Neural tracking of linguistic and acoustic speech representations decreases with advancing age

Abstract

Background: Older adults process speech differently, but it is not yet clear how aging affects different levels of processing natural, continuous speech, both in terms of bottom-up acoustic analysis and top-down generation of linguistic-based predictions. We studied natural speech processing across the adult lifespan via electroencephalography (EEG) measurements of neural tracking.

Goals: Our goals are to analyze the unique contribution of linguistic speech processing across the adult lifespan using natural speech, while controlling for the influence of acoustic processing. Moreover, we also studied acoustic processing across age. In particular, we focus on changes in spatial and temporal activation patterns in response to natural speech across the lifespan.

Methods: 52 normal-hearing adults between 17 and 82 years of age listened to a naturally spoken story while the EEG signal was recorded. We investigated the effect of age on acoustic and linguistic processing of speech. Because age correlated with hearing capacity and measures of cognition, we investigated whether the observed age effect is mediated by these factors. Furthermore, we investigated whether there is an effect of age on hemisphere lateralization and on spatiotemporal patterns of the neural responses.

Results: Our EEG results showed that linguistic speech processing declines with advancing age. Moreover, as age increased, the neural response latency to certain aspects of linguistic speech processing increased. Also acoustic neural tracking (NT) decreased with increasing age, which is at odds with the literature. In contrast to linguistic processing, older subjects showed shorter latencies for early acoustic responses to speech. No evidence was found for hemispheric lateralization in neither younger nor older adults during linguistic speech processing. Most of the observed aging effects on acoustic and linguistic processing were not explained by age-related decline in hearing capacity or cognition. However, our results suggest that the effect of decreasing linguistic neural tracking with advancing age at word-level is also partially due to an age-related decline in cognition than a robust effect of age.

Conclusion: Spatial and temporal characteristics of the neural responses to continuous speech change across the adult lifespan for both acoustic and linguistic speech processing. These changes may be traces of structural and/or functional change that occurs with advancing age.

Keywords: Aging; Lifespan; Linguistic processing; Neural tracking; Speech processing; Speech understanding.

Fig. 1

Different electrodes selections: frontal (red), centro-parietal (green) and temporal (blue). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 2

The distribution of age across the participants, pure tone average (PTA), Stroop color-word test (SCWT) scores and reading span test (RST) across age. The data points of older adults ($>$ 60 years old) are represented by orange triangles, while those of the youngest adults ($<$ 40 years old) are represented by green squares. All other participants, referred to as middle-aged participants, are represented by grey dots. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 3

Top panel A-D: Age affects acoustic neural tracking in regions of interest and interacts with hemisphere. A. Average topography of younger participants below 40 years of age and older participants above 60 years of age for visualization purpose. B. A linear mixed effects model revealed a main effect of age over 3 ROIs, which did not differ from each other. C. Interaction effect between age and hemisphere when values are averaged across all left and all right hemisphere electrodes respectively. D. Interaction effect between age and hemisphere when values are averaged across left temporal and right temporal electrodes respectively. Bottom panel E-G: Age affects phoneme- and word-level linguistic neural tracking. E. Average topography of the younger participants and older participants for visualization purpose. F. Linear age effect when values are averaged across all electrodes. G. A linear mixed effects model revealed an interaction between age and ROI. Post-hoc comparisons showed that the centro-parietal ROI is significantly different from the frontal one. The insets in B, C, D, F and G show the EEG electrodes included in the analysis. All statistics are reported in Table 3. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 4

Older adults show shorter latencies for the early acoustic response. The neural responses to acoustic speech representations are visualized as a function of age for both speech representations: spectrogram (left) and acoustic onsets (right). A. The average temporal response functions across frontocentral electrodes (indicated on the middle inset) for younger (green), middle-aged (grey), and older (orange) participants. The shaded area indicates the time region used to find the peak topography. B. This panel shows the corresponding peak topographies associated with the peak found in the grey vertical panel. C. The corresponding normalized peak topographies associated with the peak found in the grey vertical panel. To test whether the normalized topographies significantly differed, the mcCarthy Wood method was applied (for more details see Section 2.10). D. The decrease in neural response latency as a function of age. The grey lines indicate the model predictions and 95% confidence interval of how the response latency increases with age. The data points of adults above 60 years (older) are annotated with the orange triangles, while those of the adults below 40 years (younger) are annotated with the green squares. Not all subjects showed a prominent peak; therefore, the plot does not show these data points. The insets show the corresponding peak topographies for respectively the younger and older subjects. n.s. = not significant. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 5

Older adults show an enhanced response to linguistic representations in frontocentral regions. The neural responses to linguistic representations are visualized in function of age for the three linguistic speech representations: phoneme surprisal, (top), phoneme entropy, word surprisal and word frequency (bottom). The left plot shows the average temporal response functions across frontocentral electrodes (indicated on the inset of the top plot) for younger (green), middle (grey), and older-aged (orange) participants. The shaded area indicates the standard error of the average TRF. The lines above the TRFs indicate where the TRF is significantly negative for each age category (annotated in the corresponding color). The red bold lines underneath the TRFs indicate where the TRFs are significantly different between younger and older adults. The right-sided plots show the corresponding peak topographies associated with the peaks at the time of the dashed grey line in the left-sided plots. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)