Short-term and long-term effects of reinforcers on choice.

L et al. placed pigeons in a two-key chamber. Each key gave grain on its own variable-interval schedule.

The team watched 90-minute sessions. They noted where each reinforcer fell and how long birds stayed on each key.

They asked: does a reinforcer make the bird stick with that side longer, or does overall choice stay the same?

A grain pellet kept the bird on that key for a few extra seconds. The local stay grew, then faded fast.

Across the whole session the birds’ split stayed locked at the matching law ratio. Short blips did not shift the big picture.

So momentary reinforcers nudge stay length, yet molar preference holds steady.

Attwood et al. (1988) showed that pushing every reinforcer to session-end slashed VI response rates. That paper proves timing matters; L et al. add that the timing effect is brief and washes out in molar data.

Henson et al. (1979) found pigeons still matched choice to obtained rate even when reinforcement was only 50 % reliable. Together with L et al. the rule is clear: organisms track overall rate, not local reinforcer bumps.

Tyrer et al. (2009) used bigger sucrose portions to hike progressive-ratio breakpoints. They changed incentive value; L et al. kept value fixed and only moved delivery time. The pair shows you can tweak either magnitude or timing, but long-term choice still follows the same matching equation.

You now know a single reinforcer works like a tiny magnet: it pulls the client in for a moment, then lets go. Use this during teaching: deliver the item, then quickly prompt the next response to keep momentum. Do not fear that one reinforcer will lock the learner into the wrong choice for the whole session; the molar rate, not the last treat, drives future responding. Track totals across the period, not just after each delivery, when you graph your data.