Trend in correlated proportions

A random effects probit model is developed for the case in which the same units are sampled repeatedly at each level of an independent variable. Because the observed proportions may be correlated under these conditions, estimating their trend with respect to the independent variable is no longer a standard problem for probit, logit or loglinear analysis. Using a qualitative analogue of a random regressions model, we employ instead marginal maximum likelihood to estimate the average latent trend line. Likelihood ratio tests of the hypothesis of no trend in the average line, and the hypothesis of no differences in average trend lines between experimental treatments, are proposed. We illustrate the model both with simulated data and with observed data from a clinical experiment in which psychiatric patients on two drug therapies are rated on five occasions for the presence or absence of symptoms.Supported by a grant from the MacArthur Foundation and National Science Foundation Grant BNS85-11774.The authors are indebted to James Heckman for calling our attention to the Clark algorithm.     

Optimal temporal differentiation

In order to illuminate a light signaling a correct response, adult humans had to space their button presses according to a range of time requirements. In some conditions, the spacing needed only to exceed a minimum duration; in others, it had to fall between lower and upper bounds. Mean interresponse times always exceeded the lower limit, and decreased the more stringent were the upper bounds. Variability of interresponse times increased with larger lower bounds, but was unaffected by the size of the upper bound. Feedback about the direction of errors in conditions involving both upper and lower bounds did not affect the means, but it did reduce variability. Predictions were derived from optimality theory, based on the assumption that the critical factor was minimization of the time between correct responses. Without upper bounds, the theory overestimated the mean interresponse times by about 10%; with upper bounds, the theoretical predictions corresponded closely to the actual data. The results did not appear to reflect a scalar timing process. Optimality theory, in contrast to Weber's law, correctly predicted the variety of curves relating sensitivity to duration requirements.     

Feedback effects on sequential ordering in humans

Under various feedback conditions, 38 college undergraduates were asked to rearrange abstract graphic characters on a computer screen, placing them in arbitrarily designated “correct” sequences. Two sets of seven horizontally arrayed stimuli were used. In Experiment 1, subjects in Group 1 learned to arrange the first set under Selection Feedback in which a “+” appeared above each character after it was selected in the correct order and to arrange the second set under Order Feedback in which a correct response produced a copy of the character in its correct ordinal position at the top of the screen. For Group 2 the order of these conditions was reversed. In Experiment 2, for subjects in Group 3, correct responses produced neither of these types of feedback. Subjects in Group 4 received Order Feedback only until the first set was correctly ordered once. Order Feedback was more effective than Selection Feedback during initial acquisition of the first set but not during maintenance; no differences were found for the second set. Only 2 of 9 subjects successfully put the characters in correct sequential order under the No Feedback condition. When, in Experiment 2, Order Feedback was eliminated after the first correctly arranged sequence, the steady-state criteria were met more slowly than in Experiment 1.     

DOES CONDITIONAL DISCRIMINATION LEARNING BY PIGEONS NECESSARILY INVOLVE HIERARCHICAL RELATIONSHIPS?

Four experiments demonstrate that when putative conditional and discriminative cues are presented simultaneously in the single reversal procedure, it is not possible to ascribe a uniquely conditional or uniquely discriminative function to either of the cues. In Experiment 1, pigeons were trained to respond to a blue key and not to a red key while the houselight was on; then in a different session they learned the reversal of this discrimination with the houselight off (single reversal). Separate groups were tested for color generalization with houselight conditions alternating in blocks of trials or for houselight intensity generalization with blue and red key colors alternating in blocks of trials. Both test procedures revealed a conditional relationship between houselight and key color conditions. Experiment 2 produced the same result following training in which the key colors were held constant across training sessions while the houselight and no houselight conditions varied within sessions. In Experiment 3, separate groups were trained with the two procedures but were tested with randomly ordered combinations of key colors and houselight intensities. The two groups yielded indistinguishable bidimensional generalization gradients with peaks at both previously reinforced stimulus combinations. In Experiment 4 the subjects were switched from one of these training procedures to the other with no decrement in their discriminative performance. We conclude that for successive discriminations between conditional- and discriminative-stimulus combinations, the notion of a hierarchical relation between conditional and discriminative stimuli must be extended to include a symmetrical relationship or the notion should be abandoned altogether.     

INDUSTRIAL SAFETY HAZARD REDUCTION THROUGH PERFORMANCE FEEDBACK

A “feedback package” system, designed to prevent occupational accidents and to fit directly into the normal operations of an industrial organization, was analyzed. Eighteen hazardous conditions in six production departments were assessed during seven observation sessions over a 12-week period, plus four follow-up observations over 4 months. The “feedback package” was presented in multiple baseline fashion, across subjects (department supervisors). It consisted of presenting the supervisor with copies of observational data, accompanied by a note which congratulated good practices and suggested ways for improving safety conditions, along with occasional comments from a senior executive. The results indicated that during the feedback phase, hazard rates were lower and less variable than during the baseline phase. Baseline data were highly variable with peaks ranging from 20 to 55 hazards per department. Following intervention, hazard frequencies dropped by 60%, averaged across departments, with decreases ranging from 29% to 88%. During treatment, data stabilized, with the highest frequency reaching 33. A modified feedback system was implemented by the organization following termination of the study, validating the assumption that such a system would tend to maintain.     

TEACHING MENTALLY RETARDED ADULTS TO TIME-MANAGE IN A VOCATIONAL SETTING

Three retarded adults who had minimal ability to tell time were trained to “time-manage.” Each was given a card with clock face representations on which the hands of the clock were drawn, representing each trainee's assigned lunch and break times. Instruction was given before work to perform each of the required behaviors when the “real” clock matched the clock faces. Praise was given following correct responses, and reprimands, instruction, and, in some instances, delay or omission of the scheduled activity followed incorrect responses. Pre-instruction and instructional feedback were then sequentially withdrawn. Results indicated that the package consisting of pre-instruction, instructional feedback, and picture cues was effective in producing independent time management responding. When the first two components were withdrawn, two trainees maintained high levels of correct responding. Correct responding decreased for one trainee when pre-instruction was withdrawn. Reintroduction and subsequent withdrawal of the components resulted in maintenance by this trainee. Little improvement in time-telling ability resulted.     

Reduction of police vehicle accidents through mechaniically aided supervision

Tachograph recorders were installed in 224 vehicles of a metropolitan police department to monitor vehicle operation in an attempt to reduce the rate of accidents. Police sergeants reviewed each tachograph chart and provided feedback to officers regarding their driving performance. Reliability checks and additional feedback procedures were implemented so that upper level supervisors monitored and controlled the performance of field sergeants. The tachograph intervention and components of the feedback system nearly eliminated personal injury accidents and sharply reduced accidents caused by officer negligence. A cost-benefit analysis revealed that the savings in vehicle repair and injury claims outweighed the equipment and operating costs.     

LARGE-SCALE REDUCTIONS IN SPEEDING AND ACCIDENTS IN CANADA AND ISRAEL: A BEHAVIORAL ECOLOGICAL PERSPECTIVE

We assessed the effects of posted feedback and warning ticket programs on speeding and accidents in two cities. In Experiment 1, speeding feedback signs were effective even when 10 were used in Dartmouth, Nova Scotia, and reductions in speeding were associated with reductions in accidents. The effectiveness of the signs was correlated with the number of intersections and residences within 0.5 km beyond them, and the signs had no effect on untreated streets. In Experiment 2, posted feedback and a warning program reduced speeding and accidents on 14 streets in Haifa, Israel.     

Error Feedback Frequency Affects Automaticity But Not Accuracy and Consistency After Extensive Motor Skill Practice

Earlier studies addressed the effects of feedback frequency on movement accuracy and consistency. The authors additionally addressed the effects on motor automatization. High error feedback frequencies may induce attentional control processes and impede motor automatization. In a pre-post design, 42 participants were assigned to 2 groups with different feedback frequencies and practiced an arm movement sequence with 760 trials in 5 sessions. The 100% group practiced with feedback on 3 movement reversals of the sequence after each trial. The 14% group practiced with 14% frequency according to a fading schedule. Only the 14% group showed a decrease in dual-task costs indicating an increase in automaticity. Group differences in movement accuracy and consistency were not evident.