to a described action was analyzed using a five-way ANOVA. The variables (a) antecedent condition, (b) disposition of the actor, (c) intent of the actor, (d) likely social outcome of the act, and (e) actual social outcome of the act, were given a high friendliness value and a low friendliness value. Thirty-two stories which described the same act, but which systematically differed in background information, were made by combining one of the two values of the five variables in all possible combinations. Senior secondary school girls read the stories and rated the friendliness of the act and the friendliness of the actor on 9-place, unipolar rating scales. Results showed that the intent of the actor contributed the only significant source of variance to the trait attribution to the act. The trait attribution to the actor was significantly influenced by intent, antecedent condition, and disposition.     

The effects of constant vs varied reinforcer presentation on responding by autistic children

The autistic child's characteristic lack of motivation can create considerable problems for educators. Since there is some evidence that stimulus variation may influence motivation, this study was designed to assess the differential effects of constant vs varied reinforcer presentation. Specifically, two questions were addressed: (1) Would there be any differences in the total number of responses emitted by autistic children when the reinforcer was held constant as opposed to varied; and (2) would there be any differences in the interresponse interval for constant vs varied reinforcer presentation? The results clearly demonstrated that the children in this investigation made significantly more and faster responses when the reinforcer presentation was varied as opposed to constant. The results are discussed in terms of the effects of stimulus variation on satiation and implications for teachers of autistic children.     

Loss of control by a less valid or by a less salient stimulus compounded with a better predictor of reinforcement

An unsalient stimulus, or one imperfectly correlated with reinforcement, may acquire significant control over responding, provided that it is the only available signal for reinforcement, but may fail to acquire control if it is reinforced only in conjunction with a second, more salient or more valid stimulus. A stimulus imperfectly correlated with reinforcement may also lose control over responding if having initially been reinforced in isolation, it is subsequently reinforced only in conjunction with another, more valid stimulus. If the effects of relative salience are to be explained in exactly the same way as those of relative validity, we should expect a similar loss of control by an unsalient stimulus, A, if, after initial consistently reinforced trials to A alone, subjects subsequently receive reinforcement only in the presence of a compound stimulus, A + B. Two experiments on discrete-trial discrimination learning in pigeons and one on conditioned suppression in rats confirm this expectation. The results have implications for theories of selective association in conditioning and discrimination learning.     

The role of vivid information in fear appeals and attitude change

Fear-arousing persuasive communications frequently incorporate so many types of information that the results can be difficult to interpret. R. Nisbett and L. Ross' (1980, Human inference: Strategies and shortcomings of social judgment, Englewood Cliffs, NJ: Prentice-Hall) analysis of informational vividness offers an attractive conceptual framework that identifies and predicts the effects of several types of information that have been confounded with other variables in fear appeal studies. Thus, the first purpose of this factorial experiment was to investigate the persuasive impact of three types of vivid information frequently used in fear appeals: emotional interest, concreteness, and proximity. It was predicted that vivid information would be more persuasive than pallid information. The results disclosed that emotionally interesting information was highly effective in changing intentions to seek protection from the depicted health threat. Also, the impact of information of high emotional interest and greater temporal and spatial proximity was enhanced with the passage of time, but information of low emotional interest and proximity diminished over time. Additional findings indicated that concrete, specific information was memorable and affected beliefs. The data rejected the hypothesis that memorial availability mediated the effects of vivid information. Implications for designing and strengthening fear appeals were discussed.     

Relative deprivation and referent cognitions

Determinants of relative deprivation were examined in a laboratory experiment. Discontent associated with relative deprivation was hypothesized to be a function of likelihood, or the perceived probability of actually obtaining a desired outcome under existing circumstances, and referent cognitions, which involve the subjunctive probability of obtaining that outcome under alternative (hypothetical) circumstances. Male and female undergraduates discovered that their alleged chances of receiving experimental credit were either poor (low likelihood) or good (high likelihood). They also learned that if a different procedure had been used, their chances of receiving credit would have been either poor (low referent) or good (high referent). Responses to a subsequent questionnaire indicated that whereas discontent did not differ by referent conditions under high likelihood, greater discontent was shown by low-likelihood/high-referent subjects than by low-likelihood/low-referent subjects. This low- versus high-referent difference under low likelihood was also found in two conceptual replications designed to eliminate any influence of social comparison. The relevance of this model to other findings and theoretical formulations is discussed.