Stability of daily preference across multiple individuals

Kelley and colleagues ran paired-stimulus preference assessments day after day. They wanted to know if today’s top item would still win tomorrow.

The team tracked how each person’s rankings moved across sessions. They used simple correlations to see if the order stayed put.

The ranks held steady. High correlations showed that what a person picked first on Monday was still near the top later in the week.

In plain words, your last assessment is probably still valid, so you can grab those top items without re-testing every day.

Gilroy et al. (2021) took the idea one step further. They added an elasticity probe to see which highly preferred items actually keep clients working when the schedule gets lean. Kelley tells you the list is stable; Gilroy tells you which items on that list are worth their weight in tokens.

Fahmie et al. (2013) looked similar but came out unsure. They tried brief extinction probes and saw rankings flip. The difference is method: Kelley used repeated paired-stimulus trials, while A et al. used short disruption tests that may not reflect true value.

Mace et al. (1990) and LeBlanc et al. (2003) both showed that richer schedules make behavior tougher to disrupt. Kelley’s stable ranks fit right in: if preference holds, you can safely assign the same high-rate reinforcers and expect the same sturdy persistence.

You can stop running daily preference checks. Trust last week’s top three and spend that time on teaching or play instead. Just re-test when the client’s motivation clearly shifts or when the task gets harder and you need the extra punch Gilroy’s elasticity check provides.