The formation of a generalized categorization repertoire: effect of training with multiple domains, samples, and comparisons.
Teach with many examples from many settings, but limit the choices on each trial to grow true generalization.
01Research in Context
What this study did
Fields et al. (2002) taught adults to sort new pictures into groups. They used matching-to-sample drills with shapes, colors, and lines.
Some people saw many kinds of items and several examples per kind. Others saw fewer. The team then tested who could sort brand-new pictures without help.
What they found
More training domains and more sample examples helped people sort new pictures on their own.
Surprise: giving too many choice pictures during training hurt this flexible sorting.
How this fits with other research
Luciano et al. (2007) saw the same boost with babies. Their 15-month-old formed groups after many matching examples, even before talking. This shows the trick works before language.
van Timmeren et al. (2016) got the same curve with birds. Nutcrackers needed 64-128 examples to reach peak concept learning, just like the humans in Lanny’s lab. The set-size rule crosses species.
Brown et al. (2025) looked rare items in rats. Fewer trials on rare shapes slowed learning. Together these papers warn: give enough examples, but don’t swamp the learner with extra choices.
Why it matters
When you teach categories like “food” versus “not food,” rotate many real items across places and times. Keep the number of comparison cards or objects small during drills. This mix—rich examples, few choices—builds flexible sorting that holds with new items later.
Get CEUs on This Topic — Free
The ABA Clubhouse has 60+ on-demand CEUs including ethics, supervision, and clinical topics like this one. Plus a new live CEU every Wednesday.
Pick one target class, gather five real exemplars from different rooms, and run matching trials with only two comparison items.
02At a glance
03Original abstract
The present experiment explored the effects of three variables on the spontaneous categorization of stimuli in perceptually distinct and novel domains. Each of six stimulus domains was created by morphing two images that were the domain endpoints. The endpoints of the domains were male and female faces, two abstract drawings, a car and a truck, two banded-elevation satellite land images, a tree and a cat, and two false-color satellite images. The stimulus variants at each end of a domain defined two potential perceptual classes. Training was conducted in a matching-to-sample format and used stimuli from one or two domains, one or three variants per class as samples, and one or three variants per class as comparisons. The spontaneous categorization of stimuli in the untrained stimulus domains showed the emergence of a generalized categorization repertoire. The proportion of spontaneously categorized stimuli in the new domains was positively related to the number of domains and samples used in training, and was inversely related to the number of comparisons used in training. Differential reaction times demonstrated the discriminability of the stimuli in the emergent classes. This study is among the first to provide an empirical basis for a behavior-analytic model of the development of generalized categorization repertoires in natural settings.
Journal of the experimental analysis of behavior, 2002 · doi:10.1901/jeab.2002.78-291