Verbal fluency – i.e. the ability to name as many words of a given category as possible in a defined time interval – is an integral part of neuropsychological test batteries for the diagnosis of dementia. Verbal fluency can be easily administered and thus may also be implemented in computerized dementia screening tests.

In the present study we sought to investigate the capability of phonemic verbal fluency as a potential predictor of cognitive decline and its associations with acoustic and lexical features directly inferred from recordings of spoken language. Data from 246 ILSE participants born in the years 1930 to 1932 were included. Subjects were investigated three times (t1-t3) over a period of more than 10 years by two board certified geriatric psychiatrists: the mean age of the sample was 62.77 at t1, 66.36 at t2, and 74.22 at t3. To evaluate the sensitivity of phonemic fluency performance on cognitive impairment, scores obtained at t1 and t2 were compared with repeated measures ANOVA between subjects who were cognitively intact and those who were diagnosed with mild cognitive impairment (MCI) or Alzheimer's dementia (AD) at t3. In addition, acoustic and lexical features of spoken language were extracted with automatic speech recognition from semi-structured autobiographical interviews recorded at baseline. Scores were correlated with verbal fluency performance at t1. In all analyzes education was considered as a covariate. Regarding verbal fluency obtained at t1 and t2, repeated measures ANOVA revealed significantly higher scores between subjects who were cognitively intact (HC) at t3 compared with those diagnosed with MCI or AD at t3. At baseline verbal fluency was significantly correlated with important acoustic features of spontaneous speech: word rate, phoneme rate, speech-duration and silence-to-word-ratio.

Verbal fluency deficits were proven already in young old age and precede marked cognitive decline years before in a large group recruited from the general population. Verbal fluency is associated with acoustic rather than lexical features of spoken language which can directly be inferred from spontaneous speech.

口语流利度（即在定义的时间间隔内尽可能多地命名给定类别的单词的能力）是神经心理测试电池中不可或缺的诊断痴呆症的组成部分。口语流利性易于管理，因此也可以在计算机痴呆筛查测试中实现。

在本研究中，我们试图研究语音口语流利性作为认知能力下降的潜在预测因子的能力，以及其与直接从口语录音中推断出的声学和词汇特征的关联。包括来自1930年至1932年的246名ILSE参与者的数据。由两名董事会认证的老年精神病医生在超过10年的时间里对受试者进行了三次调查（t1-t3）：样本的平均年龄在t1时为62.77，在t2时为66.36，在t3时为74.22。为了评估音素流利能力对认知障碍的敏感性，将在t1和t2时获得的得分与在t3时认知完整的受试者和被诊断为轻度认知障碍（MCI）或阿尔茨海默氏痴呆（AD）的受试者之间的重复测量ANOVA进行比较。此外，从基线记录的半结构自传访谈中，通过自动语音识别提取口语的声学和词汇特征。分数与t1时的语言流利度相关。在所有分析中，教育都被视为协变量。关于在t1和t2时获得的口语流利度，重复测量ANOVA显示，与在t3时被诊断为MCI或AD的受试者相比，在t3时具有认知完好性（HC）的受试者之间的得分明显更高。在基线时，口语流利度与自发语音的重要声学特征显着相关：字速率，音素速率，语音持续时间和静默比。分数与t1时的语言流利度相关。在所有分析中，教育都被视为协变量。关于在t1和t2时获得的口语流利度，重复测量ANOVA显示，与在t3时被诊断为MCI或AD的受试者相比，在t3时具有认知完好性（HC）的受试者之间的得分明显更高。在基线时，口语流利度与自发语音的重要声学特征显着相关：字速率，音素速率，语音持续时间和静默比。分数与t1时的语言流利度相关。在所有分析中，教育都被视为协变量。关于在t1和t2时获得的口语流利度，重复测量ANOVA显示，与在t3时被诊断为MCI或AD的受试者相比，在t3时具有认知完好性（HC）的受试者之间的得分明显更高。在基线时，口语流利度与自发语音的重要声学特征显着相关：字速率，音素速率，语音持续时间和静默比。

在从老年人口招募的一大批人中，已经在年轻人中证明了语言流利性不足，并在明显的认知能力下降之前。口语流利度与口语的听觉而非词汇特征相关，而语音特征可以直接从自发语音中推断出来。