Electrophysiology Reveals the Neural Dynamics of Naturalistic Auditory Language Processing: Event-Related Potentials Reflect Continuous Model Updates

Abstract

The recent trend away from ANOVA-based analyses places experimental investigations into the neurobiology of cognition in more naturalistic and ecologically valid designs within reach. Using mixed-effects models for epoch-based regression, we demonstrate the feasibility of examining event-related potentials (ERPs), and in particular the N400, to study the neural dynamics of human auditory language processing in a naturalistic setting. Despite the large variability between trials during naturalistic stimulation, we replicated previous findings from the literature: the effects of frequency, animacy, and word order and find previously unexplored interaction effects. This suggests a new perspective on ERPs, namely, as a continuous modulation reflecting continuous stimulation instead of a series of discrete and essentially sequential processes locked to discrete events.

M/EEG analysis of naturalistic stories: a review from speech to language processing

“…Nonetheless, the naturalistic paradigms remain crucial for understanding speech and language processing in its full complexity, where interactions between both local linguistic and global contextual features, both linguistic (cf. Alday, Schlesewsky, & Bornkessel-Schlesewsky, 2017) and paralinguistic (Tromp, Peeters, Meyer, & Hagoort, 2017), matter as well as their constant, continuous overlap (cf. Steinberg, Truckenbrodt, & Jacobsen, 2012, who found that stimulus splicing and the resultant misleading formant transitions impact phonological processing).…”

“…The authors used ridge regression to calculate the TRF and found a negativity with a latency and distribution corresponding to a traditional N400. Broderick and colleagues hesitated to use the N400 label, however, as they had no cloze probability manipulation; nonetheless, in light of more recent N400 perspectives (Kutas & Federmeier, 2011) beyond discrete components (Alday et al, 2017), the resulting component can safely be classified as an N400. Furthermore, noise and cocktail-party attention manipulations attenuated this response.…”

“…In contrast to the above, Alday et al (2017) used a more traditional ERP analysis, focusing on mean EEG voltage in a fixed time window (300-500 ms), time locked to word onset. Although basic results were presented for an rERPperspective on TRF (no autocorrelation, regression coefficients interpreted as response), the core focus was complex interactions in a longer (23 minute) narrative in German.…”

“…All of them are based essentially on linear models. Alday et al (2017) and Hale et al (2018) both used a straightforward application of contemporary single-trial ERP analysis with mixed-effects (and two-stage) regression models to accommodate the covariates outside of experimental control (cf. Sassenhagen & Alday, 2016).…”

Phasic alerting increases visual processing speed in amnestic mild cognitive impairment

“…Compared to traditional null-hypothesis significance testing, i.e. repeated-measures ANOVAs, multilevel models do not only address the potential significance of a difference between conditions but estimate the difference itself (for more information about multilevel modelling see Alday, Schlesewsky, & Bornkessel-Schlesewsky, 2017; for general information regarding estimation see Cumming, 2014) (Alday et al, 2017). Parameters estimates can be interpreted as statistically significant (in line with traditional null-hypothesis significance testing) if their credible intervals do not contain zero.…”