The effects of preference assessment type on problem behavior

Tung and team compared three ways to find reinforcers for kids whose problem behavior is kept going by access to toys or snacks. They ran paired-stimulus, MSWO, and free-operant preference assessments in an alternating-treatments design with four children who had intellectual or developmental disabilities.

Each session was five minutes long. The team counted how often problem behavior happened during the assessment itself.

Free-operant sessions produced the lowest rates of problem behavior—often zero. Paired and MSWO sessions sparked two to five times more disruptions, even when the same items were used.

All kids showed the same pattern, so the effect was large and consistent.

The result lines up with Smith et al. (1997). That study showed high problem behavior during escape extinction made teachers drop the protocol. Here, high problem behavior during paired or MSWO assessments could punish staff for doing those formats.

Dudley et al. (2019) used function-based differential reinforcement to cut food stealing maintained by tangibles. Tung et al. take a different road: they avoid evoking the behavior in the first place by picking a low-cost assessment.

Smith et al. (2023) found cycling DRA lowered resurgence. Tung’s free-operant method may also lower resurgence because it never creates dense reinforcement periods that later disappear.

If you work with clients who hit or scream to get items, start with free-operant preference assessments. You will see fewer disruptions, finish faster, and keep staff happy. Once you know what the child likes, you can move to more structured formats only if you need rank order data.