On the time course of letter perception: a masked priming ERP investigation

Abstract

In an experiment measuring event-related brain potentials (ERPs), single-letter targets were preceded by briefly presented masked letter primes. Name and case consistency were manipulated across primes and targets so that the prime was either the same letter as the target (or not), and was presented in the same case as the target (or not). Separate analyses were performed for letters whose upper- and lowercase forms had similar features (or not). The results revealed an effect of prime-target visual similarity between 120 and 180 msec, an effect of case-specific letter identity between 180 and 220 msec, and an effect of case-independent letter identity between 220 and 300 msec. We argue that these ERP results reflect processing in a hierarchical system for letter recognition that involves both case-specific and case-independent representations of alphabetic stimuli.

Figure 1

A typical trial.

Figure 2

Electrode montage and analysis columns. Twenty-nine active tin electrodes were held in place by an elastic cap (ElectroCap International, Inc., Eaton, OH) and maintained < 5 KΩ. Additional electrodes were placed below the left eye (LE) and beside the right eye (HE) to monitor eye movements and blinks. All electrodes were referenced to the left mastoid (A1), and the right mastoid (A2) was recorded actively to detect left/right mastoid asymmetry (none was detected). the four analysis columns (C1, C2, C3, and midline) are indicated by the gray bars interconnecting the various anterior/posterior sites.

Figure 3

(A) target ERPs from the four all letter scheme conditions. Note that the time intervals for analysis as well as prime and target onset latencies are indicated on the time axis in the lower left-hand corner and that target onset is indicated by the calibration bar. (B) Target ERPs at the CP1 electrode site for the four all letter scheme conditions. the three analysis epochs are color coded and the two primary ERP components analyzed are labeled. In the first epoch (brown) there are clear effects of letter case. At the beginning of the second epoch (blue) the same-name/same-case condition undergoes a much stronger deflection than any of the other conditions. In the third epoch (pink), the different-name conditions (for same and different case) start to pattern together and are clearly distinct from both of the same-name conditions.

Figure 4

Target ERPs at two posterior and two anterior sites contrasting same-name letters that in their upper- and lowercase versions are either similar (left) or dissimilar (right).