Harnessing Speech-Derived Digital Biomarkers to Detect and Quantify Cognitive Decline Severity in Older Adults.

INTRODUCTION: Current cognitive assessments suffer from floor/ceiling and practice effects, poor psychometric performance in mild cases, and repeated assessment effects. This study explores the use of digital speech analysis as an alternative tool for determining cognitive impairment. The study specifically focuses on identifying the digital speech biomarkers associated with cognitive impairment and its severity. METHODS: We recruited older adults with varying cognitive health. Their speech data, recorded via a wearable microphone during the reading aloud of a standard passage, were processed to derive digital biomarkers such as timing, pitch, and loudness. Cohen's d effect size highlighted group differences, and correlations were drawn to the Montreal Cognitive Assessment (MoCA). A stepwise approach using a Random Forest model was implemented to distinguish cognitive states using speech data and predict MoCA scores based on highly correlated features. RESULTS: The study comprised 59 participants, with 36 demonstrating cognitive impairment and 23 serving as cognitively intact controls. Among all assessed parameters, similarity, as determined by Dynamic Time Warping (DTW), exhibited the most substantial positive correlation (rho = 0.529, p < 0.001), while timing parameters, specifically the ratio of extra words, revealed the strongest negative correlation (rho = -0.441, p < 0.001) with MoCA scores. Optimal discriminative performance was achieved with a combination of four speech parameters: total pause time, speech-to-pause ratio, similarity via DTW, and intelligibility via DTW. Precision and balanced accuracy scores were found to be 88.1 ± 1.2% and 76.3 ± 1.3%, respectively. DISCUSSION: Our research proposes that reading-derived speech data facilitates the differentiation between cognitively impaired individuals and cognitively intact, age-matched older adults. Specifically, parameters based on timing and similarity within speech data provide an effective gauge of cognitive impairment severity. These results suggest speech analysis as a viable digital biomarker for early detection and monitoring of cognitive impairment, offering novel approaches in dementia care.

Can irony regulate negative emotion? Evidence from behaviour and ERPs.

This study used ratings and event-related potentials (ERPs) to compare the mechanisms through which verbal irony and cognitive reappraisal mitigate negative emotion. Verbal irony is when the literal meaning of words contrasts with their intended meaning. Cognitive reappraisal is when we reconsider emotional stimuli to make them less intense. Our hypothesis was that cognitive reappraisal is a potential mechanism through which irony reduces negative emotion. Participants viewed mildly negative pictures first, then read an ironic or literal statement about it in one block, and used cognitive reappraisal of or attending to the picture in the other block. Participants then viewed the picture for a second time, before rating how negative they felt. Behaviourally, irony reduced negative feelings more than literal statements, and reappraisal reduced negative feelings more than attending, with a larger reduction from reappraisal than from irony. In ERPs, irony elicited a prolonged N400 compared to literal, indexing an initial contrast between picture and word affect and sustained processing of their combination. Cognitive reappraisal elicited a larger late positivity compared to attending at the instruction screen. No differences were found during second picture presentation. These findings suggest that irony and cognitive reappraisal can reduce negative affect in different ways.

How ready is speech-to-text for psychological language research? Evaluating the validity of AI-generated English transcripts for analyzing free-spoken responses in younger and older adults.

For the longest time, the gold standard in preparing spoken language corpora for text analysis in psychology was using human transcription. However, such standard comes at extensive cost, and creates barriers to quantitative spoken language analysis that recent advances in speech-to-text technology could address. The current study quantifies the accuracy of AI-generated transcripts compared to human-corrected transcripts across younger (n = 100) and older (n = 92) adults and two spoken language tasks. Further, it evaluates the validity of Linguistic Inquiry and Word Count (LIWC)-features extracted from these two kinds of transcripts, as well as transcripts specifically prepared for LIWC analyses via tagging. We find that overall, AI-generated transcripts are highly accurate with a word error rate of 2.50% to 3.36%, albeit being slightly less accurate for younger compared to older adults. LIWC features extracted from either transcripts are highly correlated, while the tagging procedure significantly alters filler word categories. Based on these results, automatic speech-to-text appears to be ready for psychological language research when using spoken language tasks in relatively quiet environments, unless filler words are of interest to researchers.

Narratives about Cancer: What Metaphors can tell us about Depressive Symptoms in Breast Cancer Patients.

Metaphors are pervasive in cancer discourse. However, little is known about how metaphor use develops over time within the same patient, and how metaphor use and its content relate to the mental health of the patient. Here, we analyzed metaphor use in personal essays written by breast cancer patients shortly after the time of diagnosis and nine months later, in relation to their depressive symptoms at both time points. Results show that metaphor use can provide important insight into a patient's current mental state. Specifically, patients who had no change in their depressive symptom levels used metaphors more densely after nine months. In addition, metaphor valence in the later essay was associated with depressive symptoms at study entry and nine months after. Lastly, we observed a shift in metaphor reference pattern for different symptom trajectories, such that those who recovered from initially elevated depressive symptoms used fewer self-referencing metaphors and more cancer-referencing metaphors in their later essay. Our work suggests that metaphor use reflects how a patient is coping with their diagnosis.

Social acquisition context matters: Increased neural responses for native but not nonnative taboo words.

This study examined whether the context of acquisition of a word influences its visual recognition and subsequent processing. We utilized taboo words, whose meanings are typically acquired socially, to ensure that differences in processing were based on learned social taboo, rather than proficiency. American English-speaking participants made word/non-word decisions on American taboo (native dialect), British taboo (non-native dialect), positive, neutral, and pseudo- words while EEG was recorded. Taboo words were verified as taboo by both American and British English speakers in an independent norming survey. American taboo words showed a more positive amplitude of the Late Positive Complex (LPC), a neural correlate of emotionality and social processing, compared with British taboo words and all other word categories. Moreover, in an item-wise analysis, LPC amplitudes of American taboo words were positively correlated with their taboo ratings. British taboo words did not show this effect. This indicates that American participants, who had very limited social contact with British English, did not have the same perception of social threat from British taboo words as they had from American taboo words. These results point to the importance of social context of acquisition in establishing social-affective meaning in language.

Mixed and ambiguous emotions can be studied with verbal irony.

In this commentary we draw attention to a context involving mixed and ambiguous emotions: verbal irony. Irony is frequently used, evokes mixed emotional responses (e.g., criticism and amusement), and has been the focus of recent cognitive neuroscience research. Yet, irony has primarily been studied as a linguistic device, and has rarely been considered by emotion researchers. Similarly, linguistics has not considered mixed and ambiguous emotion when studying verbal irony. We argue that verbal irony offers rich opportunities to evoke and study mixed and ambiguous emotions, and might provide advantages for testing the MA-EM model.

Cursed Concepts: New insights on combinatorial processing from ERP correlates of swearing in context.

Expressives (damn) convey speaker attitude and when used in context (Tom lost the damn dog) can be flexibly applied locally to the noun (dog) or globally to the whole sentence (the situation). We used ERPs to explore brain responses to expressives in sentences. Participants read expressive, descriptive, and pseudoword adjectives followed by nouns in sentences (The damn/black/flerg dog peed on the couch). At the adjective late-positivity-component (LPC), expressives and descriptives showed no difference, suggesting reduced social threat and that readers employ a 'wait-and-see' strategy to interpret expressives. Nouns preceded by expressives elicited a larger frontal P200, as well as reduced N400 and LPC than nouns preceded by descriptives. We associated the frontal P200 with emotional salience, the frontal N400 with mental imagery, and the LPC with cognitive load for combinatorics. We suggest that expressive adjectives are not bound to conceptual integration and conclude that parsers wait-and-see what is being damned.

The comprehension of irony in high and low emotional contexts.

Verbal irony is when words intend the opposite of their literal meaning. We investigated the emotional function of irony by asking whether irony intensifies or mitigates negative feelings. Experiment 1 used ratings to assess the mental state of a speaker using irony or literal language following a negative event in either a high- or a low-emotional context. We found that regardless of context emotionality, speakers using irony were perceived as being in a less negative and less aroused mental state than speakers using literal language. In Experiment 2, we examined the time course of this process with ERPs. Initially, literal statements elicited a larger N100 than irony, regardless of context emotionality, suggesting that irony mitigates negative feelings overall. Later on, irony elicited a larger LPC than literal statements in high emotion contexts, but not in low emotion contexts. This suggests that irony required more mental state processing or/and more speaker emotion processing than literal language in emotionally loaded situations. These results indicate that whether irony intensifies or mitigates negative feelings depends on context and the point in time at which we assess its function. (PsycInfo Database Record (c) 2021 APA, all rights reserved).