Evidence of syntactic working memory usage in MEG data

Abstract

While reading times are often used to measure working memory load, frequency effects (such as surprisal or n-gram frequencies) also have strong confounding effects on reading times. This work uses a naturalistic audio corpus with magnetoencephalographic (MEG) annotations to measure working memory load during sentence processing. Alpha oscillations in posterior regions of the brain have been found to correlate with working memory load in non-linguistic tasks (Jensen et al., 2002), and the present study extends these findings to working memory load caused by syntactic center embeddings. Moreover, this work finds that frequency effects in naturally-occurring stimuli do not significantly contribute to neural oscillations in any frequency band, which suggests that many modeling claims could be tested on this sort of data even without controlling for frequency effects.

Probabilistic language models in cognitive neuroscience: promises and pitfalls

“…Current applications of probabilistic language models in cognitive neuroscience show that probabilistic language models can be used with hemodynamic and 360 electrophysiological methods and allow researchers to investigate and focus on spatial fingerprints for specific linguistic computations in cortical regions (Willems et al, 2016;Henderson et al, 2016) or to compare predictions of different models against each other on the basis of same neurobiological data, be it fMRI time courses (Brennan et al, 2016), language event-related M/EEG 365 components (Frank et al, 2015;Wehbe et al, 2015), or spectral contents of electrophysiological signals (Van Schijndel and Schuler, 2015;Nelson et al, 2017). The studies employed language stimuli in both auditory and visual modalities and, with the exception of the studies by Frank et al (2015) and Nelson et al (2017), used language stimuli in naturalistic, narrative contexts.…”

“…From the perspective of neurophysiological explanation, current fMRI-based applications stay within what has been dubbed the "cartographic imperative" (Poeppel, 2012) with the goal of tentatively localizing hypothesized computations to gross-level brain areas (as in Willems et al, 2016;Henderson et al, 2016). On the other hand, electrophysiological results are predominantly informing cognitive theories (as in Frank et al, 2015;Van Schijndel and Schuler, 2015). However, it is becoming increasingly clear in cognitive and systems neuro-580 sciences that brain signals are not only indices representing diagnostic evidence for theories cast at the cognitive-computational levels of analyses, but are biophysically meaningful signals reflecting underlying neuronal computations and circuit configurations (Cohen, 2017) occurring at lower levels of spatio-temporal cortical organizations (this is conveyed by the upper part of our schematic in Figure 1).…”

Probabilistic language models in cognitive neuroscience: Promises and pitfalls

“…Current applications of probabilistic language models in cognitive neuroscience show that probabilistic language models can be used with hemodynamic and electrophysiological methods and allow researchers to investigate and focus on spatial fingerprints for specific linguistic computations in cortical regions (Willems et al, 2016;Henderson et al, 2016) or to compare predictions of different models against each other on the basis of same neurobiological data, be it fMRI time courses , language event-related M/EEG components (Frank et al, 2015;Wehbe et al, 2015), or spectral contents of electrophysiological signals (van Schijndel et al, 2015;Nelson et al, 2017). The studies employed language stimuli in both auditory and visual modalities and, with the exception of the studies by Frank et al (2015) and Nelson et al (2017), used language stimuli in naturalistic, narrative contexts.…”

“…From the perspective of neurophysiological explanation, current fMRI-based applications stay within what has been dubbed the "cartographic imperative" (Poeppel, 2012) with the goal of tentatively localizing hypothesized computations to gross-level brain areas (as in Willems et al, 2016;Henderson et al, 2016). On the other hand, electrophysiological results are predominantly informing cognitive theories (as in Frank et al, 2015;van Schijndel et al, 2015). However, it is becoming increasingly clear in cognitive and systems neurosciences that brain signals are not only indices representing diagnostic evidence for theories cast at the cognitive-computational levels of analyses, but are biophysically meaningful signals reflecting underlying neuronal computations and circuit configurations (Cohen, 2017) occurring at lower levels of spatio-temporal cortical organizations (this is conveyed by the upper part of our schematic in Fig.…”

“…van Schijndel et al (2015) investigated the role of syntactic memory load during auditory story comprehension. The strength of spectral coherence of MEG oscillatory neural activity in the 10 Hz (alpha-band) range was taken as a neural indicator of increased working memory usage.…”