Unequal reinforcer magnitudes and relative preference for cooperation in the dyad.
Cooperation sticks when partners can easily fix unequal pay-offs.
01Research in Context
What this study did
College students sat side-by-side at two levers.
One lever paid only the student who pressed it.
The other lever paid both students, but in unequal amounts.
A light told them when the pay-offs flipped.
They could press a third button to swap roles at any time.
What they found
Some students kept cooperating even when their partner earned more.
Other students hit the swap button to even out the money.
The swap let them fix the gap without losing their own cash.
Cooperation stayed high as long as the gap could be closed.
How this fits with other research
Malagodi et al. (1975) also got students to drop competition, but they did it by making the lever ratio very high.
Both labs show people will switch strategies when the cost gets big, just through different levers.
Landon et al. (2003) found bigger reinforcers always pull more responses.
Whitehead et al. (1975) adds that the size of your partner’s reinforcer can matter just as much as your own.
Together the papers say magnitude drives choice in or out of social games.
Why it matters
You can use unequal team pay-offs to keep cooperation going.
Let a client earn a bigger token pile for helping a peer, then give the peer a way to catch up next turn.
The built-in chance to rebalance keeps both sides playing without extra cost to you.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Set up a cooperative game where the helper earns 3 tokens and the helpee earns 1, then let them swap roles after five correct responses.
02At a glance
03Original abstract
College-student subjects, who were paired with a confederate, chose to respond either independently or cooperatively for money reinforcers. The subject's relative preference for cooperation was assessed by a procedure (analogous to the psychophysical method of limits) in which response choice was monitored as reinforcer magnitude for one response mode was systematically varied while the other remained constant. Relative preference for cooperation was assessed when the confederate's payoff for cooperation was greater than the subject's (Experiment I) and when the confederate's payoff for independent responding was less than the subject's (Experiment II). For some subjects, changes in the confederate's reinforcer magnitudes resulted in shifts in relative preference for cooperation, which reduced the earnings differences, even though these preference shifts reduced the subject's absolute earnings. For those subjects for whom within-dyad differences in reinforcer magnitude produced no effect, a changeover button was introduced that allowed the subject to eliminate the payoff difference without reducing her own earnings; some subjects used this changeover button to eliminate earnings differences. Thus, the behavior of subjects varied, in part, as a function of reinforcer magnitudes provided for the confederate.
Journal of the experimental analysis of behavior, 1975 · doi:10.1901/jeab.1975.24-1