Using a Discourse Task to Explore Semantic Ability in Persons With Cognitive Impairment

Abstract

This paper uses a discourse task to explore aspects of semantic production in persons with various degree of cognitive impairment and healthy controls. The purpose of the study was to test if an in-depth semantic analysis of a cognitive-linguistic challenging discourse task could differentiate persons with a cognitive decline from those with a stable cognitive impairment. Both quantitative measures of semantic ability, using tests of oral lexical retrieval, and qualitative analysis of a narrative were used to detect semantic difficulties. Besides group comparisons a classification experiment was performed to investigate if the discourse features could be used to improve classification of the participants who had a stable cognitive impairment from those who had cognitively declined. In sum, both types of assessment methods captured difficulties between the groups, but tests of oral lexical retrieval most successfully differentiated between the cognitively stable and the cognitively declined group. Discourse features improved classification accuracy and the best combination of features discriminated between participants with a stable cognitive impairment and those who had cognitively declined with an area under the curve (AUC) of 0.93.

Learning implicit sentiments in Alzheimer's disease recognition with contextual attention features

“…In addition, these articles are not exhaustive because of our limited ability. Of all the studies in Table 4, the first set of studies (Becker et al, 1994;Clarke et al, 2013;Yancheva and Rudzicz, 2016;Sirts et al, 2017;Hernández-Domínguez et al, 2018;Fraser et al, 2019;Li et al, 2019;Antonsson et al, 2021;R'mani and James, 2021;Zehra et al, 2021) used a feature extraction + machine learning method, and the best accuracy was 85.4%. The second set of studies (Karlekar et al, 2018;Orimaye et al, 2018;Fritsch et al, 2019;Pan et al, 2019;Balagopalan et al, 2021;Guo et al, 2021;Meghanani et al, 2021) used deep learning methods, of which the best accuracy was 91.1% (Karlekar et al, 2018).…”

“…Finally, ASR-extracted features in combination with a Random forest classifier manifested the best results (75% accuracy). For example, Antonsson et al (2021) quantitatively measured the semantic ability, used the Support Vector Machine (SVM) classifier to recognize AD, and finally obtained the best area under the curve (AUC) of 0.93. Clarke et al (2013) measured 286 linguistic features to train the SVM classifier, and the final accuracy obtained was 50-78% for MCI vs. HC, 59-90% for AD vs. HC, and 62-78% for AD+MCI vs. HC.…”

Naming as a window to word retrieval changes in healthy and pathological ageing: Methodological considerations

“…In addition to ecological validity, analysing (semi‐)spontaneous speech samples offers certain supplementary benefits over more constrained methods in clinical settings. Among others, it can be used to assess language deficits in the absence of standardized tools (Abuom & Bastiaanse, 2012), to detect subtle deviant patterns unnoticeable for standard neuropsychological evaluations (e.g., see Jaecks et al., 2012, for residual aphasias; and Antonsson et al., 2021, for MCI), and to obtain baseline measures to assess recovery after intervention (Brookshire & Nicholas, 1994). Hence, given its susceptibility to reveal subtle changes, spontaneous speech analysis stands a priori as a good tool to detect changes in the word retrieval skills of both healthy and pathological groups of ageing individuals, especially in cases of preclinical AD and MCI.…”

“…(2000) report significant differences in the number of semantic errors between elderly and young adults, thus claiming that discourse tasks may be better than structured tasks for assessing word‐finding difficulties in healthy adults. Although further evidence is still needed to confirm the adequacy of (semi‐)spontaneous speech tasks as a meaningful screening tool in cases of preclinical AD and MCI (Antonsson et al., 2021), task selection seems to be crucial to accurately detect early symptoms. Despite the paucity of data, narrative tasks stand as the most adequate task to differentiate these individuals from healthy elderly adults (see Clarke et al., 2021, passim , for dementia).…”

An approach for assisting diagnosis of Alzheimer's disease based on natural language processing

“…Guo et al ( 2021 ) extracted low-level acoustic features such as the number of utterances, speech rate, and vocal events, then employed the Bayesian classifier to train on low speech datasets extracted from the recordings, and finally obtained an accuracy of 68% for classifying AD and older adults controls. Antonsson et al ( 2020 ) measured semantic ability quantitatively and employed the support vector machine (SVM) to recognize AD from NC, and the area under curve (AUC) value is 0.93. Clarke et al ( 2021 ) measured 286 linguistic features to train the SVM model, and the final accuracy is 62–78% for AD+MCI vs. HC, 59–90% for AD vs. HC, and 50–78% for MCI vs. HC.…”