Laganaro, M. 1 , Valente, A. 1 & Perret, C. 2
1 University of Geneva, FAPSE
2 University of Paris XIII
Single word production has been intensively investigated with neuroimaging techniques allowing high temporal resolution, leading to good estimation of the time course of the underlying processes. The identified time-windows represent an average across speakers, who often display very variable mean reaction times. The question then is which time-windows vary according to the speed of response (production latencies, RTs) across speakers. To answer this question we take advantage of the manipulation of a linguistic variable known to affect picture naming times (word age of acquisition, AoA) and of the combination of waveform and spatio-temporal analysis on stimulus- and response-aligned ERPs (Laganaro and Perret, 2010) in order to cover the entire planning period.
High density EEG recordings were carried out during a picture naming task of 60 early-acquired and 60 late-acquired words, matched on main psycholinguistic variables. The 36 subjects were split in two subgroups according to their mean RTs (fast subjects, RTs = 738ms; slow subjects, RTs = 903ms).
Behavioral analysis showed a significant effect of AoA in both subgroups.
Converging results from waveform analysis and spatio-temporal segmentation indicate differences between fast and slow subjects starting around 200 ms after picture onset. The stable topography (functional microstate, Lehman, 1987; Michel et al., 2009) starting around 200 ms lasts 40 ms longer in slow subjects. The next period of stable topography (starting between 240-270 ms) lasts 80 ms longer in the slow group, with no further differences across groups.
Only the duration of the stable topography starting between 240-270ms correlates with RTs. In conclusion, between-subjects variability seems to be distributed in time-windows associated with lexical selection and phonological encoding processes (Indefrey and Levelt, 2004), with higher contribution of phonological encoding to RTs variability.