Discriminative functions and over-training as class-enhancing determinants of meaningful stimuli.
Running 500 successive-discrimination trials lifts equivalence-class yields to 85 %, giving you a picture-free way to match the boost once thought to need meaningful stimuli.
01Research in Context
What this study did
College students learned to match abstract shapes into three-member equivalence classes. Some students also practiced telling apart two shapes that looked alike. They kept doing these 'same-different' drills until they hit 500 correct trials.
The team wanted to know if extra discrimination practice could stand in for using a picture that already meant something, like a cat or a tree.
What they found
Students who got the long discrimination run ended up forming 85 % of the classes. That matched the score seen when a meaningful photo is tucked into the set.
Without the extra practice, yields stayed lower. The data say practice alone can replace meaning.
How this fits with other research
Fields et al. (2012) first showed that dropping in one meaningful or easy-to-tell-apart stimulus lifts class yields. The new study keeps the same lab and age group but swaps the boost source: extra trials instead of a meaningful picture.
Weissman-Fogel et al. (2015) went one year further. They gave an abstract shape five prior conditional relations and got the same jump. Both papers agree: the helper is not 'meaning' itself, but the history of telling things apart.
Fields et al. (2021) doubled yields with a timing tweak—separating the response window from the comparisons. Like the target, they prove you can hit high equivalence scores by tuning the drill, not the stimuli.
Why it matters
If you teach conditional discriminations, you can now pick your boost. Add a meaningful icon, stack extra successive-discrimination trials, or separate response timing. Each path lands near 85 % class formation. For clients who lack real-world pictures, just run more 'same-different' trials. You keep the abstract set and still get the gain.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Pick two hard-to-tell-apart stimuli in your set and run extra 'same-different' trials until the learner hits 50 straight correct, then move to equivalence training.
02At a glance
03Original abstract
Likelihood of equivalence class formation (yield) was influenced by pre-class formation of simultaneous and successive discriminations, their mastery criteria, and overtraining of the successive discriminations. Each undergraduate in seven groups attempted to form two 3-node, 5-member equivalence classes (ABCDE). In the pictorial (PIC) group, meaningless nonsense syllables were used as the A, B, D, and E stimuli and meaningful pictures as the C stimuli. Nonsense syllables only were used in the other groups. The abstract (ABS) or 0-0-0 group involved no pre-class training. In the 84-0-0, 84-5-0 and 84-20-0 groups, simultaneous discriminations were trained among C stimuli to a mastery criterion of 84 trials, followed by successive discriminations trained to mastery criteria of 0, 5, and 20 trials, respectively. In the 84-20-0, 84-20-100, and 84-20-500 groups, simultaneous and successive discriminations were trained as noted, followed by overtraining with 0, 100, 500 successive-discrimination trials, respectively. The ABS group produced a 6% yield with the 84-0-0, 84-5-0, and 84-20-0 groups producing further modest increments. Overtraining produced a linear increase in yield, reaching 85% after 500 overtraining trials, a yield matching that produced by classes containing pictures as C stimuli (PIC). Thus, acquired discriminative functions and the overtraining of at least one function can account for class enhancement by meaningful stimuli.
Journal of the experimental analysis of behavior, 2014 · doi:10.1002/jeab.91