Preference and discrimination between response-dependent and response-independent schedules of reinforcement.

Brinker et al. (1975) let quail choose between two keys. One key gave food only after a peck. The other gave food on a timer, no peck needed.

The birds could switch any time. The team counted pecks and time on each key for many sessions.

The quail did not stick with either key. They pecked more on the response-dependent side, but they did not stay there.

In plain words, higher peck rates did not mean the birds liked that schedule better.

Neuringer et al. (1968) saw pigeons match their time to the rate of response-independent food. The birds clearly preferred the richer side. P et al. found no such preference in quail. The clash fades when you note the species swap and the different measure: time matching versus stay-put choice.

Parrott (1984) later showed that pigeons can flip their choice when the rules inside each schedule change. That work extends P’s point: schedule labels like "dependent" or "independent" don’t predict choice by themselves; the local details matter.

Green et al. (1987) added that pigeons pick the more reliable schedule when probability differs. Together these papers tell us animals weigh many schedule features, not just whether a response is required.

For BCBAs, the takeaway is to watch what clients actually choose, not what we think they should prefer. A token board that demands a response may boost work rate, yet the learner might still drift to a no-response activity. Check time allocation, not just response speed, before you claim a reinforcer is preferred.