Operant Stuttering: the Control of Stuttering Behavior through Response-contingent Consequences.

Flanagan et al. (1958) wrote a theory paper. They asked: What if stuttering is just an operant behavior?

The authors sketched how stuttering could start and stop based on what happens right after it. No lab test yet—just the blueprint.

They argued stuttering is not locked in by inner anxiety. Instead, it rises or falls because of its own pay-offs.

If a stutter gets attention, escape, or any other quick reward, it will grow. Remove the reward and it should shrink.

Sainsbury (1971) later showed kids will press a button more when a silly word is paired with candy. This gave real data to the idea that any response can be shaped by its consequences.

Hamm et al. (1978) then proved adults will do boring tasks if that unlocks fun tasks. Again, the same rule: tie a less-liked act to a liked outcome and the less-liked act grows.

Bland et al. (2018) pushed the idea further. They found that even a simple “no” signal can punish pigeons’ key pecks. This opens a gentle way to cut stuttering without harsh penalties.

Embregts (2000) wrapped it all up by grouping reinforcers into classes. One candy, one praise, or one token can act the same if they sit in the same class. This helps you pick the best of rewards that best fit each client.

You can treat stuttering like any other operant. Pinpoint what follows each stutter—attention, break from talking, extra time—and change that payoff. Swap the old reward for a better one, or simply withhold it. Start small: pick one client, log every stutter and its consequence for one session, then test a new payoff the next day.