The Time Course of Visual Letter Perception

Abstract

We describe a novel method for tracking the time course of visual identification processes, here applied to the specific case of letter perception. We combine a new behavioral measure of letter identification times with single-letter ERP recordings. Letter identification processes are considered to take place in those time windows in which the behavioral measure and ERPs are correlated. A first significant correlation was found at occipital electrode sites around 100 msec poststimulus onset that most likely reflects the contribution of low-level feature processing to letter identification. It was followed by a significant correlation at fronto-central sites around 170 msec, which we take to reflect letter-specific identification processes, including retrieval of a phonological code corresponding to the letter name. Finally, significant correlations were obtained around 220 msec at occipital electrode sites that may well be due to the kind of recurrent processing that has been revealed recently by TMS studies. Overall, these results suggest that visual identification processes are likely to be composed of a first (and probably preconscious) burst of visual information processing followed by a second reentrant processing on visual areas that could be critical for the conscious identification of the visual target.

The time course of visual influences in letter recognition

“…This small time window (i.e., 120 ms) is certainly computationally too short to encompass all of these nonvisual processes, suggesting that visual and nonvisual processes cannot be encapsulated, and have to interact earlier in time (Carreiras et al, 2014). Similarly, the fact that abstract or phonological effects have been reported much earlier than 290-300 ms (e.g., Madec et al, 2012;Petit et al, 2006) also indicates that visual and nonvisual levels of processing are coactivated at some point in time during letter identification, and would therefore interact following a cascaded interactive-activation framework (McClelland & Rumelhart, 1981).…”

“…It has been proposed that visual templates would be processed around 150-170 ms (Rey, Dufau, Massol, & Grainger, 2009), followed by abstract and phonological processing around 220 ms (Petit et al, 2006). Recent evidence has suggested that phonological processes (i.e., access to the letter's name) may start earlierthat is, between 150 and 190 ms-and that letter identification processes would be supported by recurrent interactions between visual and phonological codes until response selection (Madec, Rey, Dufau, Klein, & Grainger, 2012).…”

The unbearable articulatory nature of naming: on the reliability of word naming responses at the item level

“…This result could be explained by assuming that the amount of variance devoted to visual identification processes is not fixed, either, among word categories. Indeed, if these central processes are fast and efficient (as for high-frequency monosyllabic words), then the total amount of the variance in naming could be influenced less by these processes, and the contribution of articulatory factors might therefore be stronger (for a similar result in single-letter naming, see Madec et al, 2012). Conversely, if central processes are slower and more time consuming, then they would account for a larger amount of the total variance, and as a consequence, the amount of variance captured by onset phonemes would decrease.…”

“…Finally, in a recent study using a large database of letternaming and delayed-naming response times, Madec, Rey, Dufau, Klein, and Grainger (2012) found that the delayed-naming times predicted 79 % of the naming time variance. This result suggests that output articulatory processes might be responsible for a much greater percentage of the total item variance than was initially estimated on the basis of the phonetic properties of the initial phoneme (which might, therefore, not capture all of the articulatory-related variance).…”

“…The new database is composed of naming response times for 400 words randomly extracted from the database, in order to directly compare item means between the old and new databases. This naming task is combined with the same conditional delayed-naming procedure that had been used in Madec et al (2012). For half of the words (i.e., 200), after the presentation of a target signal (a green circle), participants had to repeat the word that they had just read aloud.…”