Alzheimer's disease (AD) is characterized by progressive cognitive decline, including impairments in speech production and fluency. Mild cognitive impairment (MCI), a prodrome of AD, has also been linked with changes in speech behavior but to a more subtle degree.
Objective
This study aimed to investigate whether speech behavior immediately following both filled and unfilled pauses (post-pause speech behavior) differs between individuals with MCI and healthy controls (HCs), and how these differences are influenced by the cognitive demands of various speech tasks.
Methods
Transcribed speech samples were analyzed from both groups across different tasks, including immediate and delayed narrative recall, picture descriptions, and free responses. Key metrics including lexical and syntactic complexity, lexical frequency and diversity, and part of speech usage, both overall and post-pause, were examined.
Results
Significant differences in pause usage were observed between groups, with a higher incidence and longer latencies following these pauses in the MCI group. Lexical frequency following filled pauses was higher among MCI participants in the free response task but not in other tasks, potentially due to the relative cognitive load of the tasks. The immediate recall task was most useful at differentiating between groups. Predictive analyses utilizing random forest classifiers demonstrated high specificity in using speech behavior metrics to differentiate between MCI and HCs.
Conclusions
Speech behavior following pauses differs between MCI participants and healthy controls, with these differences being influenced by the cognitive demands of the speech tasks. These post-pause speech metrics can be easily integrated into existing speech analysis paradigms.
Forbes-McKayKEVenneriA. Detecting subtle spontaneous language decline in early Alzheimer’s disease with a picture description task. Neurol Sci2005; 26: 243–254.
2.
AhmedSHaighA-MFde JagerCA, et al.Connected speech as a marker of disease progression in autopsy-proven Alzheimer’s disease. Brain2013; 136: 3727–3737.
3.
KavéGGoralM. Word retrieval in picture descriptions produced by individuals with Alzheimer’s disease. J Clin Exp Neuropsychol2016; 38: 958–966.
4.
FraserKCMeltzerJARudziczF. Linguistic features identify Alzheimer’s disease in narrative speech. J Alzheimers Dis2015; 49: 407–422.
5.
YamadaYShinkawaKNemotoM, et al.Speech and language characteristics differentiate Alzheimer’s disease and dementia with Lewy bodies. Alzheimers Dement (Amst)2022; 14: e12364.
6.
YuanJCaiXBianY, et al.Pauses for detection of Alzheimer’s disease. Front Comput Sci2021; 2: 624488.
7.
BoschiVCatricalàEConsonniM, et al.Connected speech in neurodegenerative language disorders: a review. Front Psychol2017; 8: 269.
ThemistocleousCEckerströmMKokkinakisD. Voice quality and speech fluency distinguish individuals with mild cognitive impairment from healthy controls. PLoS One2020; 15: e0236009.
10.
MuellerKDKoscikRLHermannBP, et al.Declines in connected language are associated with very early mild cognitive impairment: results from the Wisconsin Registry for Alzheimer’s Prevention. Front Aging Neurosci2018; 9: 437.
11.
GoodglassHKaplanEWeintraubS. BDAE: The Boston diagnostic aphasia examination. Philadelphia: Lippincott Williams & Wilkins, 2001.
12.
WechslerD. Wechsler memory scale-revised. New York: The Psychological Corporation, 1987.
13.
CraftSNewcomerJKanneS, et al.Memory improvement following induced hyperinsulinemia in Alzheimer’s disease. Neurobiol Aging1996; 17: 123–130.
14.
BeltramiDGagliardiGRossini FavrettiR, et al.Speech analysis by natural language processing techniques: a possible tool for very early detection of cognitive decline?Front Aging Neurosci2018; 10: 369.
15.
RoarkBMitchellMHollingsheadK. Syntactic complexity measures for detecting mild cognitive impairment. In: Proceedings of the workshop on BioNLP 2007 biological, translational, and clinical language processing - BioNLP ‘07. Prague, Czech Republic: Association for Computational Linguistics, 2007, p.1.
16.
ClarkeNBarrickTRGarrardP. A comparison of connected speech tasks for detecting early Alzheimer’s disease and mild cognitive impairment using natural language processing and machine learning. Front Comput Sci2021; 3: 634360.
17.
TalerVKousaieSSheppardC. Lexical access in mild cognitive impairment: the effects of context. Mental Lexicon2015; 10: 271–285.
18.
FraserKCLundholm ForsKEckerströmM, et al.Predicting MCI status from multimodal language data using cascaded classifiers. Front Aging Neurosci2019; 11: 205.
19.
FraserKCMeltzerJAGrahamNL, et al.Automated classification of primary progressive aphasia subtypes from narrative speech transcripts. Cortex2014; 55: 43–60.
20.
MeteyardLQuainEPattersonK. Ever decreasing circles: speech production in semantic dementia. Cortex2014; 55: 17–29.
21.
Goldman-EislerF. Psycholinguistics: Experiments in spontaneous speech. London & New York: Academic Press, 1968.
22.
GildenDMezaraupsT. Laws for pauses. J Exp Psychol Learn Mem Cogn2022; 48: 142–158.
23.
WomackKMcCoyWOvesdotter AlmC, et al.Disfluencies as extra-propositional indicators of cognitive processing. In: MoranteRSporlederC (eds) Proceedings of the workshop on extra-propositional aspects of meaning in computational linguistics. Jeju, Republic of Korea: Association for Computational Linguistics, 2012, pp.1–9.
24.
ClarkHHFox TreeJE. Using uh and um in spontaneous speaking. Cognition2002; 84: 73–111.
25.
ShribergE. To ‘errrr’ is human: ecology and acoustics of speech disfluencies. J Int Phon Assoc2001; 31: 153–169.
26.
TothLHoffmannIGosztolyaG, et al.A speech recognition-based solution for the automatic detection of mild cognitive impairment from spontaneous speech. Curr Alzheimer Res2017; 14: 130–138.
27.
VinczeVSzatlóczkiGTóthL, et al.Telltale silence: temporal speech parameters discriminate between prodromal dementia and mild Alzheimer’s disease. Clin Linguist Phon2021; 35: 727–742.
28.
Egas-LópezJVBaloghRImreN, et al.Automatic screening of mild cognitive impairment and Alzheimer’s disease by means of posterior-thresholding hesitation representation. Comput Speech Langu2022; 75: 101377.
29.
WangTHongYWangQ, et al.Identification of mild cognitive impairment among Chinese based on multiple spoken tasks. J Alzheimers Dis2021; 82: 185–204.
30.
BesserLMChrisphonteSKleimanMJ, et al.The healthy brain initiative (HBI): a prospective cohort study protocol. PLoS One2023; 18: e0293634.
31.
MorrisJC. The clinical dementia rating (CDR): current version and scoring rules. Neurology1993; 43: 2412–2412.
32.
WeintraubSBesserLDodgeHH, et al.Version 3 of the Alzheimer disease Centers’ neuropsychological test battery in the uniform data set (UDS). Alzheimer Dis Assoc Disord2018; 32: 10–17.
33.
BolognaniSAPMirandaMCMartinsM, et al.Development of alternative versions of the logical memory subtest of the WMS-R for use in Brazil. Dement Neuropsychol2015; 9: 136–148.
34.
BerubeSNonnemacherJDemskyC, et al.Stealing cookies in the twenty-first century: measures of spoken narrative in healthy versus speakers with aphasia. Am J Speech Lang Pathol2019; 28: 321–329.
35.
HeuerS. The influence of image characteristics on image recognition: a comparison of photographs and line drawings. Aphasiology2016; 30: 943–961.
36.
HuxKFrodshamK. Speech and language characteristics of neurologically healthy adults when describing the Modern Cookie Theft Picture: mixing the new with the old. Am J Speech Lang Pathol2023; 32: 1110–1130.
BittnerRHumphreyEJBelloJ. PySOX: Leveraging the audio signal processing power of SOX in Python. New York: Proceedings of the 17th International Society for Music Information Retrieval Conference Late Breaking and Demo Papers, 2016.
40.
HonnibalMMontaniIVan LandeghemS, et al.spaCy: Industrial-strength natural language processing in Python. Zenodo2020. doi: https://doi.org/10.5281/zenodo.1212303
HansenLOlsenLREnevoldsenK. Textdescriptives: a Python package for calculating alarge variety of metrics from text. J Open Source Softw2023; 8: 5153.
43.
ColemanMLiauTL. A computer readability formula designed for machine scoring. J Appl Psychol1975; 60: 283–284.
44.
BrysbaertMNewB. Moving beyond Kucera and Francis: a critical evaluation of current word frequency norms and the introduction of a new and improved word frequency measure for American English. Behav Res Methods2009; 41: 977–990.
45.
YngveVH. A model and an hypothesis for language structure. Proc Am Philos Soc1960; 104: 444–466.
46.
CovingtonMAMcFallJD. Cutting the gordian knot: the moving-average type–token ratio (MATTR). J Quant Linguist2010; 17: 94–100.
47.
McCarthyPMJarvisS. MTLD, vocd-D, and HD-D: a validation study of sophisticated approaches to lexical diversity assessment. Behav Res Methods2010; 42: 381–392.
48.
BortfeldHLeonSDBloomJE, et al.Disfluency rates in conversation: effects of age, relationship, topic, role, and gender. Lang Speech2001; 44: 123–147.
49.
VallatR. Pingouin: statistics in Python. J Open Source Softw2018; 3: 1026.
50.
NasreddineZSPhillipsNABédirianV, et al.The Montreal Cognitive Assessment, MoCA: a brief screening tool for mild cognitive impairment. J Am Geriatr Soc2005; 53: 695–699.
51.
Cahn-HidalgoDBenabouRKewinS. Validity, reliability, and psychometric properties of Cognivue®, a quantitative assessment of cognitive impairment. Am J Geriatr Psychiatry2019; 27: S212.
52.
BrandtJ. The Hopkins Verbal Learning Test: development of a new memory test with six equivalent forms. Clin Neuropsychol1991; 5: 125–142.
53.
GalvinJEToleaMIMooreC, et al.The Number Symbol Coding Task: a brief measure of executive function to detect dementia and cognitive impairment. PLoS One2020; 15: e0242233.
54.
KleimanMJGalvinJE. The Vulnerability Index: a weighted measure of dementia and cognitive impairment risk. Alzheimers Dement (Amst)2021; 13: e12249.
55.
GalvinJEKleimanMJChrisphonteS, et al.The Resilience Index: a quantifiable measure of brain health and risk of cognitive impairment and dementia. J Alzheimers Dis2021; 84: 1729–1746.
56.
KeGMengQFinleyT, et al. LightGBM: A highly efficient gradient boosting decision tree. In: Proceedings of the 31st international conference on neural information processing systems. Red Hook, NY, USA: Curran Associates, Inc., 2017, pp.3149–3157.
57.
PedregosaFVaroquauxGGramfortA, et al.Scikit-learn: machine learning in Python. J Mach Learn Res2011; 12: 2825–2830.
58.
KeanyE. BorutaShap: A wrapper feature selection method which combines the Boruta feature selection algorithm with Shapley values. Zenodo2020. doi: https://doi.org/10.5281/zenodo.4247618.
59.
AkibaTSanoSYanaseT, et al.Optuna: a next-generation hyperparameter optimization framework. In: Proceedings of the 25th ACM SIGKDD international conference on knowledge discovery & data mining. New York: Association for Computing Machinery, 2019, pp.2623–2631.
MüllerCGroßmann-HutterBJamesonA, et al.Recognizing time pressure and cognitive load on the basis of speech: an experimental study. In: BauerMGmytrasiewiczPJVassilevaJ (eds) User modeling 2001. Berlin, Heidelberg: Springer, 2001, pp.24–33.
62.
SwertsM. Filled pauses as markers of discourse structure. J Pragmat1998; 30: 485–496.
63.
DegandLGilquinGMeurantL, et al.Fluency and disfluency across languages and language varieties. Louvain-la-Neuve: Presses universitaires de Louvain, 2019.
64.
EarlesJLKerstenAWBerlin MasB, et al.Aging and memory for self-performed tasks: effects of task difficulty and time pressure. J Gerontol B Psychol Sci Soc Sci2004; 59: P285–P293.
65.
LibbyLAHannulaDERanganathC. Medial temporal lobe coding of item and spatial information during relational binding in working memory. J Neurosci2014; 34: 14233–14242.
66.
KühnSGallinatJ. Segregating cognitive functions within hippocampal formation: a quantitative meta-analysis on spatial navigation and episodic memory. Hum Brain Mapp2014; 35: 1129–1142.
67.
RussoMBendazzoliCDefrancqB. Making way in corpus-based interpreting studies. Singapore: Springer, 2017.
68.
KleimanMJBarenholtzEGalvinJE. Screening for early-stage Alzheimer’s disease using optimized feature sets and machine learning. J Alzheimers Dis2021; 81: 355–366.
69.
YuanJBianYCaiX, et al.Disfluencies and fine-tuning pre-trained language models for detection of Alzheimer’s disease. Proc Interspeech2020: 2162–2166.
70.
MuellerKDKoscikRLDuL, et al.Proper names from story recall are associated with beta-amyloid in cognitively unimpaired adults at risk for Alzheimer’s disease. Cortex2020; 131: 137–150.
71.
RoarkBMitchellMHosomJ-P, et al.Spoken language derived measures for detecting mild cognitive impairment. IEEE Trans Audio Speech Lang Process2011; 19: 2081–2090.
72.
MuellerKDHermannBMecollariJ, et al.Connected speech and language in mild cognitive impairment and Alzheimer’s disease: a review of picture description tasks. J Clin Exp Neuropsychol2018; 40: 917–939.
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
