The control of "classroom attention": a group contingency for complex behavior.

Cumulative time measures of classroom attention, as delineated by the teacher, were taken of four elementary classes (kindergarten, third, fifth, and sixth grades) and of 16 randomly chosen students in these same classes. Each class of students was viewed as an individual responding organism. Base rates showed considerable variability. Explicit instructions alone concerning student attention produced temporary increase for some students and for some grades. Adding group contingencies (i.e., contingencies dependent on the attention of every student in the class) and token-mediated reinforcement to class achievement of a gradually increasing attention criterion raised group measures to a consistent 70 to 85% level of time attending to task as instructed, and raised individual student measures to a stable 90 to 100% level. Reversals and other data indicate that the elementary teacher can, by herself and with little effort, maximize what she considers the "paying-attention behavior" of all her students by her less-than-precise measure and consequation of the attention of the class as a whole.