Linguistic markers in Parkinson’s disease (Smith et al., 2018)

<b>Purpose:</b> Early cognitive symptoms such as word-finding difficulty (WFD) in daily conversation are common in Parkinson’s disease (PD), but studies have been limited by a lack of feasible, quantitative measures. Linguistic analysis, focused on pauses in speech, may yield markers of impairment of cognition and communication in PD. The objective of this study was to evaluate the relationship of linguistic markers in semistructured speech to WFD symptoms and cognitive function in PD.<b>Method: </b>Speech recordings of description of the Cookie Theft picture in 53 patients with PD without dementia and 23 elderly controls were analyzed with Praat software. Montreal Cognitive Assessment (MoCA; Nasreddine et al., 2005), category naming fluency, and confrontation naming tests were administered. Questionnaires rating WFD symptoms and cognitive instrumental activities of daily living were completed. We determined the relationships between (a) pause length and location, (b) MoCA score, and (c) WFD symptoms, using Pearson’s correlations and multivariate regression models.<b>Results:</b> Compared with controls, patients with PD had more pauses within utterances as well as fewer words per minute and a lower percentage of well-formed sentences. Pauses within utterances differed significantly between PD–mild cognitive impairment and normal cognition (<i>p</i> &lt; .001). Words per minute and percentage of well-formed sentences were predictive of MoCA in multivariate regression models. Pauses before verbs were associated with patient-reported severity of WFD symptoms (<i>p</i> = .006).<b>Conclusions:</b> Linguistic markers including pauses within utterances distinguish patients with PD with mild cognitive symptoms from elderly controls. These markers are associated with global cognitive function before the onset of dementia. Pauses before verbs and grammatical markers may index early cognitive symptoms such as WFD that may interfere with functional communication.<br><b>Supplemental Material S1.</b> Baseline characteristics of Parkinson’s disease (PD) with mild cognitive impairment (MCI) and normal cognition (NC). <br><b>Supplemental Material S2.</b> Correlations between clinical, neuropsychological, and linguistic biomarkers and word-finding difficulty (WFD) symptoms.<br><b>Supplemental Material S3.</b> Transcription conventions.<br>Smith, K. M., Ash, S., Xie, S. X., &amp; Grossman, M. (2018). Evaluation of linguistic markers of word-finding difficulty and cognition in Parkinson’s disease. <i>Journal of Speech, Language, and Hearing Research, 61,</i> 1691–1699. https://doi.org/10.1044/2018_JSLHR-L-17-0304

**研究目的**：帕金森病（Parkinson’s Disease, PD）患者日常对话中常出现早期认知症状，如找词困难（Word-Finding Difficulty, WFD），但现有研究因缺乏可行的量化评估手段而受限。以言语停顿为核心的语言学分析，或可作为PD患者认知与沟通功能受损的标志物。本研究旨在探究半结构化言语中的语言学标志物与PD患者的找词困难症状及认知功能之间的关联。
**方法**：本研究对53例无痴呆的PD患者与23例老年对照者的“偷饼干图片”描述任务语音录音进行分析，分析工具为Praat软件。所有受试者均接受蒙特利尔认知评估量表（Montreal Cognitive Assessment, MoCA; Nasreddine等，2005）、范畴命名流畅性测验与对质命名测验，同时完成评估找词困难症状的问卷以及认知性工具性日常生活活动能力问卷。本研究采用皮尔逊相关分析与多元回归模型，分析（1）停顿的时长与位置、（2）MoCA评分、（3）找词困难症状三者之间的关联。
**结果**：与对照组相比，PD患者的语句内停顿频次更高，每分钟词汇产出量更低，且合语法语句占比更低。PD伴轻度认知损害（Mild Cognitive Impairment, MCI）亚组与认知正常（Normal Cognition, NC）亚组的语句内停顿存在显著差异（*p* < .001）。多元回归模型显示，每分钟词汇产出量与合语法语句占比可预测MoCA评分。动词前停顿与受试者自评的找词困难严重程度显著相关（*p* = .006）。
**结论**：包括语句内停顿在内的语言学标志物，可区分伴轻度认知症状的PD患者与老年对照人群。此类标志物与痴呆发病前的整体认知功能相关。动词前停顿与语法标志物或可作为早期认知症状（如找词困难）的评估指标，此类症状可能会损害功能性沟通能力。
**补充材料S1**：伴轻度认知损害（Mild Cognitive Impairment, MCI）与认知正常（Normal Cognition, NC）的PD患者基线特征。
**补充材料S2**：临床、神经心理学与语言学生物标志物与找词困难（Word-Finding Difficulty, WFD）症状间的相关性。
**补充材料S3**：语音转录规范。
Smith, K. M., Ash, S., Xie, S. X., & Grossman, M. (2018). 帕金森病患者找词困难与认知的语言学标志物评估. 《言语、语言与听力研究杂志》, 61, 1691–1699. https://doi.org/10.1044/2018_JSLHR-L-17-0304