Judging the strength of designated evidence

Judgments of evidence strength were obtained in a series of vignettes, each containing one or two items of evidence for each of two logically independent hypotheses. Each judgment concerned a designated subset of the presented evidence. Different groups of subjects encountered different but overlapping sets of evidence. We compared groups that judged the same designated evidence for the same hypothesis, but differed in exposure to surrounding (nondesignated) evidence. Results showed clear separation of relevant from irrelevant evidence and of designated from surrounding relevant evidence. This was particularly clear in the second experiment, where judgments of designated evidence remained invariant whether the surrounding evidence supported or contradicted the specified hypothesis. The conjunction fallacy was observed for judgments of evidence strength, but was substantially reduced by a simple instruction. Because subjects can separate evidence, it becomes possible to construct cardinally scaled standards of evidence strength using conjoint-measurement methods.     

A multidimensional latent trait model for measuring learning and change

A latent trait model is presented for the repeated measurement of ability based on a multidimensional conceptualization of the change process. A simplex structure is postulated to link item performance under a given measurement condition or occasion to initial ability and to one or more modifiabilities that represent individual differences in change. Since item discriminations are constrained to be equal within a measurement condition, the model belongs to the family of multidimensional Rasch models. Maximum likelihood estimators of the item parameters and abilities are derived, and an example provided that shows good recovery of both item and ability parameters. Properties of the model are explored, particularly for several classical issues in measuring change.     

Dimensions of maltreated children's home behavior: A factor analytic approach

This study identified dimensions of home behavior of maltreating parents and their children through exploratory factor analysis. Five child factors were identified: self-directed behavior, affiliative behavior, negativity, positive social interaction, and questioning. Five parent factors were identified: positive social interaction, caretaking, discipline, refusal, and hostility. Multivariate analyses indicated that neglected children and abused children had less positive social interaction than adequately reared children. Children showed more questioning and less negativity at older ages. Abusing parents were most hostile, whereas neglecting parents showed the least positive social interaction. Abusing parents directed less hostility toward older children than the other child-rearing groups. Parents exhibited less caretaking behavior toward older children and directed more positive social interaction toward children with higher developmental quotients. The results suggest that factor analysis is a viable means of identifying dimensions of child and parent behavior which could facilitate comparison of effects across studies.     

Using aptitude measurements for the optimal assignment of subjects to treatments with and without mastery scores

For assigning subjects to treatments the point of intersection of within-group regression lines is ordinarily used as the critical point. This decision rule is critized and, for several utility functions and any number of treatments, replaced by optimal monotone, nonrandomized (Bayes) rules. Both treatments with and without mastery scores are considered. Moreover, the effect of unreliable criterion scores on the optimal decision rule is examined, and it is illustrated how qualitative information can be combined with aptitude measurements to improve treatment assignment decisions. Although the models in this paper are presented with special reference to the aptitude-treatment interaction problem in education, it is indicated that they apply to a variety of situations in which subjects are assigned to treatments on the basis of some predictor score, as long as there are no allocation quota considerations.     

Enhancement of conditioned autonomic responses in monkey when preshock signals occasion operant suppression.

Classical pairings of a sound stilulus with shock elicited larger magnitude and more rapidly conditioned autonomic responses when subjects were responding on variable-interval schedules for food than when they were eating freely available food. The difference was not attributable to changes in control values of heart rate and blood pressure, or to alterations in motor activity, but appeared related to operant suppression.     

SOCIAL-SKILLS TRAINING TO MODIFY ABUSIVE VERBAL OUTBURSTS IN ADULTS

Social-skills training was used to modify abusive verbal outbursts displayed by two adult psychiatric patients. Five target behaviors—looking, irrelevant comments, hostile comments, inappropriate requests, and appropriate requests were monitored during role-played situations. Social-skills training, consisting of behavior rehearsal with modelling, focused instructions, and feedback, was introduced in a multiple-baseline design across individuals. Training improved all target behaviors. The improved behavior generalized to: (1) novel scenes role-played with the original respondent, (2) training and novel scenes role-played with a different respondent, and (3) interpersonal situations on the hospital ward.     

AN EXPERIMENTAL ANALYSIS OF “SPILLOVER” EFFECTS ON THE SOCIAL INTERACTION OF BEHAVIORALLY HANDICAPPED PRESCHOOL CHILDRENsp1

The effects of prompting and social reinforcement directed to target subjects on their social behavior and that of peers who never received prompting and reinforcement for positive social behavior, were examined. In a combined reversal and multiple-baseline design, three behaviorally handicapped preschool boys who exhibited divergent social behavior repertoires and varied histories with social reinforcement events, were sequentially exposed to intervention conditions in order to investigate “spillover” of treatment effects. Prompting and reinforcement increased positive social behavior and decreased negative social behavior emitted by all target subjects. The results also demonstrated a “spillover” effect on two target subjects, who at various times were not under intervention, and on the peers as well. The findings suggest that: (a) the direct and indirect effects of intervention procedures may be enhanced by designing treatment based on the social repertoire and reinforcement histories of the subjects; and (b) the treatment “spillover” effect may be increased by applying procedures to two children at once, rather than to one at a time.     

OVERCORRECTION: GENERALIZATION AND MAINTENANCE1

The present study examined the effects of one component of overcorrection—verbal reprimands—and compared them with the effects of the entire overcorrection procedure. Generalization and maintenance of effects were also evaluated. A combined reversal and multiple-baseline experimental design was used to evaluate the effect of the various contingencies on the inappropriate searching behavior of a severely and profoundly mentally retarded adult living in a group home. Searching was defined as carefully and thoroughly touching floors, sofas, cabinets, etc. Data were collected by group-home personnel during three separate observation periods each day of the 115-day study: before breakfast, after breakfast, and after dinner. Observers also served as experimenters. Following an initial baseline phase, verbal reprimands followed searching behavior during two daily experimental periods in multiple-baseline fashion. An immediate reduction of searching during experimental periods occurred with the onset of verbal reprimands. No reduction was observed in the third daily observation period, the control session. After a second baseline phase, during which searching behavior was not shown to return to the original baseline levels, overcorrection procedures (washing hands) were instituted. Overcorrection virtually eliminated inappropriate searching during both daily experimental periods. Further, the effects generalized to the daily control session. In a final experimental phase, decreased searching behavior was maintained with verbal reprimands alone.     

Time-based and count-based measurement of preference

Rats' pressing on two levers was reinforced according to two independent variable-interval schedules that were varied during the experiment. Since the levers were connected directly to the programming equipment, bypassing the standard pulseformers, reinforcement could occur while a lever was held down. Although the time a lever was pressed might, therefore, have varied independently of number of presses, these two measures covaried substantially, because the average duration of the presses remained roughly constant. This rough invariance may have resulted from the rats' tendency to make bursts of brief presses (i.e., to jiggle the levers), even though the contingencies encouraged holding. When duration did vary, presses on the two levers tended to vary together. As a result, relative time spent pressing corresponded closely to relative number of presses. Both of these measures conformed well to the matching law. Absolute behavioral frequency at a lever, measured either way, varied directly with proportion of reinforcement for that lever, in accordance with the generalized version of the matching law. Number of presses seemed, on balance, to be a slightly more reliable measure than pressing time. The substantial interchangeability may prove more significant than the slight disparity, however, because it supports the notion that all behavior can be measured on a common scale of time.     

An evaluation of time-sample measures of behavior

Continuous and time-sample measures of the in-seat behavior of a secretary were obtained. Measurement error, i.e., the extent to which the sample measures deviated from the continuous measure, was a function of the frequency of the sample measurements and the criterion used to score an example of the behavior. If the behavior had to be exhibited throughout the observational interval (whole-interval time sampling), there was a consistent underestimate of the continuous measure. If the behavior had to be exhibited only briefly within the observational interval (partial-interval time sampling), there was a consistent overestimate of the continuous measure. And, if the behavior had to be exhibited at the end of the observational interval (momentary time sampling), overestimations and underestimations of the continuous measure occurred about equally often. As expected, the more frequently the sample measures were made the closer was the agreement between the sample and continuous measures. Two conclusions concerning measurement error in interval time sampling were made. The first was that the error will be a function of the mean time per response. The second is that this error will not be consistent across experimental conditions.