When ultrarapid is ultrarapid: on importance of temporal precision in neuroscience of language

• Published in: Frontiers in human neuroscience 2015-10, Vol.9, p.576-576
• Main Authors: Shtyrov, Yury Y, Stroganova, Tatyana A
• Format: Article
• Language: English
• Subjects: Acoustics; Brain; Brain research; Cognition & reasoning; Information processing; Language; Language comprehension; Magnetoencephalography; Neocortex; Nervous system; Neuroscience; Neurosciences; Semantics; Time course
• ISSN: 1662-5161; 1662-5161
• DOI: 10.3389/fnhum.2015.00576

[...]this creates psycholinguistic variance, when stimuli diverge in their linguistic features, including word recognition parameters in spoken words. While later deflections (e.g., N400, P600) are smeared by such averaging but, being large in amplitude and long-lasting, still survive it, this strategy could be fatal for capturing the earliest short-lived transient small-scale activity. [...]to capture the entire neural dynamics of language processing, it is important to (1) maximally reduce stimulus variance, e.g., by using a fixed set of tightly controlled stimuli, and (2) time-lock electrophysiological response to key psycholinguistic markers in the auditory stream, most importantly—to the point in time, when the available information allows for differentiating the stimulus from other similar sounds and, ultimately, for identifying it. The vast majority of stimuli were meaningless pseudowords fully sharing their onsets with the critical verbs and nouns. [...]if the semantic information were available during the onset, the similar semantically-specific motor activations should also take place for the frequent pseudowords as, up to the disambiguation point, they were identical to the words. The very same model, however, also suggested that semantic information is only processed at 300–500 ms, in line with M350 and N400 research (Embick et al., 2001; Stockall et al., 2004; Kutas and Federmeier, 2011). [...]even if the logic of recalculating brain responses to the word onset were correct and the effects were indeed in the classical N400 time range, the very same classical framework would place them together with the rest of lexico-semantic dynamics, and not with post-comprehension phenomena.