ABA Fundamentals

Superior-subordinate dyads: Dependence of leader effectiveness on mutual reinforcement contingencies.

Rao et al. (1991) · Journal of the experimental analysis of behavior 1991
★ The Verdict

Leadership is a two-way vending machine—both sides must dispense value.

✓ Read this if BCBAs training staff, parents, or peer mentors in any setting.
✗ Skip if Clinicians who only run one-way token systems with no reciprocal component.

01Research in Context

01

What this study did

Bickel et al. (1991) built a lab game with two adults. One played “boss,” the other “worker.”

The boss earned points only when the worker hit a button. The worker earned points only when the boss hit a button.

They compared this two-way setup to one-way pay and free pay.

02

What they found

When both people needed each other’s responses to get points, work rates shot up.

One-way or free pay barely moved the needle. Leadership popped up only when the pay was mutual.

03

How this fits with other research

Vergason et al. (2020) ran a zoo gift-token game that mirrors the lab finding. Guests handed staff tokens for friendly greetings; staff greeting scores jumped about 40%. Both studies show the same rule: the “follower” must also control reinforcers for the “leader.”

Rusch et al. (1981) adds a parent twist. After normal parent training stalled, they taught moms to set goals and track their own behavior. Child problems then dropped in grocery stores and parks. Self-management gave parents a way to keep the mutual loop alive without a clinician hovering.

Berkovits et al. (2014) stretches the idea to workers with autism. Interviews revealed that bosses who varied praise, choice, and small tasks sparked better work. Mutual reinforcement again, just with different currencies.

04

Why it matters

If you want staff, parents, or peers to lead, build two-way streets. Tie their reinforcers to each other’s responses. A supervisee’s smile, a child’s compliance, or a coworker’s prompt can all become the “points” that keep the leader responding. Start small: let the learner deliver a token, sticker, or verbal praise that the adult needs to earn a break. Watch the loop tighten and rates rise.

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

Let your learner hand you a token each time you give correct feedback; cash your tokens for a 2-min coffee break only after they deliver five.

02At a glance

Intervention
other
Design
single case other
Population
neurotypical
Finding
positive
Magnitude
large

03Original abstract

Task contingencies were modeled from bureaucratic organizations in which vague job descriptions provide incomplete contingency specifications. Response rates within dyads were examined using two nonsocial, two social, and two control contingencies. In the first social contingency, responses by the superior produced monetary reinforcement for a subordinate while the superior received no reinforcement from his subordinate. A second social contingency was identical to the first except that the subordinate's rate of responding determined the rate of reinforcement delivered to his superior. Within this contingency, mutual reinforcement occurred whenever rates of superior and subordinate responding were correlated. Two control contingencies were identical to the second social contingency except that either the superior or the subordinate received a rate of response-independent reinforcement virtually identical to the rate received during the second social contingency. Leadership, in this context, was the difference between rates of subordinate responding produced by a nonsocial contingency and rates produced by each of the two social contingencies. The two nonsocial contingencies supported almost no responding among subjects. The first social contingency produced minimal levels of leadership within every dyad. The second social contingency produced high levels of leadership. Response-independent reinforcement generally reduced or eliminated responding.

Journal of the experimental analysis of behavior, 1991 · doi:10.1901/jeab.1991.56-105