Primes, contingent attention, and training: effects on a child's motor behavior.

The use of primes, contingent attention, and training sessions to assess a child's engagement and skill in six large motor activities was examined using a combination reversal and multiple-baseline design. Assessment was based on four levels: proximity to equipment, touching equipment, unskilled participation, and skilled participation. Before training, priming (suggestion to the child) was more effective than contingent attention for increasing the subject's engagement (but not skill) in five activities and for increasing skilled participation in one activity. Training of four activities in the natural environment effectively increased the subject's skill level in five activities. Thus, training appeared to generalize to one of these five activities in this setting and also to skillfully executed stair climbing in an adjoining setting. After training, primes and contingent attention were sufficient to maintain both the subject's skill level and engagement in all activities. Postchecks in the same setting the following semester with different teachers revealed only slight increases in participation, as compared to previous baselines, but all participation was at the skilled level. Social interaction, which was not experimentally manipulated, did not systematically vary in relation to changes in experimental conditions.