Improving speech perception in noise in young and older adults using transcranial magnetic stimulation
Introduction
Older adults often report difficulties following conversations in noisy environments (e.g., at the restaurant or during family reunions), which can hinder social interactions (Aydelott et al., 2010, O, 1988). Several studies have shown that peripheral hearing only partially explains these difficulties, which suggests a contribution of central mechanisms (e.g. Anderson et al., 2013, Dubno et al., 2008, Fostick et al., 2013, Humes and Dubno, 2010, Pichora-Fuller and Souza, 2003). One possibility is that speech perception in noise (SPiN) difficulties are triggered by age-related decline in brain regions supporting central auditory mechanisms in the primary auditory cortex (transverse temporal gyrus; TTG). Alternatively, these difficulties could be related to aging of the brain networks supporting speech functions. Indeed, in addition to involving the primary auditory cortex, processing speech also activates regions involved in phonological (e.g., the superior temporal cortex; STC), motor (e.g., the primary motor and premotor cortex; PM) and lexical (e.g., the middle temporal gyrus; MTG) processes, as well as regions of the executive network, such as the middle frontal cortex, cingulate cortex, frontal operculum and anterior insula (e.g. Adank, 2012, Alain et al., 2018, Eckert et al., 2016).
Consistent with the notion of a role for higher-order processes in age-related decline in SPiN, several brain imaging studies have shown that structural decline in brain areas and white matter tracts that support phonological (e.g., the bilateral STC), motor (e.g., the PM), lexical (e.g., the MTG), and attention-related processes (e.g., the insula) is associated with speech performance decline (e.g. Bilodeau-Mercure et al., 2015, Eckert et al., 2008, Sheppard et al., 2011, Tremblay et al., 2021, Wong et al., 2010). For instance, in a group of older adults, accuracy during a sentence repetition task was found to be predicted by cortical thickness of the left superior frontal gyrus and hemispheric volume of the left pars triangularis gyrus (Wong et al., 2010).
Differences in cerebral activation patterns between young and older adults have also been found during SPiN tasks within auditory, phonological and lexical processing areas (e.g. Du and Alain, 2016, Hwang et al., 2007, Manan et al., 2015, Manan et al., 2017, Tremblay et al., 2020, Wong et al., 2009), as well as sensorimotor (e.g. Bilodeau-Mercure et al., 2015, Du and Alain, 2016, Eckert et al., 2008, Manan et al., 2017, Peelle et al., 2010) and attention-related areas (e.g. Bilodeau-Mercure et al., 2015, Du and Alain, 2016, Eckert et al., 2008, Peelle et al., 2010; Vaden et al., 2015, Wong et al., 2009). For instance, a study from our group showed a detrimental indirect effect of age on syllable repetition errors (Bilodeau-Mercure et al., 2015). In that study, aging was associated with lower activation in the left anterior insula, which in turn was associated with decreased accuracy. In a recent study, it was found that when performance in a syllable identification task was matched between younger and older adults, older adults showed higher activity in areas including the bilateral STG, MTG, MFG and precentral gyrus, as well as the left ventral premotor cortex (PMv) and left IFG, suggesting that compensation within speech processing areas contributed to maintaining performance (Du & Alain, 2016). Another study found that the bilateral anterior cingulate cortex (ACC) and the left MFG were more engaged in older adults during degraded word recognition for correct compared to incorrect answers (Eckert et al., 2008). Together, brain imaging studies show that the network supporting SPiN declines with age, and that this decline is associated with a decline in performance. The question that we address here is whether SPiN performance can be enhanced using non-invasive brain stimulation methods.
Non-invasive brain stimulation methods, such as transcranial magnetic stimulation (TMS), can induce beneficial short and longer-term plasticity in the brain, which can lead to enhanced performance in a variety of domains, including motor performance (e.g. Hoyer and Celnik, 2011, Lüdemann-Podubecká et al., 2015, Schambra, 2018), cognition (e.g. Guse, Falkai, & Wobrock, 2010; Kim et al., 2019, Widhalm and Rose, 2019), hearing (e.g. Chen et al., 2020, Schoisswohl et al., 2019, Soleimani et al., 2016), and speech/language (e.g. Devlin and Watkins, 2007, Li et al., 2020). Intermittent theta-burst stimulation (iTBS), a repetitive TMS (rTMS) protocol that can increase cortical excitability, is a promising method to enhance performance during speech tasks in healthy adults and in those with communication disorders. Consistent with this notion, one study has shown that iTBS over the pIFG increased accuracy in sentence repetition in healthy young adults (Restle, Murakami, & Ziemann, 2012). Another study showed enhanced vocal pitch regulation in healthy young adults after iTBS over the right somatosensory laryngeal cortex, during a pitch-matching singing task with masked feedback (Finkel et al., 2019). Several studies of post-stroke aphasic patients have shown that language functions (e.g., semantic fluency, picture naming, auditory comprehension) can be improved by applying iTBS over the left IFG or left temporal cortex in young and older adults (Griffis et al., 2016, Szaflarski et al., 2018, Szaflarski et al., 2011, Versace et al., 2019, Vuksanović et al., 2015). Together, these studies suggest that the adult speech/language system is plastic, and that performance can be boosted using faciliatory rTMS such as iTBS.
To our knowledge, iTBS has never been used to induce changes in the speech network in healthy older adults. A limited number of studies have shown that iTBS applied to the motor cortex can successfully increase cortical excitability in healthy older adults (e.g. Dickins et al., 2015, Gedankien et al., 2017, Young-Bernier et al., 2014). Yet, a recent meta-analysis reported reduced motor evoked potential (MEP) amplitudes and longer latencies in older compared to younger individuals after single-pulse, paired pulse or rTMS was applied (Tang et al., 2019). Specifically, the analysis of 20 studies using electromyography to measure cortical excitability in the primary motor cortex showed that MEP responses to TMS had a significantly lower amplitude in elderly compared to younger adult groups, though the MEP responses were significant in both groups. Another analysis including seven studies showed that post TMS MEP latency was delayed in the elderly group compared to the younger group. Although MEP activity measures can be influenced by age-related factors other than brain plasticity itself (e.g., skin and skull characteristics), at least one study has shown that MEP amplitude changes after motor cortex iTBS correlate with TMS-evoked EEG potentials. This suggests that post-iTBS MEP changes are representative of brain activity changes (Gedankien et al., 2017). Together with findings from clinical populations, these findings suggest that TMS can induce neuroplasticity in the aging brain, though the potential for plasticity may be reduced in older compared to younger adults. Additional evidence is needed regarding the potential for plasticity within specific functional systems such as the speech/language system in the aging brain.
The main objectives of this two-part study were (1) to investigate the mechanisms underlying age-related SPiN decline, and (2) to determine if SPiN performance in young and older adults can be enhanced by excitatory rTMS to two areas involved in processing sublexical speech: the left posterior superior temporal sulcus or pSTS, and the left ventral premotor cortex or PMv. The selection of these areas was based on knowledge of their role in SPiN, and prior evidence that TMS to these regions can successfully induce behavioural changes in healthy young adults. The left pSTS is involved in sublexical phonological processing (e.g. Hickok and Poeppel, 2007, Turkeltaub and Coslett, 2010). For instance, a recent study has shown that inhibitory TMS to this region is associated with phonological errors during auditory word comprehension, syllable repetition, syllable identification and pseudo-word repetition (Murakami, Kell, Restle, Ugawa, & Ziemann, 2015). Another study found that rTMS applied to the anterior STS led to poorer performance during sentence repetition in noise (Kennedy-Higgins, Devlin, Nuttall, & Adank, 2020). The left PMv is involved in speech perception and comprehension (e.g. McGettigan and Tremblay, 2018, Pulvermuller and Fadiga, 2010, Tremblay and Small, 2011, Walenski et al., 2019). Although the specific contribution of this region is still unclear and a subject of debate, previous studies have shown that inhibitory rTMS to this region is associated with reduced speech perception performance (Krieger-Redwood, Gaskell, Lindsay, & Jefferies, 2013; I. G. Meister et al., 2007, Sato et al., 2009) or enhanced selective adaptation to speech during speech perception (Grabski, Tremblay, Gracco, Girin, & Sato, 2013).
The specific objectives of Experiment 1 were to develop and test an age-sensitive sub-lexical SPiN test and to examine the impact of cognition and hearing on performance at this test. Our main hypothesis was that aging would be associated with reduced SPiN performance operationalized as lower accuracy and longer reaction times (RT), after controlling for hearing and cognition. The specific objective of Experiment 2 was to determine whether excitatory rTMS can enhance SPiN in younger and elderly adults via stimulation of the left pSTS and/or left PMv. We hypothesized that performance gain would be more limited in older compared to younger adults because of known reduced plasticity in the aging brain. An effect of target (pSTS, PMv) would suggest that one region has a stronger functional contribution to SPiN and might represent a better option to reduce or prevent SPiN decline in aging using non-invasive brain stimulation.
Section snippets
Participants
22 healthy native Canadian French speakers were recruited through the laboratory database, emails, posts on the lab website (www.speechneurolab.ca), lab Facebook page (https://www.facebook.com/speechneurolab/) and flyers distributed in various institutions, including shops and retirement homes throughout Québec City. One participant was excluded from the analyses due to inability to complete the main task. The remaining participants were aged 20–85 years (mean 53.33 ± 20.52 years). A telephone
Participants
A sample of 34 healthy right-handed healthy native French speakers aged 32–79 years (M = 57.35, SD = 14.69) was recruited through emails sent to the university community and the Centre intégré universitaire de santé et des services sociaux de la Capitale-Nationale, posts on the lab website (www.speechneurolab.ca) and Facebook page (https://www.facebook.com/speechneurolab/) and flyers distributed in various institutions, including shops and retirement homes throughout Québec City, as well as
Discussion
The objectives of this study were to determine if SPiN performance in young and older adults can be enhanced by excitatory rTMS to two areas involved in processing sublexical speech (i.e., the left pSTS, and the left PMv). To achieve this goal, we created a reliable syllable discrimination task. The results of Experiments 1 show that older adults are less accurate in our syllable discrimination task. In Experiment 2, we found this same age effect on accuracy in the baseline condition, which
Limitations
One potential limitation of this experiment was that TMS sessions were delivered on the same day. While this design has the advantage of eliminating variability related to the participant’s health, mood and state of mind during the administration of the tests, the TMS effects could have interacted (i.e., spillover effect). We controlled for this by 1) adding a delay of at least 60 min after a real stimulation before performing a second session (e.g. Chung, Hill, Rogasch, Hoy, & Fitzgerald, 2016
Conclusion
Our TMS study is the first to show that hugely prevalent age-related SPiN difficulties can be reduced by enhancing cortical excitability within the speech-processing network, especially when targeting the left PMv. Importantly, initial performance—not age—was the main driving factor for TMS-induced performance improvement. This study paves the way for the development of approaches to enhance speech processing using neurostimulation methods. Future studies are needed to determine how to maximize
Funding
This work was supported by P.T.’s grants from the Natural Sciences and Engineering Research Council of Canada [RGPIN-2019-06534] and the Canadian Foundation for Innovation [31408]. P.T. also holds a Career Awards from the “Fonds de Recherche du Québec – Santé” (FRQS) [35016]. V.B. was supported by fellowships from the CERVO foundation and from Université Laval (Department of Rehabilitation).
CRediT authorship contribution statement
Valérie Brisson: Conceptualization, Methodology, Investigation, Project administration, Formal analysis, Visualization, Writing - original draft, Data curation. Pascale Tremblay: Conceptualization, Funding acquisition, Methodology, Investigation, Supervision, Resources, Project administration, Writing - review & editing, Data curation.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgments
We thank all participants. Thanks also to Maxime Perron for his precious help developing the stimuli and task, to Audrey Desjardins and Alison Arseneault for their contribution to participant recruitment and data collection, and to Catherine Fontaine-Lavallée for administrative support.
References (113)
The neural bases of difficult speech comprehension and speech production: Two Activation Likelihood Estimation (ALE) meta-analyses
Brain and Language
(2012)- et al.
A dynamic auditory-cognitive system supports speech-in-noise perception in older adults
Hearing Research
(2013) - et al.
Cortical oscillations and sensory predictions
Trends in Cognitive Sciences
(2012) - et al.
A meta-analysis of the effects of aging on motor cortex neurophysiology assessed by transcranial magnetic stimulation
Clinical Neurophysiology: Official Journal of the International Federation of Clinical Neurophysiology
(2016) - et al.
Use of theta-burst stimulation in changing excitability of motor cortex: A systematic review and meta-analysis
Neuroscience & Biobehavioral Reviews
(2016) - et al.
Hearing speech sounds: Top-down influences on the interface between audition and speech perception
Hearing Research
(2007) - et al.
Effects of aging on the human motor cortical plasticity studied by paired associative stimulation
Clinical Neurophysiology
(2010) - et al.
Speech recognition in noise and presbycusis: Relations to possible neural mechanisms
Hearing Research
(1997) - et al.
Intermittent theta-burst stimulation induces correlated changes in cortical and corticospinal excitability in healthy older subjects
Clinical Neurophysiology
(2017) - et al.
A mediating role of the auditory dorsal pathway in selective adaptation to speech: A state-dependent transcranial magnetic stimulation study
Brain Research
(2013)