Abstract
This study examined the brain activity elicited by the hemispheric asymmetries and morpheme transposition of two-character Chinese words (canonical and transposed word) and pseudowords using event-related potentials (ERPs) with a dual-target rapid serial visual presentation (RSVP) task. Electrophysiological results showed facilitation effects for canonical words with centrally presented visual field (CVF) and right visual field (RVF) presentations but not with left visual field (LVF) presentations, as reflected by less negative N400 amplitudes. Moreover, more positive late positive component (LPC) amplitudes were observed for both canonical words and transposed words irrespective of the visual fields. More importantly, transposed words elicited a more negative N400 amplitude and a less positive LPC amplitude compared with the amplitudes elicited by canonical words for CVF and RVF presentations. For LVF presentations, transposed words elicited a less negative N250 amplitude compared with canonical words, and there was no significant difference between canonical words and transposed words in the N400 effect. Taken together, we concluded that character transposition facilitated the mapping of whole-word orthographic representation to semantic information in the LVF, as reflected by the N250 component, and such morpheme transposition influenced whole-word semantic processing in CVF and RVF presentations, as reflected by N400 and LPC components.
Highlights
The human reading system is a fast, automatic, and comparably robust system, but it is influenced by factors such as letter space, relative position, and lexical organization (Grainger and Whitney, 2004; Rayner et al, 2006)
When examining the visual field effects of each stimulus type separately, pairwise comparisons showed that there were no significant differences between the LVF and RVF in identification accuracies canonical words (p > 0.05) or transposed words (p > 0.05), except for pseudowords (p < 0.01), which showed an RVF advantage
These results indicated that visual field asymmetry and character transposition modulated the Event-related potential (ERP) throughout the word processing stages, beginning at 250–300 ms and ending close to 550–600 ms
Summary
The human reading system is a fast, automatic, and comparably robust system, but it is influenced by factors such as letter space, relative position, and lexical organization (Grainger and Whitney, 2004; Rayner et al, 2006). Significant transposed-letter similarity effects were found independent of the morphological processing of the letter transposition (Beyersmann et al, 2012). Event-related potential (ERP) studies revealed that transposed-letter non-word pairs (e.g., jugde-judge) could produce a lower N250 effect compared with the substituted-letter control condition (e.g., jupte-judge), but no differences were observed. The N400 effect elicited by transposed-letter non-words (e.g., relovution) was reduced compared with that of replacementletter non-words (e.g., retosution) (Carreiras et al, 2007). These findings suggested that non-words created by transposing letters were very effective in activating the form-level and semantic-level representation of their original word regardless of letter transposition. These findings suggested that non-words created by transposing letters were very effective in activating the form-level and semantic-level representation of their original word regardless of letter transposition. Beyersmann et al (2012) provided further evidence for both whole-word access and morphological decomposition at the initial stages of visual word recognition
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
