Classroom applications of self-determined reinforcement.
Letting students pick or give their own tokens teaches just as much as teacher control.
01Research in Context
What this study did
Glynn (1970) asked ninth-grade girls to set their own token goals for history and geography work.
Some girls picked how many tokens they needed to earn. Others had the teacher pick. A third group got tokens by chance. A fourth group got no tokens at all.
What they found
Girls who chose their own goals learned just as much as girls who got teacher-set goals. Both groups beat the chance and no-token groups.
The learning gains stuck even after tokens stopped.
How this fits with other research
Richman et al. (2001) later showed sixth graders could also hand themselves tokens and still boost work. The idea keeps working across ages.
Ajibola et al. (1995) took the same trick to boys with ADHD. Self-reinforcement plus medication gave the biggest reading gains. The method stretches to kids with diagnoses.
Peltier et al. (2023) swapped tokens for the Good Behavior Game. First graders who ran the game themselves cut disruption just as well as teacher-run games. Student power keeps proving its worth.
Why it matters
You can let learners set or deliver their own reinforcers without losing control. High school, elementary, or ADHD—results stay strong. Try adding a self-point step to your next token board and watch students own their progress.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Ask each student to write how many points they think they can earn today, then honor that goal.
02At a glance
03Original abstract
Self-determined, experimenter-determined, and chance-determined token reinforcement treatments were compared with a no-token treatment, in terms of effect on the learning of history and geography material in the classroom. Each treatment was assigned to one of four heterogeneous classes of Grade nine girls. An initial baseline period preceded the differential reinforcement period, and a token withdrawal period followed. Subsequently, the self-determined treatment was employed in all three token reinforcement classes, before a final baseline period occurred. Findings included a similarity of initial baseline performance for all classes, an equal superiority of self-determined and experimenter-determined treatments to chance-determined and no-token treatments, and significant improvement from initial baseline to final baseline for self-determined and experimenter-determined treatments, but not chance-determined and no-token treatments. Differential token reinforcement experience was found to influence subsequent rate of self-determined token reinforcement.
Journal of applied behavior analysis, 1970 · doi:10.1901/jaba.1970.3-123