Re-establishing conversational skills in overtly psychotic, chronic schizophrenic patients. Discrete trials training on the psychiatric ward.

The team worked with four adults who had lived with schizophrenia for years. All were in the hospital and still hearing voices.

Staff used short, clear trials to teach three simple conversation skills. They gave prompts, praised small steps, and mixed in snacks and cigarettes as rewards.

Each skill needed about 70 trials. The workers tracked answers to personal questions, greetings, and short comments.

Three of the four patients learned to answer greetings and personal questions. The fourth made only small gains.

Skills stayed when staff faded prompts, but only spread to new people when the questions felt familiar. For example, "How are you?" carried over, yet brand-new topics did not.

KELLEHEBERRYMAELLIOTT et al. (1962) used the same ward chaining trick thirty years earlier. They fixed eating, not talking, by making patients hand over a tray and say one word before food. The method still works; the target just moved to social words.

Northup et al. (1991) also taught conversation to adults with severe disabilities, but they used voice-output devices instead of mouth words. Both studies got gains, showing the skill matters more than the response form.

Glenn (1993) stretched the idea further. That team dropped the trial table and taught kids during play. Naturalistic talk worked for children, yet the slow, trial-by-trial style in Zigler et al. (1989) was still needed for adults stuck in psychosis.

You can shape basic conversation even when voices are loud. Break the exchange into tiny steps, deliver dozens of prompted trials, and reinforce every correct response. Expect slow progress and test with new people to see what really transfers. If a patient plateaus, try switching to an AAC device or move practice into real ward activities.