Is Reference a Relation of Equivalence or Sameness?
Word–referent links are not equivalence or sameness, so check each direction of transfer instead of assuming symmetry.
01Research in Context
What this study did
Plazas (2025) looked at how we talk about words and their meanings. The paper asked: when a word 'points' to a thing, is that link an equivalence relation or a sameness relation?
The author checked the math rules that define equivalence and sameness. He then tested whether the word–thing link passes those rules.
What they found
The word–referent link fails every rule. It is not reflexive, symmetric, or transitive. So it is neither equivalence nor sameness.
That means we cannot assume that functions will transfer evenly across symbols. What you feel about the word 'dog' may not match what you feel about the animal itself.
How this fits with other research
Pear et al. (1984) warned that the old operant-respondent split needs tweaks, not trashing. Plazas echoes that spirit: keep the SE/RFT toolbox, but drop the symmetry assumption.
Branch et al. (1980) said sloppy verbal behavior nudges us toward mentalism. Plazas gives a clear example—calling a word–thing link 'equivalence' is the kind of slop that hides asymmetry.
Craig (2023) argued momentum theory still helps even when data clash. Plazas takes the same stance: the larger framework survives if we refine its core claim about referential symmetry.
Why it matters
When you test for derived relations, stop expecting perfect bidirectional transfer. Probe each direction separately. If you see uneven emotional responses to pictures versus their printed names, that is normal, not error. Design probes and graphs that show the asymmetry instead of hiding it.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Add a reverse probe trial: after teaching A=B, test B=A with new emotions or responses to see if transfer is lopsided.
02At a glance
03Original abstract
The stimulus equivalence (SE) paradigm has become a central explanatory framework for language and complex symbolic behavior within behavior analysis. Its explanatory power rests on three core assumptions: (1) human symbolic behavior is grounded in the semantic relation between words and their referents; (2) this relation is one of equivalence; and in consequence (3) there is a transfer of stimulus functions between words and their referents. These assumptions are also endorsed by relational frame theory (RFT), although considering equivalence as a consequence of a relation of sameness within a relational frame of coordination. However, this article shows that the referential relation is neither reflexive, symmetrical, nor transitive, and therefore cannot be characterized as one of equivalence or sameness, invalidating (2) and (3). It is also shown that other attempts to support (2) or (3), based on the Fields-Place principle or contextual control, fail to achieve their aim. It is argued that between the behavior of the speaker and the listener, there is a functional asymmetry that grounds the asymmetry of the referential relation, and typical SE and RFT experimental paradigms cannot capture it. Finally, some consequences for the study of the SE phenomenon and the study of symbolic behavior from the perspective of behavior analysis are discussed.
Perspectives on Behavior Science, 2025 · doi:10.1007/s40614-025-00473-1