Alpha and gamma band oscillations index differential processing of acoustically reduced and full forms

Abstract

Reduced forms like yeshay for yesterday often occur in conversations. Previous behavioral research reported a processing advantage for full over reduced forms. The present study investigated whether this processing advantage is reflected in a modulation of alpha (8)(9)(10)(11)(12) Hz) and gamma (30+ Hz) band activity. In three electrophysiological experiments, participants listened to full and reduced forms in isolation (Experiment 1), sentence-final position (Experiment 2), or mid-sentence position (Experiment 3). Alpha power was larger in response to reduced forms than to full forms, but only in Experiments 1 and 2. We interpret these increases in alpha power as reflections of higher auditory cognitive load. In all experiments, gamma power only increased in response to full forms, which we interpret as showing that lexical activation spreads more quickly through the semantic network for full than for reduced forms. These results confirm a processing advantage for full forms, especially in non-medial sentence position.3

Hearing and seeing meaning in noise: Alpha, beta, and gamma oscillations predict gestural enhancement of degraded speech comprehension

“…Studies on unimodal auditory degraded speech comprehension have demonstrated that parietal alpha power is enhanced when speech is degraded (Becker, Pefkou, Michel, & Hervais‐Adelman, ; Drijvers, Mulder, & Ernestus, ; Obleser & Weisz, ; Weisz, Hartmann, Müller, Lorenz, & Obleser, ; Wostmann, Herrmann, Wilsch, & Obleser, ). These results were interpreted as reflecting increased auditory cognitive load when the language processing system is inhibited due to degradation.…”

Not just a function of function words: Distal speech rate influences perception of prosodically weak syllables

“…Similar to words that are lexically vowel-initial (e.g., align), variant vowel-initial word forms (e.g., [ d] for had) may show reduced or totally absent proximal acoustic discontinuities (e.g., glottalization or frication noise for /h/), which could otherwise serve as overt segmentation cues to the start of the word or presence of an extra syllable. The problem of how listeners recover representations under the highly variable pronunciations that occur in casual, everyday speech is a critically important and challenging problem in understanding the neurocognitive mechanisms of speech perception (Alexandrou, Saarinen, Kujala, & Salmelin, 2018;Drijvers, Mulder & Ernestus, 2016;Ernestus & Warner, 2011;Tucker & Ernestus, 2016).…”

Analyzing EEG Signals in Auditory Speech Comprehension Using Temporal Response Functions and Generalized Additive Models

“…The EEG data used for this paper was taken from [13,14]. In three passive listening experiments, right-handed native listeners of Dutch were instructed to listen attentively to the presented speech input and were told that they would get questions about the words and sentences they were about to hear.…”

“…For the filler sentences, there was no specific instruction. The reduced and unreduced verb forms were spliced out of their original sentences and were pasted into the carrier sentence or presented in isolation (these were segmented from the sentences in which these targets occurred at sentence-final position; see [13,14] for more details). This was done to make sure that the reduced and unreduced sentences only differed with respect to the realization of the target verb form.…”