Understanding and the listener: Conflicting views.
Skinner's three-part 'understanding' is not the same as Parrott's one relational network—pick your definition before writing language programs.
01Research in Context
What this study did
Fushimi (1990) compared two ways to define 'understanding' in verbal behavior.
Skinner said a listener 'understands' if three things happen: the words echo in the head, the person acts right, and the person can tell why they acted.
Parrott said 'understanding' is simpler: the listener's words and the speaker's words become part of the same relational network.
What they found
The two views do not match.
Skinner splits understanding into separate pieces. Parrott fuses them into one relational act.
If you teach one way, you may miss skills the other way catches.
How this fits with other research
Davison et al. (2003) kept the debate alive. Michael and Malott argued the same clash between Skinner boxes and RFT networks.
Hatton et al. (2004) defended RFT. They said relational operants are just operants shaped by many examples. This backs Parrott's single-network view.
Espinosa (2026) shows how to use Skinner's listener roles in real lessons. The 1990 paper gives the map; Espinosa shows the road.
Why it matters
Pick one definition before you write goals. If you want echoic plus correct action plus self-talk, write three targets. If you want a fluent relational net, probe for derived relations instead. Tell your team which lens you chose so everyone scores the same skill the same way.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Open your current VB program and label each 'understanding' target as echoic, action, or self-talk—then check if you need all three.
02At a glance
03Original abstract
Skinner's (1957, 1974) distinction between three senses of the term understanding is presented. For Skinner, a listener understands if she (a) can repeat back to the speaker what he has said; or (b) can respond appropriately; or (c) knows about the controlling variables. Next, a critique of Skinner's view by Parrott (1984; now L.J. Hayes) is presented. Parrott criticizes the first sense of understanding for simplifying a complex activity; the second for equating understanding with reinforcement mediation; and the third for defining understanding as potential behavior. Next, Parrott's two alternative views are presented. Understanding is (a) having perceptual responses of things when only their "names" are present, and (b) organizing objects and words into relational networks. Lastly, Skinner's and Parrott's views on understanding are evaluated, and Parrott's views are critiqued.
The Analysis of verbal behavior, 1990 · doi:10.1007/BF03392854