Task modulates ERP effects of orthographic neighborhood for pseudowords but not words

Abstract

Words and pseudowords from high-density orthographic neighborhoods elicit larger amplitude N400s than similar items from low-density orthographic neighborhoods in the lexical decision task. This pattern could be interpreted as an increase in the amount of lexico-semantic information to be processed or as an increase in difficulty identifying a word (or rejecting a pseudoword) amongst many co-activated alternatives. In order to dissociate between these mechanisms, we compared neighborhood effects between a lexical decision task (LDT) and a letter search task (LST). Behaviorally, we found the standard neighborhood and lexicality effects in the LDT, but no significant effects in the LST. Thus, behavioral responses were sensitive to the decisions required by the respective tasks. Electrophysiologically, we found similar N400 neighborhood effects between tasks for words, but the N400 neighborhood effect for pseudowords was only present in the LDT. Moreover, the effect of neighborhood in the LDT occurred earlier for words than for pseudowords. These nuanced differences in the time course and automaticity of word and pseudoword neighborhood effects lend insight into the processes that underlie N400 effects of orthographic neighborhood and how they unfold over time. We propose that the early neighborhood effects for words across tasks were driven by highly automatized word identification processes that were sensitive to the lateral inhibition generated by orthographic neighbors. In contrast, the later neighborhood effects for pseudowords in the LDT could have been driven by task-specific processes tied to how global lexical activity is used to make a lexical decision.

Keywords: ERPs; Lexical status; N400; Orthographic neighborhood; Task effects.

Figure 1.. Electrode Montage.

Sites highlighted in grey were included in analyses.

Figure 2.. LDT Group: Lexicality Effect.

Grand average ERP waveforms elicited by words (red) and pseudowords (black) in the LDT group. Each vertical tick marks 100 ms and negative is plotted up. The calibration bar marks 2 μV. The scalp voltage maps show the distribution of the lexicality effect (pseudowords-words) in each of the time windows that were analyzed.

Figure 3.. LDT Group: Neighborhood Effects.

Grand average ERP waveforms showing the effect of neighborhood density for words (A) and pseudowords (B) in the LDT group. Each vertical tick marks 100 ms and negative is plotted up. The calibration bar marks 2 μV. Scalp voltage maps show the distributions of the two neighborhood effects (high-low) in the time windows that were analyzed.

Figure 4.. LDT Group: Neighborhood Difference Waves.

Difference waves showing the time course of the effects of neighborhood density (high-low) for words (red) and pseudowords (black) in the LDT group. Each vertical tick marks 100 ms and negative is plotted up. The calibration bar marks 2 μV.

Figure 5.. LST Group: Lexicality Effect.

Grand average ERP waveforms elicited by words (red) and pseudowords (black) in the LST group. Each vertical tick marks 100 ms and negative is plotted up. The calibration bar marks 2 μV. The scalp voltage maps show the distribution of the lexicality effect (pseudowords-words) in each of the time windows that were analyzed.

Figure 6.. LST Group: Neighborhood Effects.

Grand average ERP waveforms showing the effect of neighborhood density for words (A) and pseudowords (B) in the LST group. Each vertical tick marks 100 ms and negative is plotted up. The calibration bar marks 2 μV. Scalp voltage maps show the distributions of the two neighborhood effects (high-low) in the time windows that were analyzed.

Figure 7.. LST Group: Neighborhood Difference Waves.

Difference waves showing the time course of the effects of neighborhood density (high-low) for words (red) and pseudowords (black) in the LST group. Each vertical tick marks 100 ms and negative is plotted up. The calibration bar marks 2 μV.