Some experiments on reinforcement principles within a psychiatric ward for delinquent soldiers.

Researchers set up a point economy on a locked psychiatric ward for delinquent soldiers.

They mixed rewards and punishments. Soldiers earned points for jobs done, lost points for rule breaks.

Points bought snacks, passes, and late lights. The team tracked rule violations, work done, and social fights.

Ward rule breaks dropped when points were tied to good behavior.

Work done and polite talk went up. Fights and insults went down.

The full package—tokens, rewards, and fines—kept the gains going.

Tracey et al. (1974) ran a near-copy study on a chronic adult ward. They also saw gains, but only for talk about activities, not talk about people. The 1970 study looked at broad ward life; the 1974 study zoomed in on what talk carries over to new places.

Scull et al. (1973) moved the same token idea to adult stutterers. They added delayed auditory feedback and fines for stuttered words. Speech fluency rose, showing the ward model works for narrow clinical targets too.

Pomerleau et al. (1973) flipped the tokens upside-down: they paid staff, not patients, when patients behaved. Patient gains still rose, proving the economy can work from either side of the nurse’s station.

If you run an adult day program, group home, or locked ward, layer your token system. Pair points for good acts with small fines for problem acts. Track both; post the counts daily. This old paper shows the mix keeps behavior change strong.