How children learn to listen: The effects of modeling feedback styles on children's performance in referential communication.

48 kindergartners and 48 2nd graders participated as listeners in a referential communication task with an adult speaker. Ss were presented with sets of 4 pictures and asked to choose the referent on the basis of a speaker's message. The messages were either adequate or ambiguous, describing 2 of the 4 possible referents. Ss were instructed to ask questions if they needed more information. Prior to these trials, Ss were exposed to 1 of 3 modeling conditions in which they viewed an adult listener responding to ambiguous messages by asking either general questions, specific questions, or specific questions accompanied by an explanation of the underlying strategy. A 4th group received no modeling. Although each of the modeling groups was successful in increasing the frequency of the modeled question, older Ss benefited more from modeling than younger children. Age differences suggested that the effectiveness of modeling interacted with children's existing cognitive abilities. Auxiliary evidence pointed to the role of information processing in children's abilities to deal with ambiguous communications as listeners. (16 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Generalization and maintenance of appropriate behavior through self-control.

8 7-11 yr old disruptive children attended an after school reading tutorial program 1 hr a day for 72 days. After baseline, Ss evaluated their academic and social behavior. A token program was instituted and then modified to include the following procedures: (a) Points and backup reinforcers were made contingent upon accurate self-ratings; (b) the requirement of accurately matching teacher ratings was faded until the children had complete control over point distribution; and (c) backup reinforcers were also faded and eliminated. While there was a lack of generalization of appropriate social behavior to the regular classroom situation, generalization was demonstrated in the 15-min control period of every class, and maintenance was demonstrated in the final week of the program after all backup reinforcers were withdrawn. (PsycINFO Database Record (c) 2010 APA, all rights reserved)