A discriminated rapid‐acquisition laboratory procedure for human continuous choice

Klapes et al. (2020) built a 37-minute lab game called PRESS-B. Adults pressed two buttons for points on concurrent VI-VI schedules.

The goal was speed. One short session had to give clean matching-law data without days of training.

Every participant showed strong matching. Response ratios tracked reinforcement ratios with a median R² of 0.94.

The whole curve was stable before the session ended. No extra visits were needed.

Pierce et al. (1983) already said humans match, but their evidence came from long, multi-day experiments. PRESS-B delivers the same fit in one sitting, so it extends the review into a faster era.

Thompson (1975) also found matching with humans, yet used time allocation instead of response counts. The new tool keeps the law but swaps the measure, showing the result holds across recording methods.

Bulla et al. (2026) looks like a clash: they claim restricted-operant teaching beats free-operant speed. The studies do not truly disagree. Bulla tested acquisition of new skills; Klapes tested how fast we can map already-stable choice. Different questions, different winners.

If you run basic-choice probes in clinic or classroom, PRESS-B gives you a full matching curve before lunch. Use it to screen sensitivity to reinforcement before you design an intervention. One strong data point in 37 minutes beats three weak ones across three days.