SIZE GENERALIZATION IN CHILDREN AS A FUNCTION OF TEST METHOD AND AGE

Children, ranging in age from 31/2-41/2 years and 41/2-51/2 years were provided with successive discrimination training in which S+ was a white square on a black background and S- was a black background. Bidirectional size generalization testing with either a “booster” or “instructional” method indicated that (l)test method did not differentially affect generalization patterns; (2) younger children responded maximally to all test stimuli while spontaneously labeling the size stimuli “big” and “little”; (3) older children generated a group gradient composed of one subgroup who responded to all test stimuli and a second subgroup who responded to S+ only; (4) posttest verbal behavior suggested that the size cue controlled S+-only responding while controlling stimuli associated with flat patterns were color (“white”) and form (“window”). The results drew attention to stimulus selection processes in the stimulus generalization experiment. © 1969, Psychonomic Journals, Inc.. All rights reserved.