A selectionist approach to reinforcement.
Reinforcement may be nothing more than keeping brain connections that work and dropping the rest.
01Research in Context
What this study did
The authors built a computer model of a simple neural network. They let the network "learn" by keeping connections that worked and dropping ones that did not.
No animals or people took part. The paper is pure theory. It asks whether one rule—keep winning links, prune losing ones—can copy the basic facts of both Pavlovian and operant conditioning.
What they found
The network showed acquisition, extinction, and stimulus control without any talk of "elicited responses." Just by selecting the right wiring, the model acted like a rat pressing a lever or a dog salivating to a bell.
The same single rule handled both types of learning. The authors say this backs a selectionist view: reinforcement is survival of the fittest synapse.
How this fits with other research
Atnip (1977) set the stage. That paper said reinforcement is about the relation between response rate and reinforcer rate, not about absolute size or amount. Nangle et al. (1993) keeps the relational spirit but moves it inside the brain.
Shearn et al. (1997) came next and pushed the idea further. They argue reinforcers mainly change antecedent control, not response strength. The two papers look opposite—one says "pick synapses," the other says "watch what comes before"—but both reject the old "stamp-in" story. They are updating different parts of the same house.
Soto (2020) gives you a way to test the idea. It urges single-case designs in neuroscience so you can show that a brain change really changes one person’s behavior. Think of it as the method arm for the theory arm Nangle et al. (1993) built.
Why it matters
If reinforcement is just neural selection, your intervention can aim at strengthening useful connections. Try teaching a skill in rich, varied contexts, then let natural consequences prune what does not help. Track one client closely—Soto (2020) style—to see if behavior shifts when the environment picks the "winning" responses. You are not just reinforcing acts; you are curating circuits.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Run a brief probe session with varied stimuli and consequences, then plot which responses survive—talk about "selected" behavior with your team.
02At a glance
03Original abstract
We describe a principle of reinforcement that draws upon experimental analyses of both behavior and the neurosciences. Some of the implications of this principle for the interpretation of behavior are explored using computer simulations of adaptive neural networks. The simulations indicate that a single reinforcement principle, implemented in a biologically plausible neural network, is competent to produce as its cumulative product networks that can mediate a substantial number of the phenomena generated by respondent and operant contingencies. These include acquisition, extinction, reacquisition, conditioned reinforcement, and stimulus-control phenomena such as blocking and stimulus discrimination. The characteristics of the environment-behavior relations selected by the action of reinforcement on the connectivity of the network are consistent with behavior-analytic formulations: Operants are not elicited but, instead, the network permits them to be guided by the environment. Moreover, the guidance of behavior is context dependent, with the pathways activated by a stimulus determined in part by what other stimuli are acting on the network at that moment. In keeping with a selectionist approach to complexity, the cumulative effects of relatively simple reinforcement processes give promise of simulating the complex behavior of living organisms when acting upon adaptive neural networks.
Journal of the experimental analysis of behavior, 1993 · doi:10.1901/jeab.1993.60-17