Linking descriptive and experimental analyses in the treatment of bizarre speech.

The team worked with an adult who said odd things to get staff to talk to him. First they watched and wrote down what happened before and after the odd talk. Then they ran short tests where staff either ignored him, talked to him often, or gave tasks.

After the tests showed attention kept the odd talk going, they tried two fixes. One: staff gave attention every few minutes no matter what. Two: they taught the man to start chats and add to chats with simple phrases.

Both fixes wiped out the bizarre speech. Scheduled attention worked even when the man did not ask for it. Teaching normal chat worked just as well and gave him new skills.

The brief talk tests matched the long watch notes, so the team knew they had the right cause before treating.

DeLeon et al. (2003) later showed not just any attention matters. When staff echoed the weird words, bizarre talk rose. When staff answered with normal words, it fell. That extends Duker et al. (1991) by proving the content of your reply counts.

Fabbretti et al. (1997) used the same test-then-teach plan with teens in a class. Self-monitoring replaced odd social bids and gains were fast. The pattern shows the 1991 model works across ages and settings.

Taylor et al. (1993) found attention worked only when staff spoke to another adult, not to the client. This adds a layer: check whom the attention is aimed at during your test.

You can kill odd attention-seeking talk in two ways. Give brief friendly words on a timer, or teach the client to start and stretch normal chats. Run a quick attention test first to be sure attention is the fuel. Either route takes no extra staff once planned.