Priming and response selection processes in letter classification and identification tasks.

Abstract

The processing of isolated visual letters was studied by means of a priming paradigm. In alphabetic (letter vs. nonletter) classification, any letter prime reduced response times to letter targets. Additional facilitation occurred only with primes physically identical to the target. In letter naming, facilitation was seen with primes nominally identical to the target even when they were physically different. This result is not due to phonological priming because phonologically similar primes had no effect on naming times. Primes nominally different from the target but physically similar to it increased naming times. The classification task seems to be performed through the global monitoring of stored visual knowledge of letters. In contrast, the absolute identification of letters appears to rest on a signal-to-noise statistic derived from an abstract encoding of letter identities. Connectionist simulations provide support for these proposals.

Unconscious task set priming with phonological and semantic tasks

“…Consistent with previous studies (Arguin & Bub, 1995;Neumann & Klotz, 1994), the subliminal primes were found to affect identification of the conscious instruction letter in the repetition priming procedure in both experiments, with shorter RTs when the prime was congruent, as opposed to incongruent, with the instruction. However, the extent of this repetition priming effect was greater in Experiment 1 than Experiment 2, when the lag between the prime and target was shorter (36 ms and 84 ms in Experiment 1 and 2, respectively).…”

Visual and Phonological Codes in Letter and Word Recognition: Evidence from Incremental Priming

“…An identity condition and the traditional unrelated condition were added to form a total of four priming conditions: (1) physically identical (ID) primes (e.g., B±B); (2) SIM primes (e.g., c±C); (3) DIS primes (e.g., a±A), and (4) unrelated primes (e.g., x±T). Based on previous results (Arguin & Bub, 1995;Bowers et al, 1998), it was predicted that DIS primes should produce facilitation in naming but not in alphabetic decision. In contrast, ID and SIM primes should produce facilitation in both alphabetic decision and naming.…”

“…For example, Posner and Mitchell (1967) argued that cross-case priming (a±A) constitutes evidence for phonological processing (i.e., the involvement of a name code) because, in cross-case priming, primes and targets have the same name albeit a different visual shape. However, other studies that looked speci®cally for effects of phonetic/acoustic similarity in letter recognition typically failed to ®nd such phonetic/acoustic effects (Arguin & Bub, 1995;Boles & Eveland, 1983;Carrasco, Kinchla, & Figueroa, 1988). Moreover, cross-case priming does not necessarily need to result from overlap in phonological information.…”

“…Moreover, cross-case priming does not necessarily need to result from overlap in phonological information. It could result from primes activating abstract orthographic letter representations that are shared by both upper-and lower-case letters (e.g., Arguin & Bub, 1995). Another possibility would be that primes automatically and rapidly generate the opposite case in memory, without the mediation of a phonological code (Boles, 1992).…”

“…Recently, Bowers, Vigliocco, and Haan (1998) followed up on Arguin and Bub's (1995) ®nding. They investigated cross-case letter priming with both visually similar (SIM) and visually dissimilar (DIS) primes (e.g., SIM: x±X; DIS: a±A).…”