Reports on two treatment studies of special education students markedly helpless with regard to arithmetic: one comparing attribution retraining treatment to control treatment matched to it in schedule of successes and failures; the other comparing effects of three success-failure schedules on attributions and persistence. (Author/NRB)