|
|
ENQUIRE PROJECT DETAILS BY GENERAL PUBLIC |
| Project Details |
| Funding Scheme : | General Research Fund | ||||||||||||||||||||||||||||||||||||||||||||||||
| Project Number : | 15601718 | ||||||||||||||||||||||||||||||||||||||||||||||||
| Project Title(English) : | Functional Connectome of Semantic Processing across the Adult Lifespan: A Multi-modal Investigation using EEG and fMRI | ||||||||||||||||||||||||||||||||||||||||||||||||
| Project Title(Chinese) : | 成年人語義處理的功能連結圖譜：腦電圖及功能性磁振造影的跨年齡層研究 | ||||||||||||||||||||||||||||||||||||||||||||||||
| Principal Investigator(English) : | Prof Wang, William Shi-yuan | ||||||||||||||||||||||||||||||||||||||||||||||||
| Principal Investigator(Chinese) : | |||||||||||||||||||||||||||||||||||||||||||||||||
| Department : | Department of Chinese and Bilingual Studies | ||||||||||||||||||||||||||||||||||||||||||||||||
| Institution : | The Hong Kong Polytechnic University | ||||||||||||||||||||||||||||||||||||||||||||||||
| E-mail Address : | wsywang@polyu.edu.hk | ||||||||||||||||||||||||||||||||||||||||||||||||
| Tel : | 34008557 | ||||||||||||||||||||||||||||||||||||||||||||||||
| Co - Investigator(s) : |
|
||||||||||||||||||||||||||||||||||||||||||||||||
| Panel : | Humanities, Social Sciences | ||||||||||||||||||||||||||||||||||||||||||||||||
| Subject Area : | Psychology and Linguistics | ||||||||||||||||||||||||||||||||||||||||||||||||
| Exercise Year : | 2018 / 19 | ||||||||||||||||||||||||||||||||||||||||||||||||
| Fund Approved : | 813,520 | ||||||||||||||||||||||||||||||||||||||||||||||||
| Project Status : | Completed | ||||||||||||||||||||||||||||||||||||||||||||||||
| Completion Date : | 31-12-2022 | ||||||||||||||||||||||||||||||||||||||||||||||||
| Project Objectives : |
|
||||||||||||||||||||||||||||||||||||||||||||||||
| Abstract as per original application (English/Chinese): |
Semantic access (i.e., retrieval of meanings) and semantic integration (i.e., formation of complex meanings) are traditionally regarded as intact in normal aging. Recent functional magnetic resonance imaging (fMRI) studies have begun to challenge this notion. For example, compared with young subjects, older subjects tend to rely more on the prefrontal cortex in many semantic tasks, including speech comprehension. Resting-state fMRI studies have also revealed robust age-related changes in functional connectivity (FC), most notably an inverted U-shaped trajectory for long-range connections within the frontotemporal and frontoparietal networks (FTN/FPN), which subserve high-order language and executive functions. Intriguingly, the peak of this developmental trajectory appears to coincide with the onset of behavioral decline in semantic ability. However, how these large-scale changes in the brain’s “functional connectome” relate to semantic processing has yet to be elucidated.
This project aims to extend our knowledge regarding the age-related neural reorganization that underlies the changes in semantic ability across the adult lifespan, by clarifying the behavioral and neural consequences of the changing resting-state functional connectivities (RSFCs) in the left FTN. We aim to examine two main hypotheses: (1) the continued growth in semantic ability in middle age is accompanied by increases in RSFCs in the left FTN; (2) the decline in semantic ability during senescence is associated with weakened RSFCs in the same network. Because fMRI is only sensitive to slow-varying (< 0.1 Hz) blood oxygen-level dependent (BOLD) signals, but language processing depends on neurocomputations that operate at much faster timescales, electroencephalography (EEG) will be jointly employed in order to develop a more comprehensive understanding on the developmental trajectory of semantic processing during adulthood.
Four cohorts of cognitively normal Cantonese speakers aged 20–34, 35–49, 50–64, and 65–79, matched for handedness and gender, will be recruited. These subjects will attend resting-state experiments and perform two tasks (semantic priming and sentence comprehension), with which the possible link between functional connectivities/activations of key semantic loci and both behavioral indices of semantic ability and EEG neuromarkers will be established.
Collectively, our findings will provide an extensive account of how semantic processing changes, both neuroanatomically and spectrotemporally, across the adult lifespan. The data from the 240+ subjects will be used to launch an open-access EEG/fMRI database to advance neurophysiological research on cognitive aging and semantic processing. Our results will find application in clinical assessment and monitoring for neurological disorders in which semantic processes are strongly affected, including semantic dementia and aphasia. 傳統上認為語義擷取（即擷取詞語的意思） 和語義整合（即整合複雜的意思） 在正常老化中是不會衰退的。然而，近期的功能性磁共振成像 (fMRI) 研究開始挑戰這一觀點。 與年輕人相比，年齡較大的人在理解語言時會傾向使用更多前額皮質 (prefrontal cortex)。靜息態 fMRI 研究亦發現了功能連結 (functional connectivity) 會呈現與年齡相關的強烈變化。其中，用來處理高階語言和決定執行功能的額顳網絡 (frontotemporal network, FTN) 和額頂網絡 (frontoparietal network, FPN) 的遠程連結與年齡呈現倒 U 形趨勢。有趣的是，遠程連結的發展峰值似乎與語義能力開始下降的時間一致。 不過，這些大腦功能連結圖譜中的大規模變化是否與語義處理能力的轉變相關尚未得到闡明。 此項目旨在研究左額顳網絡靜息態功能連結 (resting state functional connectivity, RSFC) 的改變及其對行為和神經活動的影響，以理解語義能力在成人生命週期內的變化及擴展對年齡相關神經重組的知識。此項目有兩個重要假說。一方面，我們推想語義能力在中年時期持續增長的同時，左額顳網絡的 RSFC 會隨之而增長。另一方面，我們也推想語義能力在衰老過程中的下降與左額顳網絡中 RSFC 弱化是有相關性的。 由於 fMRI 僅對緩慢變化的 (<0.1 Hz) 血氧濃度相依對比 (BOLD) 信號敏感，但語言處理卻依賴更快速的神經計算，所以本項目將聯合使用腦電圖 (EEG)，以全面了解成年期語義處理的發展軌跡。 本項目將招募四組認知能力正常、並以廣東話為母語的受試者。受試者年齡分別為 20-34 歲，35-49 歲，50-64 歲和 65-79 歲，組別之間慣用手和性別將會相互匹配。 通過靜息態實驗及兩項與語義有關的任務（語義促發和句子理解），本項目將針對某些處理語義有關的腦區，以釐清其功能連結及激活與語義能力及腦電指標間的聯繫。 總括而言，研究將全面地探索語義處理在成人生命週期中的變化，尤其是在腦神經定位及時頻域分析方面提供新知識。來自240多名受試者的數據將用於建立一個開放式EEG / fMRI數據庫，以推進認知衰老和語義處理的神經生理學研究。預料研究結果可應用於一些神經疾病的臨床評估和監測，尤其是那些語義處理會受到嚴重影響的病症，如語意型腦退化症 (semantic dementia) 和失語症 (aphasia)。 |
||||||||||||||||||||||||||||||||||||||||||||||||
| Realisation of objectives: | To achieve the four (revised) project objectives, over 200 native speakers of Cantonese were tested on a range of semantic tasks (semantic fluency, semantic judgment, semantic priming, and sentence comprehension) in two rounds of data collection. The first round included a smaller-scale behavioral and EEG study comprising only younger and older adults, focusing on establishing the paradigms and obtaining initial evaluation of the hypotheses. These results were published in 4 journal publications (Fong et al., 2020; Fong et al., 2021; Hui et al., 2020; Ma et al., 2021). The second round focused on formally testing the hypotheses through a sample comprising 100+ adults uniformly sampled across the age range of 18-81. The level of education was matched in four age-groups: 18-34; 35-49; 50-64; and 65-81. For Objective 1 (To investigate the behavioral and neurophysiological (fMRI and EEG) implications of the age-related changes in resting-state functional connectivities within the left frontotemporal language network), a comprehensive cognitive and semantic task battery was compiled and was administered to the participants in both rounds of data collection. During the first round of data collection, cognitive associations of semantic performances were established, including the associations of semantic fluency with lexical retrieval speed in younger adults (Fong et al., 2020), inhibition function with an L2 synonym test (Hui et al., 2020), and the age-related decline in controlled retrieval in older adults (Fong et al., 2021). Resting-state EEG measures were also correlated with cognitive task performances (Ma, Fong, Lee, & Wang, 2022). In the second round of data collection, a Cantonese psycholinguistic database was compiled, and the database was central to the design two tasks (semantic priming and sentence comprehension), since it allows for the control of various psycholinguistic factors that modulate the behavioral performance. For semantic priming, EEG and fMRI versions were administered, while for sentence comprehension, only the fMRI experiment was conducted. For the MRI dataset, structural analysis has been conducted with FreeSurfer (Fong et al., 2023), while functional activation and connectivity analysis have been conducted with SPM and CONN toolboxes. However, because the data collection was delayed due to COVID-19, the overall dataset is still under active analysis by the project team. The project team will continue to publish the results towards completely achieving this objective. For Objective 2 (To test whether semantic access and semantic integration are maximally efficient during middle age but not during early adulthood), two auditory tasks were designed—semantic priming and sentence comprehension. Behaviorally, we found no evidence that semantic access is maximally efficient during middle age. Instead, a hyper-priming effect was found, i.e., the semantic priming effect (SPE) was the largest for older adults. This increase in priming effect was not associated with the decline in inhibition function; nor could it be attributed to decreasing processing speed. Instead, we found evidence that the SPE was associated with semantic knowledge (as measured using a verbal fluency task). This finding suggests that older adults may have improved automatic semantic access due either to their growth in semantic knowledge or accumulated exposures to the semantic relations. In contrast, controlled retrieval appeared to peak during early adulthood. For Objective 3 (To test whether abstract and concrete words have dissociable developmental trajectories due to normal aging), behaviorally, an interaction between Concreteness and Age was found by ANOVA, i.e., we found that the concreteness effect based on RT was positively related with age. A corresponding effect was observed with ERP data. These findings support the hypothesis that abstract and concrete words have dissociable developmental trajectories. Regarding Objective 4 (To launch an open-access, web-based EEG/fMRI database that includes both task-related and resting-state data for the study of cognitive aging), we originally aimed to release it through a webpage administered by PolyU. However, in view of data security, it was decided that the data would be uploaded to two platforms—ADRES managed by the Research Institute for Smart Ageing, Hong Kong Polytechnic University, as well as OpenNeuro, a widely used platform in the public domain for sharing neuroimaging data. Using these web services will enhance the impact of the present project. In total, fifteen conference papers / abstracts were presented in international conferences (including International Conference on Acoustics, Speech, and Signal Processing 2020, Cognitive Neuroscience Society 2019/2020, International Symposium on Chinese Spoken Language Processing, The 34th Pacific Asia Conference on Language, Information and Computation, The 34th International Congress of Psychology, and The 13th International Symposium on Bilingualism) and a local conference (PolyU Academy for Interdisciplinary Research Conference 2023). In addition, the PI and Dr. Manson Fong (a key member in this project) have also presented some of the findings in the present project in RGC Public Lecture. In terms of education impact, four different undergraduate student assistants have been trained, and two final-year undergraduate projects arose directly from the present project. | ||||||||||||||||||||||||||||||||||||||||||||||||
| Summary of objectives addressed: |
|
||||||||||||||||||||||||||||||||||||||||||||||||
| Research Outcome | |||||||||||||||||||||||||||||||||||||||||||||||||
| Major findings and research outcome: | The first round of data collection focused on how cognitive differences contribute to individual differences in semantic abilities. In younger adults, semantic fluency was influenced by lexical retrieval speed, but inhibitory control functions appeared to play little role (Fong et al., 2020). For older adults, the decline in inhibitory control function was posited to contribute to their declining cognitive performance. To test this hypothesis within the domain of lexico-semantic retrieval, event-related potential (ERP) was adopted to measure several inhibition-related functions. Mixed-effects modelling on semantic fluency revealed that the number of concepts retrieved in a controlled manner, but not in an automatic manner, was negatively associated with age (Fong et al., 2021). This interaction was partially accounted for by the stimulus-locked P2 and response-locked correct positivity (Pc), lending support to the hypothesis. Other relevant results were as follows. The Stroop test performance was predicted by an English synonym test, demonstrating the link between inhibition and language (Hui et al., 2020). The loss of complexity in ageing hypothesis (LOCH), which postulates that ageing is accompanied by the loss of complexity, was supported by EEG data (Ma et al., 2021). In the second round of data collection, cognitively normal participants aged 18–81 were recruited. Older adults were found to exhibit hyper-priming, i.e., the auditory semantic priming effect (SPE) was positively associated with age. This is consistent with the hypothesis that the automatic semantic processing of older adults improves with age. Also, the concreteness effect was found to increase with age (Fong et al., under preparation). Across the adult lifespan, an inverted U-shaped pattern in the complexity trajectory was found (Ma, Fong, & Wang, 2023; Ma, Fong, & Wang, under preparation). A Cantonese psycholinguistic database for over 7000 words was constructed and will be released soon (Fong, Ng, & Wang, under preparation). In total, six journal articles were published based on the data collected in the present project, along with 15 conference abstracts/presentations. Three manuscripts are being prepared. Because of the pandemic and that an MRI machine was available for use only midway through, data collection in the second round was much delayed; the research findings will be disseminated via journal publications. A postdoctoral fellow trained in this project, Dr. Manson Fong, was promoted to Research Assistant Professor. Four PhD students were directly trained in the present project, including Dr. Matthew Ma, who was awarded the RGC postdoctoral fellowship. | ||||||||||||||||||||||||||||||||||||||||||||||||
| Potential for further development of the research and the proposed course of action: |
The present research will be extended along two tracks. First, having collected detailed semantic data for a representative cohort of 100+ adults aged 18-81, the team is in a primed position to study the differential semantic impairments in pathological populations (e.g., mild cognitive impairment, dementia). The project team will seek collaboration with medical colleagues to document the semantic, cognitive, and neuroimaging profile for different sub-types of dementia (e.g., Alzheimer’s disease and semantic dementia). The results will contribute towards the early detection, diagnosis, and prognosis of dementia. Second, the brain’s cognitive reserve has been linked with bilingualism or bilingual experience. The present research can be extended in a bilingual setting (e.g., Cantonese–English) to obtain measures of bilingual semantic performance, for providing converging evidence on how bilingualism may benefit brain health and contributes towards the protection against dementia. | ||||||||||||||||||||||||||||||||||||||||||||||||
| Layman's Summary of Completion Report: | Normal ageing is typically accompanied by a growth in semantic knowledge and relatively intact semantic functions (automatic access and controlled retrieval). The present project aims to find out when semantic functions become maximally efficient, and the underlying age-related neural changes. Over 100 participants aged 18-81 were recruited to the study, with each participant completing some semantic tasks while their brain activities are measured using electroencephalography (EEG) and functional magnetic resonance imaging (fMRI). Surprisingly, we found that in the presence of a categorically or associatively related prime word, older adults are more efficient than their younger counterparts in extracting semantic relations of a target word. This phenomenon is known as hyper-priming, and it reflects the more efficient automatic access of semantic knowledge by older adults. In contrast, controlled retrieval exhibited a small but significant decline. We also found that the age-related facilitation in semantic access was more robust for concrete than for abstract words. Overall, the present study provides an initial norm for understanding the semantic development across the adult lifespan. The results will be applied to studying the semantic impairment in various forms of neurodegenerative diseases, including Alzheimer’s disease and semantic dementia. | ||||||||||||||||||||||||||||||||||||||||||||||||
| Research Output | |||||||||||||||||||||||||||||||||||||||||||||||||
| Peer-reviewed journal publication(s) arising directly from this research project : (* denotes the corresponding author) |
|
||||||||||||||||||||||||||||||||||||||||||||||||
| Recognized international conference(s) in which paper(s) related to this research project was/were delivered : |
|
||||||||||||||||||||||||||||||||||||||||||||||||
| Other impact (e.g. award of patents or prizes, collaboration with other research institutions, technology transfer, etc.): |
|||||||||||||||||||||||||||||||||||||||||||||||||
| SCREEN ID: SCRRM00542 |