Nadia Guerouaou’s research while affiliated with University of Lille and other places
What is this page?
This page lists works of an author who doesn't have a ResearchGate profile or hasn't added the works to their profile yet. It is automatically generated from public (personal) data to further our legitimate goal of comprehensive and accurate scientific recordkeeping. If you are this author and want this page removed, please let us know.
In a world where our social interactions are increasingly mediated by digital technologies, the possibility to use AI filters to artificially control how we appear and sound to one another is quickly becoming a commodity. Here, we propose the framework of Bayesian inference as a way to question how ``externalizing emotions'' in such a way would affect our social-cognitive functions, both during such interactions and in society as a whole.
The results of recent studies suggested that emotional disorders (such as anxiety and depression), cognitive impairments and cardiovascular disorders are related on the subclinical level. These major health issues are often concomitant and have complex, sex-dependent relationships; it is therefore important to study these issues concomitantly in the general population, in order to gain a better understanding of early-stage subclinical relationships between these conditions. The objective of this exploratory study was to assess correlations between anxiety, depression, cognition, and endothelial function in young adults from the general population. Endothelial function (via the reactive hyperaemia index (RHI) was assessed with a plethysmographic device. Depression and anxiety were self-reported via the Beck Disorder Inventory II and the State-Trait Anxiety Inventory, respectively. The Cambridge Neuropsychological Test Automated Battery was used to measure performances in visuospatial memory, visuospatial working memory, and sustained attention. Performances in inhibition and flexibility were evaluated with the Color Word Interference Test. Forty-four young adults (21 males; mean ± standard deviation age: 25.8 ± 1.1; 23 females; mean age: 25.6 ± 1.4) were included in the study. Anxiety was correlated with a low RHI (r = -0.40, p = 0.015, 95% CI [-0.64, -0.08]). In females, the depression score was positively correlated with the number of errors in the visuospatial memory task (r = 0.42, p = 0.049; 95% CI [-0.002, 0.70]) and visuospatial working memory (r = 0.57, p = 0.005; 95% CI [0.10, 0.79]). In males, high anxiety and depression scores were negatively correlated with the number of errors in visuospatial working memory task (anxiety: r = -0.77, p = 0.001; 95% CI [-0.91, -0.43]; depression r = -0.61, p = 0.004, 95% CI [-0.82, -0.22], respectively). However, the relationship between cognitive performance and RHI was not significant. Our data suggest that anxiety and depression could be differentially related to cognitive and endothelial functions in a non-clinical population of young adults. More research is needed to confirm these results, understand the pathophysiological mechanisms in more details, and assess the importance of a sex-specific approach.
Rapid technological advances in artificial intelligence are creating opportunities for real-time algorithmic modulations of a person’s facial and vocal expressions, or ‘deep-fakes’. These developments raise unprecedented societal and ethical questions which, despite much recent public awareness, are still poorly understood from the point of view of moral psychology. We report here on an experimental ethics study conducted on a sample of N = 303 participants (predominantly young, western and educated), who evaluated the acceptability of vignettes describing potential applications of expressive voice transformation technology. We found that vocal deep-fakes were generally well accepted in the population, notably in a therapeutic context and for emotions judged otherwise difficult to control, and surprisingly, even if the user lies to their interlocutors about using them. Unlike other emerging technologies like autonomous vehicles, there was no evidence of social dilemma in which one would, for example, accept for others what they resent for themselves. The only real obstacle to the massive deployment of vocal deep-fakes appears to be situations where they are applied to a speaker without their knowing, but even the acceptability of such situations was modulated by individual differences in moral values and attitude towards science fiction.
This article is part of the theme issue ‘Voice modulation: from origin and mechanism to social impact (Part II)’.
Rapid technological advances in artificial intelligence are creating opportunities for real-time algorithmic modulations of a person’s facial and vocal expressions, or “deep- fakes”. These developments raises unprecedented societal and ethical questions which, despite much recent public awareness, are still poorly understood from the point of view of moral psychology. We report here on an experimental ethics study conducted on a sample of N=303 participants (predominantly young, western and educated), who evaluated the acceptability of vignettes describing potential applications of expressive voice transformation technology. We found that vocal deep-fakes were generally well accepted in the population, notably in a therapeutic context and for emotions judged otherwise difficult to control, and surprisingly, even if the user lies to their interlocutors about using them. Unlike other emerging technologies like autonomous vehicles, there was no evidence of social dilemma in which one would e.g. accept for others what they resent for themselves. The only real obstacle to the massive deployment of vocal deep-fakes appears to be situations where they are applied to a speaker without their knowing, but even the acceptability of such situations was modulated by individual differences in moral values and attitude towards science-fiction.
Rapid technological advances in artificial intelligence are creating opportunities for real-time algorithmic modulations of a person’s facial and vocal expressions, or “deep- fakes”. These developments raises unprecedented societal and ethical questions which, despite much recent public awareness, are still poorly understood from the point of view of moral psychology. We report here on an empirical study conducted on N=303 online participants, who evaluated the acceptability of vignettes describing potential applications of expressive voice transformation technology. We found that vocal deep- fakes were generally well accepted in the population, notably in a therapeutic context and for emotions judged otherwise difficult to control, and surprisingly, even if the user lies to their interlocutors about using them. Unlike other emerging technologies like autonomous vehicles, there was no evidence of social dilemma in which one would e.g. accept for others what they resent for themselves. The only real obstacle to the massive deployment of vocal deep-fakes appears to be situations where they are applied to a speaker without their knowing, but even the acceptability of such situations was modulated by individual differences in moral values and attitude towards science- fiction.
Background
The prevalence of cognitive impairment and dementia is high and steadily increasing. Early detection of cognitive decline is crucial since some interventions can reduce the risk of progression to dementia. However, there is a lack of manageable scales for assessing cognitive functions outside specialized consultations. Recently, the MoCA-5 min, a short version of the Montreal Cognitive assessment (MoCA), phone-administered, was validated for screening for vascular cognitive impairment. The aim of the present study was to validate the MoCA-5 min in French in diverse clinical populations.
Methods
The Cantonese version of the MoCA-5 min was adapted for French language. Healthy volunteers and patients with possible or established cognitive impairment (Alzheimer's disease or related disorders, Parkinson's disease, Huntington's disease, type-2 diabetes) participated in the study. The original MoCA and the MoCA-5 min were administered, by phone, with a 30-day interval. Alternate forms were used to reduce learning effects.
Results
The scores of the original MoCA and MoCA-5 min correlated significantly (Spearman rho = 0.751, P < 0.0001, 95% confidence interval 0.657 to 0.819). Internal consistency was good (Cronbach alpha = 0.795). The area under the ROC curve was 0.870 and the optimal cut-off value for separating patients with and without cognitive impairment with the MoCA-5 min was ≤ 27 with 87.32% sensitivity and 76.09% specificity. Interrater and test-retest reliability were adequate.
Conclusion
This study demonstrates that the French version of the MoCA-5 min is a valid and reliable scale for detecting cognitive impairment in different clinical populations. It is administrable by phone and thus suitable for remote assessment as well as for large-scale screening and epidemiological studies.
Citations (2)
... Deep neural networks Source: Gil et al., (2023) A pair of algorithms have been employed in combat with one another in order to generate fake data. Each is frequently referred to as a marker, while the other as a machine that generates electricity (Etienne, 2021;Guerouaou et al., 2022). The tool for discrimination has been charged in figuring out if the multimedia material generated through the program that creates it synthetic or legitimate. ...
... In addition, the inclusion form will collect the town and site where the data collection will occur. Potential participants will then be prescreened for dementia and excluded if MoCA-5 minute (Mini MoCA) scores are <12 (clinically important risk of cognitive impairment) [45]. This brief task takes �5 minutes and assesses the domains of attention, executive function/language, orientation, and memory. ...