Transfer of a novel discriminative function across functional stimulus class members in rats
New responses within an equivalence class may transfer at first but still need extra training to last.
01Research in Context
What this study did
Six rats learned to poke left or right after seeing one light.
The lights were members of a trained class: if A1 then B1, if A2 then B2, and so on.
Next the rats were taught: when you see A1, poke left; when you see A2, poke right.
The team then tested if the left/right rule would jump to the untrained B and C lights.
What they found
Five rats showed the jump on the first probe.
Only two rats kept doing it when the probes were repeated.
The transfer showed up fast, but it broke easily.
How this fits with other research
Davison et al. (2002) saw strong, lasting transfer in adults after many training examples.
The rat data look weaker, but the methods differ: humans got lots of extra practice, rats got almost none.
Silguero et al. (2023) found that conflict inside a class does not kick members out.
Mason’s study adds a new layer: even when the class stays intact, a fresh response may not stick to every member.
Together the papers say: equivalence can hold, yet new functions still need extra support to stay put.
Why it matters
If you teach a client to name “dog” when she sees a picture, do not assume she will also name “dog” when she sees the written word.
Check each new form of the stimulus, and give a few extra trials if the response fades.
This quick probe-and-boost step can save you from later errors in natural settings.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →After the learner masters a conditional discrimination, probe untaught members once; if the response drops, run five more exemplars right away.
02At a glance
03Original abstract
This experiment explored a rodent model of functional class formation by assessing the transfer of a novel discriminative function across class members. Following simple successive discrimination reversal training and consistently strong probe performance indicative of the formation of two six‐member functional classes (X1–X6 and Y1–Y6), subjects were trained on a novel discrimination task to respond in the right‐side port in the presence of odor X1 and in the left‐side port in the presence of Y1. When rats demonstrated high accuracy on the left–right (LR) task as well as maintenance of the functional classes, LR probe sessions were conducted in which X2 or Y2 were presented on some trials as nonreinforced probes to test for transfer of the novel function (left or right responding). The LR probe sessions were conducted in this fashion for each pair of X and Y stimuli that had never been directly trained in the LR discrimination procedure. Transfer of the novel function was observed in five out of six subjects on the initial probe session, but only two rats showed consistent transfer across subsequent probes. The results offer preliminary evidence for transfer of novel function in rats and support further investigation using a similar approach.
Journal of the Experimental Analysis of Behavior, 2025 · doi:10.1002/jeab.70055