Quantitative analyses of matching-to-sample performance.
Extreme reinforcer ratios break matching-to-sample by turning kids (and pigeons) into position-guessers.
01Research in Context
What this study did
The team ran a three-key matching-to-sample game with pigeons. The birds saw a sample made of tiny black dots, then picked one of two side keys that looked like it.
The twist: the researchers paid the birds with grain on crazy lopsided schedules—think 9:1 or 19:1. They wanted to see if the usual math (the matching law) still predicted which key the bird would peck.
What they found
When the grain odds got extreme, the birds stopped looking at the sample. They just parked on the side that usually paid off—a classic position bias.
Old-school matching equations fell apart. A newer model that adds “signaled differential reinforcement” fit the data. It treats the sample as a cue that can lose control if one comparison is suddenly a gold mine.
How this fits with other research
Cameron et al. (1996) saw the same payoff-tracking in delayed matching, but they blamed long gaps, not extreme ratios. Together the papers show both delay and payoff can weaken sample control.
Bailey (1984) showed kids generalize matching only when each sample owns a stable “name” response. Campbell (2003) adds a warning: even perfect naming can crash if one choice starts paying nine times more.
Locurto et al. (1980) found reinforcing errors in signal detection quickly trashed accuracy. The pigeon MTS study extends that idea—reinforcer ratio alone, not error payoff, can override stimulus control.
Why it matters
If you run matching-to-sample lessons with kids, watch your reinforcer rates like a hawk. Once one comparison earns far more tokens, stickers, or praise, the child may stop looking at the sample and just pick the rich side. Keep ratios close to 1:1, or build in extra trials that re-train attending to the sample after big payouts. Your data will thank you.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Check your token board: if one correct choice earns 5 tokens and the other earns 1, rebalance to 3:2 or add sample-only review trials.
02At a glance
03Original abstract
Six pigeons performed a simultaneous matching-to-sample (MTS) task involving patterns of dots on a liquid-crystal display. Two samples and two comparisons differed in terms of the density of pixels visible through pecking keys mounted in front of the display. Selections of Comparison 1 after Sample 1, and of Comparison 2 after Sample 2, produced intermittent access to food, and errors always produced a time-out. The disparity between the samples and between the comparisons varied across sets of conditions. The ratio of food deliveries for the two correct responses varied over a wide range within each set of conditions, and one condition arranged extinction for correct responses following Sample 1. The quantitative models proposed by Davison and Tustin (1978), Alsop (1991), and Davison (1991) failed to predict performance in some extreme reinforcer-ratio conditions because comparison choice approached indifference (and strong position biases emerged) when the sample clearly signaled a low (or zero) rate of reinforcement. An alternative conceptualization of the reinforcement contingencies operating in MTS tasks is advanced and was supported by further analyses of the data. This model relates the differential responding between the comparisons following each sample to the differential reinforcement for correct responses following that sample.
Journal of the experimental analysis of behavior, 2003 · doi:10.1901/jeab.2003.79-323