When same-case letter pairs are to be physically matched as “same” or “different,” reaction times (RTs) are generally shorter for “same” responses. The advantage in RT increases when such pairs are intermixed in blocks of trials also containing mixed-case pairs to be matched for name identity. These results have been interpreted as supportive of a two-code hypothesis of letter matching: In pure blocks of same-case pairs, a visual or physical code underlies letter matching, whereas in intermixed blocks, a phonetic or name code must be used for all “different” judgments. The theory predicts, however, that there should be little discrepancy in RT for same-case and mixed-case “different” pairs in intermixed blocks. Here a partial replication of Hellige and Webster (1981) is reported, showing that in fact there is a reliable discrepancy. This outcome poses problems for the two-code hypothesis, although it may be consistent with a “generation” hypothesis of letter matching.
Boles, D.B., Hellige, J.B. Case effects in letter-name matching: A partial replication. Bull. Psychon. Soc. 22, 23–25 (1984).