Selection by consequences, behavioral evolution, and the price equation

Baum (2017) rewrote Skinner's idea of selection by consequences. He used a math tool called the Price equation. The paper is pure theory—no new lab data.

The goal was to give behavior analysts a clean formula that tracks how one response grows and another fades inside a single lifetime.

The Price equation splits behavioral change into two parts: selection and drift. Selection is when consequences make one response more likely. Drift is random wobble.

By separating the two, the paper shows exactly how reinforcement reshapes the moment-to-moment "population" of responses.

McDowell (2004) and Kulubekova et al. (2013) built digital organisms that learned through selection. Their programs matched live-animal data. Baum's math now gives those same results a one-line proof.

Stahlman et al. (2023) stretch selection across evolution, child learning, and culture. Baum keeps the lens tight on one learner, giving clinicians a ruler for single-client data.

Meyer (1999) and Malone (1999) remind us Thorndike called reinforcement "selection" back in 1905. Baum simply makes the old idea measurable.

You can plug baseline and treatment numbers into the Price equation to show parents exactly how much a behavior changed because of contingencies versus chance. It turns narrative progress notes into a one-line formula that other BCBAs can replicate.