Edward L. Thorndike: the selectionist connectionist.
Thorndike’s law of effect is a Darwinian selection process—use this framing when teaching the roots of reinforcement to students or colleagues.
01Research in Context
What this study did
The author re-read Edward Thorndike’s 1898 cat-puzzle work.
He shows that Thorndike saw learning as a Darwin-style process.
Strong responses live, weak ones die—no mind stuff needed.
What they found
The law of effect is a selection rule, not a mechanical push.
Surviving actions become more frequent, just like genes in evolution.
This view quietly previews today’s brain-behavior links.
How this fits with other research
Imam (2001) picks up the story: Skinner’s 1945 speech on verbal behavior uses the same selection logic.
Petursdottir et al. (2023) extend the idea to iPad voice devices—symbol choices survive when they get quick pay-offs.
Gallistel (2025) adds numbers: what matters is the info rate of the payoff, not just that a payoff came.
Together the four papers turn one old cat experiment into a living frame for AAC, grammar, and math-based schedules.
Why it matters
When you say “consequences select behavior,” you are using Thorndike’s Darwinian lens.
That single line helps parents, teachers, and payors see why we don’t blame the child— we rearrange the contingencies.
Next time you write a BIP, call your procedure a selection process; it roots your plan in 125 years of science.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Open your next staff training with “Behaviors are selected, not created—here’s the data” and show Thorndike’s cat puzzle photo.
02At a glance
03Original abstract
From the very outset of his work, Thorndike allied himself with the Darwinian proposition that complex phenomena can arise as the cumulative effects of a selection process, here the process envisioned by the law of effect. Thorndike's selectionist approach, when combined with his connectionism, laid the foundation for a synthesis of behavior analysis and neuroscience.
Journal of the experimental analysis of behavior, 1999 · doi:10.1901/jeab.1999.72-451