Abstract
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.