Equivalence class formation as a function of the pronounceability of the sample stimulus.

The role of naming in stimulus equivalence was studied by varying the pronounceability of sample stimulus pseudowords. Experiment 1 compared three conditions: in the first, sample stimuli consisted of phonologically correct pseudowords; in the second, sample stimuli consisted of phonologically incorrect words; in the last, sample stimuli consisted of punctuation marks. Subjects exposed to pronounceable stimuli demonstrated equivalence class formation more quickly and with fewer errors than did other subjects. In Experiment 2, subjects were trained in equivalence-class formation using only non-phonological sample stimuli. Half the subjects were exposed to a pretraining procedure in which they read a list of non-phonological pseudowords aloud. Remaining subjects transcribed the same list of pseudowords - a procedure which equated exposure to the pseudowords, but did not necessarily encourage subjects to name them. Subjects who were pretrained with the read-aloud task made significantly fewer errors than those who transcribed the words. These data are consistent with the theory that equivalence class formation is mediated by verbal behavior.