Deviations from matching as a measure of preference for alternatives in pigeons.

Researchers gave pigeons two choices of keys to peck. One choice had two keys side-by-side. Only one of those two paid off with grain. The other choice had just one key that always paid.

They wanted to know if the useless second key would wreck the birds’ usual matching pattern. Would birds still peck in the same ratio as the grain they got?

The extra dead key did nothing. The birds kept pecking so their responses matched the grain rate, just like before.

Adding a key that never pays does not bend the matching law.

Glover et al. (1976) saw big swings away from matching when they cut all payoff for one key. The difference: they removed payoff from the whole alternative, not just tacked on an extra key. That explains the seeming clash.

Hastings et al. (2001) later showed that bigger grain pieces also leave matching untouched. Together the two papers say matching stays steady even when you tweak non-rate factors.

Shimp (1967) already proved matching can survive without maximizing. White (1979) adds another boundary: it also survives pointless keys.

When you set up concurrent schedules for kids or clients, extra toys or buttons that never pay off won’t distort the rate-based choice pattern. You can safely add neutral stimuli without re-balancing reinforcer rates.