The effectiveness of a brief period of isolation (timeout)

Two aides operating a kindergarten-style program for institutionalized mental retardates were trained, using observer feedback, to apply generalized "correct" social contingencies to 10 defined classes of appropriate and inappropriate child behaviors. A multiple baseline design was used to demonstrate, sequentially, the effects of the training procedure upon the attending behavior of each teacher. After withdrawal of feedback, a posttraining follow-up served to assess the durability of training. For both aides, the effect of training was to increase the proportion of appropriate child behaviors attended to, compared with baseline data, and a follow-up over a number of weeks indicated that the effects of training were apparently durable.There is abundant evidence that social contingencies applied by staff and/or parents can modify selected problem behaviors of children. This has been shown with preschool children (e.g., Baer and Wolf, 1968;Harris, Wolf, and Baer, 1964), school children (e.g., Hall, Lund, and Jackson, 1968;McAllister, Stachowiak, Baer, and Conderman, 1969), children at home (e.g., Hawkins, Peterson, Schweid, and Bijou, 1966;Herbert and Baer, 1972), and in clinical groups with brain injury, emotional disturbance, or retardation (e.g., Hall and Broden, 1967;Zimmerman and Zimmerman, 1962 (FALL 1974)

Differential reinforcement of compliance with teacher invitations to complete a specific academic task was applied to three extremely negativistic children in a special preschool class. For each child, this technique resulted in clear and useful increases in compliance as it was applied. In addition, the technique produced a greater diversity of sampling the available tasks by all children, enabling them to contact instructional materials they had previously avoided. The reinforcement system, contingent access to free playtime, materials, and a snack, mediated by a token, was thus demonstrated to be an effective contingency. In the case of two children whose compliance was not maximized by differential reinforcement alone, further increases in compliance were produced by combining a 1-min timeout for noncompliance with the differential reinforcement procedure.Some instructional control by a teacher over the classroom activities of her students is widely considered a necessity for the successful teaching of many academic skills. On the other hand, it is sometimes characterized as just another example of the excessive repression of children practised by the typical classroom. Neither side of this controversy has produced any empirical evaluation of the actual results of establishing increased instructional control; instead, the issue is argued (spiritedly) on apply systematic reinforcing consequences for such compliance, often resorting instead to reasoning, nagging, and threats contingent on noncompliance. This research was designed to analyze the possibilities of contingent reinforcement and contingent timeout as easily implemented techniques for establishing instructional control, and to examine certain academic results of the establishment of that control.Schutte and Hopkins (1970) pointed out that an operant analysis of instructions treats them as discriminative stimuli that set the occasion for an instruction-following response. To be maintained, an operant ordinarily must be reinforced. Schutte and Hopkins' excellent review of the relevant literature showed that when compliance with instructions is not reinforced, either deliberately or by natural reinforcing consequences in the environment, it is not maintained. However, consequences can be provided for instruction-following which will maintain its rate. The research of Schutte and Hopkins demonstrated a clear relationship between instructional control in a normal kindergarten classroom and continuing teacher attention to compliance. An earlier study by Zimmerman, Zimmerman, and Russell (1969) showed that in a class of retarded boys, a combination of praise with tokens 289 1973, 61,[289][290][291][292][293][294][295][296][297][298] NUMBER 2 (SUMMER 1973)

Token-mediated access to play and snacks was made contingent on completion of academic tasks in the Baseline Experiment. This contingency produced stable completion rates that were subsequently doubled, and then tripled, for four deviant children in a special preschool. A reversal design demonstrated that the contingency was functional in maintaining the children's rates of task completion. The Guidance Experiment examined the role of a social event, teacher guidance, in the acquisition of task-completion skills, in a multiple-baseline-across-tasks design (with reversals). The analysis demonstrated that teacher guidance was an important supplement to the token-mediated contingency in establishing significant increases in task completions for a second group of three deviant children in the special class. The importance of teacher guidance was related to the difficulty level of the children's tasks.