-
Speech Communication Nov 2023To compare verbal fluency scores derived from manual transcriptions to those obtained using automatic speech recognition enhanced with machine learning classifiers.
OBJECTIVE
To compare verbal fluency scores derived from manual transcriptions to those obtained using automatic speech recognition enhanced with machine learning classifiers.
METHODS
Using Amazon Web Services, we automatically transcribed verbal fluency recordings from 1400 individuals who performed both animal and letter F verbal fluency tasks. We manually adjusted timings and contents of the automatic transcriptions to obtain "gold standard" transcriptions. To make automatic scoring possible, we trained machine learning classifiers to discern between valid and invalid utterances. We then calculated and compared verbal fluency scores from the manual and automatic transcriptions.
RESULTS
For both animal and letter fluency tasks, we achieved good separation of valid versus invalid utterances. Verbal fluency scores calculated based on automatic transcriptions showed high correlation with those calculated after manual correction.
CONCLUSION
Many techniques for scoring verbal fluency word lists require accurate transcriptions with word timings. We show that machine learning methods can be applied to improve off-the-shelf ASR for this purpose. These automatically derived scores may be satisfactory for some applications. Low correlations among some of the scores indicate the need for improvement in automatic speech recognition before a fully automatic approach can be reliably implemented.
PubMed: 38881790
DOI: 10.1016/j.specom.2023.102990 -
Scientific Reports Jun 2024Dementia is a progressive neurological disorder that affects the daily lives of older adults, impacting their verbal communication and cognitive function. Early...
Dementia is a progressive neurological disorder that affects the daily lives of older adults, impacting their verbal communication and cognitive function. Early diagnosis is important to enhance the lifespan and quality of life for affected individuals. Despite its importance, diagnosing dementia is a complex process. Automated machine learning solutions involving multiple types of data have the potential to improve the process of automated dementia screening. In this study, we build deep learning models to classify dementia cases from controls using the Pitt Cookie Theft dataset from DementiaBank, a database of short participant responses to the structured task of describing a picture of a cookie theft. We fine-tune Wav2vec and Word2vec baseline models to make binary predictions of dementia from audio recordings and text transcripts, respectively. We conduct experiments with four versions of the dataset: (1) the original data, (2) the data with short sentences removed, (3) text-based augmentation of the original data, and (4) text-based augmentation of the data with short sentences removed. Our results indicate that synonym-based text data augmentation generally enhances the performance of models that incorporate the text modality. Without data augmentation, models using the text modality achieve around 60% accuracy and 70% AUROC scores, and with data augmentation, the models achieve around 80% accuracy and 90% AUROC scores. We do not observe significant improvements in performance with the addition of audio or timestamp information into the model. We include a qualitative error analysis of the sentences that are misclassified under each study condition. This study provides preliminary insights into the effects of both text-based data augmentation and multimodal deep learning for automated dementia classification.
Topics: Humans; Deep Learning; Dementia; Aged; Female; Male; Aged, 80 and over; Databases, Factual
PubMed: 38880810
DOI: 10.1038/s41598-024-64438-1 -
NPJ Digital Medicine Jun 2024Exergaming is a combination of exercise and gaming. Evidence shows an association between exercise and cognition in older people. However, previous studies showed... (Review)
Review
Exergaming is a combination of exercise and gaming. Evidence shows an association between exercise and cognition in older people. However, previous studies showed inconsistent results on the cognitive benefits of exergaming in people with cognitive impairment. Therefore, this study aims to examine the effect of exergaming intervention on cognitive functions in people with MCI or dementia. A systematic literature search was conducted via OVID databases. Randomized controlled trials (RCTs) examined the effect of an exergaming intervention on cognitive functions in people with MCI or dementia were included. Subgroup analyses were conducted according to the type of intervention and training duration. Twenty RCTs with 1152 participants were identified, including 14 trials for MCI and 6 trials for dementia. In people with MCI, 13 studies used virtual-reality (VR)-based exergaming. Those who received VR-based exergaming showed significantly better global cognitive function [SMD (95%CI) = 0.67 (0.23-1.11)], learning and memory [immediate recall test: 0.79 (0.31-1.27); delayed recall test: 0.75 (0.20-1.31)], working memory [5.83 (2.27-9.39)], verbal fluency [0.58 (0.12-1.03)], and faster in executive function than the controls. For people with dementia, all studies used video-based exergaming intervention. Participants with exergaming intervention showed significantly better global cognitive function than the controls [0.38 (0.10-0.67)]. Subgroup analyses showed that longer training duration generated larger effects. The findings suggest that exergaming impacts cognitive functions in people with MCI and dementia. Cognitive benefits are demonstrated for those with a longer training duration. With technological advancement, VR-based exergaming attracts the attention of people with MCI and performs well in improving cognitive functions.
PubMed: 38879695
DOI: 10.1038/s41746-024-01142-4 -
Experimental Gerontology Jun 2024Given the established association between sarcopenia and cognitive impairment was mainly in the older and oldest-old population or people with relatively limited...
BACKGROUND
Given the established association between sarcopenia and cognitive impairment was mainly in the older and oldest-old population or people with relatively limited education, this study extends the investigation to community-dwelling middle-to-old age adults in urban communities, emphasizing the need for preventive intervention for muscle health and healthy longevity.
METHODS
Data of 712 participants from the Gan-Dau Healthy Longevity Plan were retrieved for analysis, and all participants were stratified by age (50-64, 65-74 and 75+ years old). Possible sarcopenia was defined by 2019 consensus report of the Asian Working Group for Sarcopenia (AWGS). This study used four neuropsychological tests for analysis, i.e., Mini-Mental Status Examination (MMSE), California Verbal Learning Test II (CVLT-SF), Digital Symbol Substitution Test (DSST) and Verbal fluency (VF) for global and domain-specific cognitive function. Multivariate generalized linear models (GLMs) were employed to investigate the associations between possible sarcopenia and cognitive function in each age-specific groups.
RESULTS
The prevalence of possible sarcopenia increased with age, with 31.8 %, 37.7 %, and 55.6 % in participants aged 5064, 65-74 and, 75+ years, respectively. On the other hand, cognitive performance declined with age. In particular, among participants aged 75+ years with possible sarcopenia, their cognitive performance were poorer than robust counterparts, including MMSE (26.6 [3.4] vs. 27.4 [2.6]), CVTL-SF (total score: 21.5 [5.4] vs. 23.8 [5.5]; 30-second delayed recall: 6.0 [1.7] vs. 6.5 [1.6]), DSST (32.8 [14.3] vs. 41.3 [18.7]), and VF (12.8 [5.1] vs. 14.8 [4.9]). Multivariate generalized linear model indicated that possible sarcopenia was associated with lower MMSE (β: -0.70, p = 0.014) and lower DSST (β: -7.00, p = 0.010) in those aged 50-64 years. Moreover, possible sarcopenia was associated with lower CVLT-SF (total score β:-1.90, p = 0.028), lower DSST (β: -6.45, p < 0.001), and lower VF (β: -1.64, p=0.026) in 75+ years group.
CONCLUSIONS
An association exists between possible sarcopenia and cognitive impairment, encompassing global cognition, delayed memory, verbal fluency, and executive function, among community-dwelling adults of mid-to-old age. Future research is warranted to explore the temporal alterations in this association and the potential effects of interventions aimed at fostering healthy longevity.
PubMed: 38879092
DOI: 10.1016/j.exger.2024.112487 -
Journal of Stroke and Cerebrovascular... Jun 2024Electroacupuncture (EA) could represent a clinically effective treatment strategy for patients with vascular cognitive impairment no dementia (VCIND). This randomized...
BACKGROUND
Electroacupuncture (EA) could represent a clinically effective treatment strategy for patients with vascular cognitive impairment no dementia (VCIND). This randomized trial aims to explore the underlying mechanism of EA in VCIND patients through cognitive function assessment and neuroimaging assessment.
METHODS
140 eligible patients with VCIND were recruited and randomly divided into EA group (n = 70) and Control group (n = 70). The Montreal Cognitive Assessment (MoCA), and the Auditory Verbal Learning Test (AVLT), the Stroop color-naming task (STROOP), and the resting-state functional magnetic resonance imaging assessment. The EA group received treatment for 30 min/day, 5 times/week, for 8 weeks.
RESULTS
EA intervention could increase the MoCA score and improve the neutral and consistency response of the STROOP test in VCIND patients (P < 0.05). fMRI functional connectivity analysis showed that, after EA, the default mode network (DMN) function of the posterior cingulate gyrus, left middle frontal gyrus, left anterior cingulate gyrus, left and right superior temporal gyrus, right insula, left precentral gyrus and other brain regions were significantly higher than that in the control group. The functional connectivity between the posterior cingulate gyrus-left middle frontal gyrus and the posterior cingulate gyrus-right superior temporal gyrus was positively correlated with cognitive function (P < 0.05). Gray Matter Volume increased in VCIND after EA(P < 0.05).
CONCLUSIONS
EA can increase the functional connectivity between posterior cingulate gyrus-other gyri in VCIND patients. The functional connectivity is positively correlated with cognitive function.
PubMed: 38878393
DOI: 10.1016/j.jstrokecerebrovasdis.2024.107788 -
Cortex; a Journal Devoted To the Study... May 2024Hebb repetition learning (HRL) refers to neurodevelopmental processes characterised by repeated stimulus exposure without feedback, which result in changes in behaviour...
Hebb repetition learning (HRL) refers to neurodevelopmental processes characterised by repeated stimulus exposure without feedback, which result in changes in behaviour and/or responses, e.g., long-term learning of serial order. Here, we investigate effects of HRL on serial order memory. The present research aimed to assess the reliability of new HRL measures and investigate their relationships with language and reading skills (vocabulary, grammar, word reading) in adolescents with intellectual disability (ID). A comparison group of children of similar mental age with typical development (TD) was also assessed. ID and TD groups were tested on HRL tasks, evaluating test-retest and split-half reliability. The relationship between HRL and language and reading was analysed after accounting for the influence of mental age and verbal short-term memory. The HRL tasks displayed moderate test-retest (and split-half) reliability, HRL tasks with different stimuli (verbal, visual) were related, and we identified issues with one method of HRL scoring. The planned regression analyses failed to show relationships between HRL and language/reading skills in both groups when mental age, a very strong predictor, was included. However, further exploratory regression analyses without mental age revealed HRL's predictive capabilities for vocabulary in the ID group and reading in the TD group, results which need further investigation and replication. HRL displays promise as a moderately reliable metric and exhibits varied and interpretable predictive capabilities for language and reading skills across groups.
PubMed: 38878338
DOI: 10.1016/j.cortex.2024.05.012 -
Progress in Brain Research 2024Verbal short-term and long-term memory are crucial neuropsychological functions involved in core cognitive abilities. They constitute vital components of subjective...
Verbal short-term and long-term memory are crucial neuropsychological functions involved in core cognitive abilities. They constitute vital components of subjective well-being and academic achievement. To date, there is limited research on the association between regular physical activity and memory abilities during young adulthood. The Individual Alpha Peak Frequency (IAPF) contributes to various cognitive abilities and also appears to be sensitive to physical activity. Consequently, the IAPF has the potential to underlie the association between physical activity and verbal memory. We examined the direct relation of physical activity and verbal memory, and the potential indirect relation via IAPF in young adults. Regular physical activity was assessed via accelerometry on seven consecutive days in 115 participants (N=115, 48% female) aged 18-35 years (M=24.1, SD=3.8). In addition, verbal memory performance was assessed using an immediate and delayed free-recall task. Brain activity during rest was recorded with EEG, and IAPF was extracted for mediation analyses. Path analysis revealed pronounced sex differences in the association between physical activity, IAPF, and verbal memory performance. Exclusively in female participants, higher vigorous physical activity levels were associated with better recall performance. In contrast, no association of physical activity and memory was found in male participants. However, being more physically active was related to a higher IAPF exclusively in male participants. Physical activity shows differential associations between IAPF and verbal memory in male and female participants. However, the lack of a mediating role of IAPF suggests that this neurophysiological marker cannot explain these specific associations in young adults.
Topics: Humans; Female; Male; Young Adult; Adult; Adolescent; Exercise; Brain; Electroencephalography; Accelerometry; Rest; Verbal Learning; Memory; Sex Characteristics; Mental Recall; Alpha Rhythm
PubMed: 38876578
DOI: 10.1016/bs.pbr.2024.01.006 -
American Journal of Speech-language... Jun 2024Repetitive transcranial magnetic stimulation (rTMS) can enhance aphasia recovery. Most studies have used inhibitory stimulation targeting the right inferior frontal...
Neuromodulation of the Right Motor Cortex of the Lips With Repetitive Transcranial Magnetic Stimulation to Reduce Phonological Impairment and Improve Naming in Three Persons With Aphasia: A Single-Case Experimental Design.
PURPOSE
Repetitive transcranial magnetic stimulation (rTMS) can enhance aphasia recovery. Most studies have used inhibitory stimulation targeting the right inferior frontal gyrus. However, the motor cortex, observed to contribute to the prediction of aphasia recovery, is involved in word production and could be an appropriate target for rTMS. We aimed to observe behavioral changes in a picture naming task induced by inhibitory rTMS targeting the right motor cortex of the lips in people with poststroke aphasia.
METHOD
Using a single-case experimental design, we included three participants with chronic poststroke aphasia who had phonological deficits. Each participant performed a verbal picture naming task 3 times a week for 2, 3, or 4 weeks (pseudorandom across participants) to establish a baseline naming ability for each participant. These were not therapy sessions, and no feedback was provided. Then, each participant received the intervention, inhibitory continuous theta burst stimulation targeting the right motor cortex of the lips, 3 times a week for 2 weeks. Naming testing continued 3 times a week, for these latter 2 weeks. No therapy was performed at any time during the study.
RESULTS
Visual analysis of the graphs showed a positive effect of rTMS for P2 and P3 on picture naming accuracy and a tendency toward improvement for P1. Statistical analysis showed an improvement after rTMS for P1 (τ = 0.544, 013, = 0.288) and P2 (τ = 0.708, 001, = 0.235). For P3, even if the intervention allowed some improvement, this was statistically nonsignificant due to a learning effect during the baseline naming testing, which lasted the longest, 4 weeks. Regarding specific language features, phonological errors significantly decreased in all patients.
CONCLUSIONS
The motor cortex of the lips could be an appropriate target for rTMS to improve naming in people with poststroke aphasia suffering from a phonological deficit. This suggests the possibility to individualize the target for rTMS, according to the patient's linguistic impairment.
PubMed: 38875479
DOI: 10.1044/2024_AJSLP-23-00215 -
Applied Psychophysiology and Biofeedback Jun 2024This review describes the historical developement and rationale of clinically relevant research on neurophysiological "mind reading" paradims: Brain-... (Review)
Review
This review describes the historical developement and rationale of clinically relevant research on neurophysiological "mind reading" paradims: Brain- Computer-Interfaces, detection of deception, brain stimulation and neurofeedback and the clinical applications in drug resistant epilepsy, chronic stroke, and communication with paralyzed locked-in persons. The emphasis lies on completely locked-in patients with amyotrophic lateral sclerosis using non-invasive and invasive brain computer interfaces and neurofeedback to restore verbal communication with the social environment. In the second part of the article we argue that success and failure of neurophysiological "mind reading" paradigms may be explained with a motor theory of thinking and emotion in combination with learning theory. The ethical implications of brain computer interface and neurofeedback approaches, particularly for severe chronic paralysis and loss of communication diseases and decisions on hastened death and euthanasia are discussed.
PubMed: 38874845
DOI: 10.1007/s10484-024-09648-z -
The Journals of Gerontology. Series B,... Jun 2024The objectives of this project were to: (1) examine the relationship between the number of biological children and hippocampal-dependent cognitive performance among...
OBJECTIVES
The objectives of this project were to: (1) examine the relationship between the number of biological children and hippocampal-dependent cognitive performance among older African American women and (2) determine the influence of socioeconomic status (i.e., age, education, marital status, median household income), if any, on this relationship.
METHODS
146 cognitively unimpaired African American women aged 60 and older were recruited from the greater Newark area and reported their number of biological children, marital status, educational level, and age. We retrieved median household income from census tract data based on the participants' addresses. Participants' cognitive performance was assessed using the Rey Auditory Verbal Learning Test (RAVLT) long delay recall and a Rutgers generalization task (Concurrent Discrimination and Transfer Task).
RESULTS
As the number of biological children a woman has had increases, the number of generalization errors also increased, indicating poorer hippocampal-dependent cognitive performance when controlling for age, education, marital status, and median household income. There was no significant relationship between the number of children and performance on a standardized neuropsychological measure of episodic memory (RAVLT), although education was a significant covariate.
DISCUSSION
Generalization tasks may better capture early changes in cognitive performance in older African American women who have had children than standardized neuropsychological assessments. This finding may be explained by the fluctuations in estrogen associated with having children. Future studies should explore how these findings can be applied to protecting cognitive function and preventing AD in older African American women who have had children.
PubMed: 38874309
DOI: 10.1093/geronb/gbae110