Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)



First Advisor

Mary Lou Kelley


The effects of the high-probability instructional sequence, with and without extinction, on increasing compliance to eat in three children with severe food refusal and selectivity were evaluated. Pediatric feeding disorders, in which children refuse most or all food and drink, are a rare and life threatening problem that requires intensive treatment. The majority of behavioral techniques used with feeding disordered children manipulate the consequence of food refusal. The use of antecedent procedures in the treatment of food refusal has been limited. One antecedent technique, the high-probability instructional sequence, involves first presenting the individual with a series of instructions that he/she almost always complies with (high-p instructions) and then immediately presenting a task that the individual has a history of low compliance to (low-p instructions). Past research has demonstrated the effectiveness of this technique in increasing compliance to demands, in the absence of competing behaviors. However, in the treatment of feeding disorders and other severe behaviors (where there may be competing refusal behaviors), the addition of an extinction procedure is often found necessary to gain appreciable results. In this evaluation, the high-p sequence was compared against a control sequence (no high-p) under two conditions: extinction and no extinction. Results indicate that the presence of the high-p sequence did not appear to provide beneficial effects in terms of increasing acceptance of food. Although, for two participants, food refusal behaviors were observed to occur less often in phases in which the high-p sequence was utilized. For all participants, increases in acceptance of low-probability foods and decreases in food refusal behaviors were only observed with the addition of the extinction procedure.