Children's identity matching and oddity: assessing control by specific and general sample-comparison relations.
Kids can learn "same as" and "different from" as broad rules without mastering every single pair first.
01Research in Context
What this study did
The team worked with children in a lab room. They ran identity and oddity matching trials.
Kids had to pick the picture that "goes with" or "does not go with" a sample. The researchers then checked if the children still chose correctly when the sample changed in small ways.
What they found
Children gave the right answers even when the sample picture was new. They showed the rule "same as" or "different from" without needing every exact pair first.
In short, kids built a broad skill, not just memorized item pairs.
How this fits with other research
Toussaint et al. (2017) later used the same identity-matching idea to teach braille. They added stimulus fading and got fast learning in children with visual impairments. This extends the 1989 finding to a new modality and population.
Peterson et al. (1971) also saw broad stimulus control in preschoolers, but with imitation instead of matching. Both studies show that young children can respond to general rules, not just specific cues.
Stoddard et al. (1967) worked on shape discrimination and found that letting children err on hard items first helped them notice finer differences later. Their focus on error patterns complements the 1989 paper’s focus on rule generalization.
Why it matters
You can teach identity and oddity as general concepts, not endless item sets. Start the lesson, vary the pictures early, and watch for broad control. If the child gets most new items right, move on—no need to drill every pair. This saves table time and builds flexible thinking.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Run 5 identity and 5 oddity trials with new pictures each round; if the child stays above 80%, drop the set and move to newer stimuli.
02At a glance
03Original abstract
After children in Experiments 1 and 2 learned identity matching or oddity, control by sample-comparison relations was assessed. Tests for generalized control displayed novel samples and two comparison stimuli, one identical to the sample. Specific relations were tested with identical or nonidentical sample-comparison stimuli from one set of stimuli and substitute comparisons from either the other training set or from a novel set. When tests displayed identical stimuli, patterns of comparison selection suggested control by generalized identity and oddity. However, selection patterns varied when stimuli were nonidentical and familiar or novel substitute comparisons were used. Therefore, control by specific relations is not a precondition for generalized identity and oddity. One set of training stimuli was used in Experiment 3, and generalized performances occurred again. Moreover, control by specific relations was shown by the oddity subjects and 2 of 6 identity subjects. Generalized and specific control may therefore exist simultaneously. In Experiment 4, selections were irregular on tests displaying substitute comparisons and samples and familiar comparison stimuli; this finding supported the relational account of specific sample-comparison control found in Experiment 3.
Journal of the experimental analysis of behavior, 1989 · doi:10.1901/jeab.1989.51-47