How animacy processing is affected in early stages of AD. (1st February 2022)

• Record Type: Journal Article
• Authors: Marefat, Haniye; Ebrahiminia, Fatemeh; Khanbagi, Mahdiyeh; Karimi, Hamed; Modarres, Mohammad Hadi; Kalafatis, Chris; Vahabi, Zahra; Khaligh‐Razavi, Seyed‐Mahdi
• Is Part Of: Alzheimer's & dementia. Volume 17(2021)Supplement 4
• Page Start: n/a
• Page End: n/a
• Publication Date: 2022-02-01
• Subjects: Alzheimer's disease -- Periodicals; Alzheimer Disease -- Periodicals; Dementia -- Periodicals; Démence; Maladie d'Alzheimer; Périodique électronique (Descripteur de forme); Ressource Internet (Descripteur de forme); 616.83
• DOI: 10.1002/alz.053837
• Languages: English
• ISSNs: 1552-5260

Abstract

Abstract: Background: Previous studies have shown that cognitive impairment in early stages of Alzheimer's disease (AD) is not limited to memory; other cognitive domains are also disrupted including impairment in visual processing (Boucart, Bubbico et al. 2014). Additionally, it is shown that in healthy human subjects, images are categorized by their animacy status (i.e. animate vs. inanimate) in the higher‐level visual areas (Kriegeskorte, Mur et al. 2008). Here we used fMRI to investigate the spatial patterns of animacy processing in the brain of patients with mild cognitive impairment (MCI)/AD in comparison to healthy controls(HC). Method: We recruited 20 HCs, 15 MCIs and 10 mild‐ADs. Participants completed the Integrated Cognitive Assessment (ICA) test (Khaligh‐Razavi, Habibi et al. 2019) in the fMRI scanner. ICA is a rapid visual‐categorization task, in which participants indicate whether a presented image contains an animal or not, as fast and accurately as they can. In addition to the conventional univariate analysis –looking at the mean brain responses– we used multivariate‐pattern‐analysis (MVPA) to investigate differences in patterns of brain activity between the three groups. In MVPA, we trained a linear classifier for each brain area, to do animal/non‐animal categorization based on the patterns of brain responses to the images. Result: In some of the key brain regions, such as R‐parahippocampus and R‐fusiform, overlapping with early tau‐pathology (Braak, Alafuzoff et al. 2006), both HC and MCI showed a significantly higher level of mean brain activity in response to all images compared to the AD group (p‐value<0.0001). Additionally, looking at the patterns of brain activity in the L‐fusiform, HC and MCI could be discriminated based on their MVPA responses(p‐value<0.0015), while there was no univariate difference between the two groups. Conclusion: We demonstrated HC and MCI/AD can be differentiated not only based on univariate analysis, but also by multivariate measures, i.e. patterns of brain responses to images. Moreover, we showed that the rapid visual categorization task used in this study could engage key brain areas affected by tau pathology in early stages of the disease. This highlights the potential application of such task for earlier detection of the disease, ideally before the onset of memory symptoms.