Positive and negative reinforcement effects on behavior in a three-person microsociety.
In group settings, negative reinforcement can spark aggression even when work output looks fine.
01Research in Context
What this study did
Three college students sat in a small lab room. They earned tokens by pressing buttons.
Some days the tokens added money to a group pot (positive reinforcement). Other days the tokens only stopped money from being taken away (negative reinforcement).
The researchers watched how much work got done and how often the students argued.
What they found
On avoidance days the students traded more insults. Aggressive words rose sharply.
Work speed stayed about the same, but one student quit early and another slowed down.
Positive days were calmer and just as productive.
How this fits with other research
Kodak et al. (2003) later saw the opposite: negative reinforcement helped two children with disabilities behave better and follow directions. The difference is the group. H et al. used typical adults who could blame each other; Tiffany worked one-on-one with kids who got personal escape.
McGonigle et al. (1982) and Davison et al. (1989) ran similar concurrent-schedule labs. They found that fine details like response rate or feedback loops can flip results. The mixed outcome in H et al. fits that pattern.
Together the papers warn: negative reinforcement is not one-size-fits-all. Context, diagnosis, and social setup decide whether you get compliance or conflict.
Why it matters
If you run group point systems, lean toward earning points rather than avoiding loss. Kids or clients may work the same either way, but the room stays kinder. When you must use avoidance, watch for snide comments or shutdowns and switch back to gain frames at the first sign of tension.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Change one ‘don’t lose your points’ rule to a ‘earn extra points’ rule and track peer arguments for the week.
02At a glance
03Original abstract
Three-person groups, either of males or of females, resided for 6 to 12 days in a continuously programmed environment. Subjects followed a behavioral program that determined the sequential and contingent relations within an inventory of activities. The program consisted of positive reinforcement days and avoidance days. During a positive reinforcement day, each work unit completed by a subject incremented a group account. The account was divided evenly among the three participants at the conclusion of the study. During a negative reinforcement day, no money was earned, and the group was assigned work unit criterion that, if completed, prevented a reduction in accumulated earnings. During negative reinforcement days, subjects made aggressive verbal responses, which differed in magnitude among the four groups. These differences were evident in several distinct behavioral measures. Performances on components of the work unit were not demonstrably affected by the reinforcement schedules in effect, although during the avoidance condition one subject stopped working and another subject's productivity declined.
Journal of the experimental analysis of behavior, 1985 · doi:10.1901/jeab.1985.44-157