The effect of contingent reinforcement on target variables in outpatient psychotherapy for depression: a successful and unsuccessful case using functional analytic psychotherapy.

Two depressed adults came to weekly therapy. One got better. One did not.

The therapists added Functional Analytic Psychotherapy (FAP) to regular CBT. They watched for small in-session improvements like eye contact or sharing feelings. When these happened, the therapist gave praise or asked deeper questions right away.

The team tracked each client's mood and these tiny target behaviors every session. They wanted to see if the real-time rewards helped reduce depression scores.

Client A’s depression scores dropped from severe to mild over the study period. Her target behaviors—like talking about emotions—increased every time the therapist praised them.

Client B’s scores stayed high. His target behaviors barely moved even with the same praise. The reinforcement simply did not work for him.

Same method, two opposite results. The paper shows both success and failure side by side.

Meier et al. (2012) came next and gave us a script. Their paper lists exact words and timing for FAP praise. This builds on Harrington et al. (2006) by turning the case study into a teachable protocol.

De Los Reyes et al. (2009) looked at hundreds of depression studies. Their Range of Possible Changes model warns that any single method helps only some clients. This matches Harrington et al. (2006) perfectly—one client improved, one did not.

LeBlanc et al. (2020) showed that small tracking sheets keep therapists honest. Their matching-task study proves that tight data collection catches drift early. Harrington et al. (2006) did the same by graphing mood and target behaviors session by session.

You can add FAP’s reinforcement to your CBT tomorrow. Pick one clear in-session behavior—like asking for help—and praise it on the spot. Track it and the mood score each visit. If the line goes up for three weeks, keep going. If it stays flat, pivot fast. This case reminds us that idiographic data beats guesswork every time.