On the current status of the evolutionary theory of behavior dynamics

McDowell (2019) wrote a theory paper. He asked: Can Darwin-style computer rules explain how behavior adapts?

He used fake genes inside a computer. The genes stood for response rules. The fittest rules lived and had baby rules.

The model kept picking rules that made more rewards happen. The fake genes slowly built a steady pattern of good moves.

The paper says this trick gives one clean story for all adaptive acts.

Kubina (2021) took the same engine and aimed it at Precision Teaching. She said you can watch fluency grow without doing a full FA. The two papers share the engine; Kubina just drives it down a narrower track.

Baum (2020) tells a different tale. He thinks aversive events induce avoidance in one shot. No slow gene-like steps. The papers clash on speed: McDowell needs many cycles; Baum needs one scary moment.

Gde Jonge et al. (2025) split the middle. They say every reinforcer also induces new responses right away. Their view lets both fast induction and slow selection live side-by-side.

You now have three levers. When a client stalls, ask: Is this a slow-fit problem that needs lots of trials and tiny steps? Or is it a one-shot induction that will vanish if I block the trigger? Try the fast test first—remove or add one event and watch. If nothing shifts, move to the slow engine: run timed fluency drills and let the data breed better moves.