Graded Delay, Enhanced Equivalence Class Formation, and Meaning

Assigned to 12 groups, 120 college students were taught the baseline relations for three equivalence classes (A -> B -> C -> D -> E) using the simultaneous protocol with trials conducted in matching-to-sample format. In 11 of the 12 groups, all stimuli were abstract shapes; in one group (PIC), the A, B, D, and E stimuli were abstract and the C stimuli were pictures. One group with abstract shapes only (ABS) did not receive preliminary training, while 10 other groups received preliminary training before class formation training. In five groups, participants learned identity relations with the C stimuli using 0-, 1-, 3-, 6-, or 9-s delays. In five others, participants learned arbitrary relations between C and X stimuli using the same five delays. Classes were formed by 70 and 0% of participants in the PIC and the ABS groups, respectively. Class formation increased systematically after preliminary training with the 0-s through 6-s delays and decreased after preliminary training with the 9-s delay. Thus, enhanced class formation was a graded, non-monotonic function of delay duration. For each delay, class formation was greater after establishing arbitrary relations than identity relations. The effects of delay in preliminary training on class formation were similar to their effects on the maintenance of the baseline relations in the derived relation tests. Two opposing processes were posited to account for the delay effects on class formation. Finally, we discussed how class enhancement by meaningful stimuli can also be attributed to their relational functions and delays.