ABA Fundamentals

Reducing delusional speech in chronic, paranoid schizophrenics.

Liberman et al. (1973) · Journal of applied behavior analysis 1973
★ The Verdict

Stop interviews at the first bizarre word and let patients earn friendly chat time for rational talk—speech can double or triple in chronic schizophrenia.

✓ Read this if BCBAs working with adults who have schizophrenia or other psychotic disorders in inpatient or day-program settings.
✗ Skip if Clinicians serving only young children with autism or clients whose problem behavior is not maintained by social attention.

01Research in Context

01

What this study did

Four adults in a state hospital had schizophrenia with daily delusional talk.

Staff ended any interview the second a patient said something bizarre. They also gave each patient a chance to earn 15 minutes of friendly evening chat by speaking rationally during the day.

The team started this plan at different times for each patient to be sure any change came from the program, not luck.

02

What they found

Rational talk jumped two to six times higher for every patient once the rules began.

Three of the four kept talking sensibly even when staff later cut back the evening chats to every other day.

03

How this fits with other research

Cullinan et al. (2001) later added a quick functional analysis first. They showed the same stop-and-earn plan still works when you prove that staff attention keeps the bizarre talk alive.

Gillberg et al. (1983) moved the idea to a group home. Four silent elderly men with intellectual disability started chatting after a short group lesson that ended when they stayed quiet and praised them for speaking up.

Williams et al. (2002) swapped the reward schedule. Instead of making dementia patients work for attention, they gave attention on a fixed timer. Disruptive vocalizations still dropped, showing you can pick either differential reinforcement or noncontingent reinforcement once you know the payoff.

04

Why it matters

You can trim psychotic speech without drugs or restraints. Just stop the social payoff for odd talk and give a clear path to earn friendly time. Try a brief functional analysis first to be sure attention fuels the problem, then set a simple contingency: end the interaction the moment delusional talk starts, save the good chat for rational moments.

Free CEUs

Want CEUs on This Topic?

The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.

Join Free →
→ Action — try this Monday

Run a 5-minute functional analysis to see if staff attention keeps delusional talk going; if yes, end conversations the instant it appears and offer a preferred social activity for 30 seconds of rational speech.

02At a glance

Intervention
differential reinforcement
Design
multiple baseline across participants
Sample size
4
Population
other
Finding
positive
Magnitude
large

03Original abstract

Four schizophrenic patients with paranoid and grandiose delusions who had been hospitalized for an average of 17 yr were exposed to social reinforcement contingencies in a multiple baseline design. During the baseline period, each patient was interviewed for four 10-min sessions each day. The elapsed time from onset of conversation to onset of delusional talk was recorded. At the end of each day, the patients engaged in a 30-min informal chat with a nurse-therapist while relaxing with coffee, snacks, and cigarettes. The intervention introduced two contingencies: (1) The 10-min interviews were terminated as soon as the patient began talking delusionally; (2) The patients earned time for their evening chat by talking rationally during their daytime interviews. Increases of from 200 to 600% in the amount of rational talk exhibited during the interviews occurred as the contingencies were introduced for each patient sequentially over time. These increases were maintained in three patients when the amount of reinforcement was halved, but declined when the patients were confronted directly with their delusional ideas. A modest amount of generalization occurred from the day-time interviews to the evening chats but did not extend to the behavior of the patients on the ward.

Journal of applied behavior analysis, 1973 · doi:10.1901/jaba.1973.6-57