Temporal dynamics of animacy categorization in the brain of patients with mild cognitive impairment

Hamed Karimi; Haniyeh Marefat; Mahdiyeh Khanbagi; Chris Kalafatis; Mohammad Hadi Modarres; Zahra Vahabi; Seyed-Mahdi Khaligh-Razavi

doi:10.1371/journal.pone.0264058

. 2022 Feb 23;17(2):e0264058. doi: 10.1371/journal.pone.0264058

Temporal dynamics of animacy categorization in the brain of patients with mild cognitive impairment

Hamed Karimi ^1,^2,^*, Haniyeh Marefat ³, Mahdiyeh Khanbagi ¹, Chris Kalafatis ^4,^5,⁶, Mohammad Hadi Modarres ⁶, Zahra Vahabi ^7,⁸, Seyed-Mahdi Khaligh-Razavi ^1,^6,^*

Editor: Stavros I Dimitriadis⁹

PMCID: PMC8865635 PMID: 35196356

Abstract

Electroencephalography (EEG) has been commonly used to measure brain alterations in Alzheimer’s Disease (AD). However, reported changes are limited to those obtained from using univariate measures, including activation level and frequency bands. To look beyond the activation level, we used multivariate pattern analysis (MVPA) to extract patterns of information from EEG responses to images in an animacy categorization task. Comparing healthy controls (HC) with patients with mild cognitive impairment (MCI), we found that the neural speed of animacy information processing is decreased in MCI patients. Moreover, we found critical time-points during which the representational pattern of animacy for MCI patients was significantly discriminable from that of HC, while the activation level remained unchanged. Together, these results suggest that the speed and pattern of animacy information processing provide clinically useful information as a potential biomarker for detecting early changes in MCI and AD patients.

1 Introduction

Mild Cognitive Impairment (MCI) is a condition in which an individual has a mild but measurable decline in cognitive abilities. This decline is noticeable to the person affected and to the family members and friends, but the individual can still carry out everyday activities [1, 2]. A systematic review of 32 cohort studies shows an average of 32 percent conversion from MCI to Alzheimer’s Disease (AD) within a five-year follow-up [3]. 5–15% of people with MCI have also been shown to develop dementia every year [4].

Electroencephalography (EEG) is widely used to study the resting-state neural activity in the brain of patients with MCI and mild AD [5–14]. A few studies have also used EEG to associate abnormalities in memory function with cognitive impairment during both encoding and decoding stages of working memory [7, 15–17] as well as episodic memory tasks [18].

The relationship between cognitive impairment and visual system changes has recently gained attention [19]. Several studies have linked deficiencies in different parts of the visual system with AD [20–22]. There are several documented cases in which visual function problems are the initial and dominant manifestation of dementia [23, 24]. A few studies have also used a visual task to report changes in the EEG responses of patients with MCI and AD [25, 26].

Several studies of the visual system in primates and healthy human subjects have demonstrated that images are categorized by their animacy status (i.e., animate vs. inanimate) in the higher-level visual areas, i.e., inferior temporal (IT) cortex [27–32]. The neural activity underlying the animacy information processing of briefly flashed images was also studied in healthy adults [33, 34]. Studies have shown that animacy information emerges in the brain of healthy human subjects as early as 80±20 ms after the stimulus onset and reaches its peak within 250±50ms after the stimulus onset [35–37]. However, the underlying neural dynamics of animacy processing in the brain of patients with MCI in comparison to healthy controls (HC) is still unknown. Several studies using autopsy [38–41] and PET imaging [42, 43] have shown that some of the visual areas in the temporal cortex are among the first regions affected by the disease.

The Integrated Cognitive Assessment (ICA) is a visual task based on a rapid categorization of natural images of animals and non-animals [44]. ICA assesses changes in the speed and accuracy of animacy processing in patients with MCI and mild AD and is shown to be sensitive in detecting MCI patients [45, 46]. The ICA primarily tests Information processing speed (IPS) and engages higher-level visual areas in the brain for semantic processing, i.e., distinguishing animal vs. non-animal images [44]. Reduced visual processing speed is reported in amnestic-MCI [47], and IPS is further reported to be related to other areas of cognitive dysfunction [48, 49].

In line with previous behavioral studies, we hypothesized that the underlying neural response of MCI patients during an animacy categorization task is both slower and less accurate at the level of representation compared to HC [44]. The absence of multivariate methods in previous studies also raises the question of whether the pattern of animacy information processing is different in patients with MCI compared to that of HC.

To address these questions, we acquired EEG data from MCI and HC participants during the completion of the ICA’s animal/non-animal categorization task. We studied the temporal neural dynamics of animacy processing in MCI patients using both univariate and multivariate analyses. By applying multivariate pattern analysis (MVPA), we compared the neural speed of animacy processing in MCI and HC. We further looked beyond the conventional univariate methods and compared MCI and HC in terms of their pattern of EEG responses to natural images of animal and non-animal stimuli.

We found that the categorical representation of animacy information emerges later in the brain of patients with MCI compared to that of HC. Furthermore, the results reveal differences between the EEG response patterns of HC vs. MCI during the time-points when univariate mean responses showed no significant difference. The EEG response patterns could further be used to discriminate HC from MCI, demonstrating that the pattern of EEG activity also carries information about the status of the disease beyond the conventional univariate analysis of mean activities.

2 Methods

2.1 Integrated cognitive assessment (ICA) task

ICA [44, 50], [51, p.] is a rapid animal vs. non-animal categorization task. The participants are presented with natural images of animals and non-animals. The images are followed by a short blank screen and then a dynamic mask. Participants should categorize the images as animal or non-animal as quickly and accurately as possible (Fig 1). See (Khaligh-Razavi et al. 2019 [44], Fig 1B) for sample images of the task.

Fig 1 — The EEG experiment contained 13 experimental runs; in each run, 32 natural images (16 animal, 16 non-animal) were presented to the participant in a random order. Each image was shown for 100 ms, followed by an inter-stimulus interval (ISI) of 20ms and a dynamic noise mask for 250ms. Participants were given 1500 ms to indicate (using pre-specified buttons) whether the presented image contained an animal or not.

2.2 Montreal cognitive assessment (MoCA)

MoCA [52] is a ten-minute pen and paper test with a maximum score of 30 and is conventionally used to assess visuospatial, memory, attention, and language abilities to detect cognitive impairment in older adults. An examiner is needed to administer the test. The results of this test were used by the consultant neurologist to help with the diagnosis of participants.

2.3 Addenbrooke cognitive examination (ACE-R)

ACE-R [53] is another pen-and-paper tool for cognitive assessment with a maximum score of 100. It mainly assesses five cognitive domains: attention, orientation, memory, fluency, language, and visuospatial. On average, the test takes about 20 to 30 minutes to administer and score. The results of this test were used by the consultant neurologist to help with the diagnosis of participants.

2.4 Subject recruitment

40 (22 Healthy, 18 MCI) participants (Table 1) completed the ICA test, MoCA, and ACE-R in the first assessment. The participants were non-English speakers, with instructions for the cognitive assessments provided in Farsi. The ICA test was administered on an iPad.

Table 1. Demographic information of participants.

Characteristic	HC (n = 22)	MCI (n = 18)	P-values
Age–mean years ±SD	63.41±6.10	63.56±6.41	0.94
Education in years–mean ±SD	15±4.18	14.72±5.02	0.85
Gender (%female)	13 (59%)	10 (55%)	0.82

Open in a new tab

SD: standard deviation.

p-values were calculated from two-sided t-test for two independent samples.

Patients were recruited into the study prospectively. A consultant neurologist made all the diagnoses according to diagnostic criteria described by the working group formed by the National Institute of Neurological and Communicative Disorders and Stroke (NINCDS) and the Alzheimer’s Disease and Related Disorders Association (ADRDA) (referred to as the NINCDS-ADRDA criteria) and the National Institute on Aging and Alzheimer’s Association (NIA-AA) diagnostic guidelines. The study was conducted at the Royan Institute, according to the Declaration of Helsinki, and approved by the local ethics committee at the Institute. The inclusion/exclusion criteria are listed below.

Inclusion criteria for the HC group:

Males and females aged between 50–85 years who are not currently on medication that may interfere with the study results and are in good general health were included in the study.
Inclusion criteria for the MCI group:

Males and females aged between 50–85 years with a clinical diagnosis of MCI who were able to provide informed consent were included in the study.
Exclusion criteria for both groups:

Individuals with a presence of significant cerebrovascular disease or major psychiatric disorder (e.g., chronic psychosis, recurrent depressive disorder, generalized anxiety disorder, and bipolar mood disorder) or major medical comorbidities (e.g., congestive cardiac failure, diabetes mellitus with renal impairment) were excluded from the study.

Additional exclusion criteria were: use of cognitive-enhancing drugs (e.g., cholinesterase inhibitors), or a concurrent diagnosis of epilepsy or any history of alcohol misuse, illicit drug abuse, severe visual impairment (e.g., macular degeneration, diabetic retinopathy, as determined by the clinical examination), or repeated head trauma.

2.5 EEG data acquisition and preprocessing

The EEG experiment included a short version of the ICA task (i.e., smaller image set). Participants completed one EEG session that included 13 runs; each run lasted 67 seconds, during which 32 images were presented in random order, and participants had a short break in between the runs. Each stimulus was repeated 13 times during the whole EEG session (once in each run). These are referred to as repetition trials throughout the manuscript. Participants 16 and 17 completed 10 runs, and participants 12 and 22 completed 12 runs. We acquired the EEG data using a 64-channel (63 electrodes on the cap and one as the reference; for electrodes layout, see S1 Fig of the supplementary materials) g.tec product at a sampling rate of 1200 Hz. Three electrooculograms (EOG) channels were set up to capture eye blinks. Additionally, we included resting trials in between the image trials (i.e., almost every 70 seconds, they were given 10 seconds to rest their eyes, blink, and swallow). Participants were instructed to only blink (or swallow) during these trials to prevent contamination of EEG signals with the eye-blink (and swallowing) artifacts. These trials were excluded from further EEG analyses. Because of such a design, we did not have to reject any of the image trials. Other potential artifacts were removed with Independent Component Analysis.

The preprocessing consisted of six general steps, which were all done using Brainstorm [54] in MATLAB:

Re-referencing the data with the mean activation and removing the reference channel (channel 33).
Neutralizing eye blinks by removing the most correlated component with the EOG channels, using the independent component analysis algorithm.
Extracting pre-stimulus data from 100 ms before to 800 ms after the stimulus onset (epoching).
Normalizing the epochs regarding the mean and standard deviation of the baseline.
Smoothing the data with a 50 Hz low-pass filter.
Resampling the data to 1000 Hz sampling rate.

2.6 Univariate pattern analysis—Event-Related Potential (ERP)

We calculated the ERPs of the extracted epochs (from 100 ms before to 800 ms after the stimulus onset) by averaging the EEG responses to all stimuli within each group of channels. We calculated the ERPs separately for each individual. The ERP of HC and MCI are the average ERP of corresponding participants.

2.7 Support vector classifier

We used a linear support vector machine (SVM) classifier throughout the analyses to decode neural data. The hyperparameters were as follows: The cost/regularization parameter (C) and the weight of classes were all set to 1. All the classifications were done using the LIBSVM software implementation [55].

2.8 Multivariate pattern analysis—Animal vs. non-animal decoding

To study the emergence of animacy categorical information in the brain, we used a linear classifier to discriminate EEG responses to animal stimuli from that of non-animal (Fig 2). Before the classification, we randomly assigned each target stimulus with all its EEG trials to bins of sizes 2, 3, and 4 stimuli and randomly sub-averaged the trials within each bin. The decoding accuracy at each time point ‘t’ is then the average accuracy of 10,000 repetitions in a leave-one-bin-out cross-validation model, using an SVM classifier.

Fig 2 — We extracted repetition trials of EEG responses to animal and non-animal stimuli at each time-point (for a given time-point t we had two matrices of 63 channels x16 stimuli x13 repetition trials of EEG responses to animal and non-animal stimuli). Before the classification, we randomly assigned each target stimulus with all its EEG trials to bins of size 2, 3 and 4 stimuli and randomly sub-averaged the trials within each bin. We trained a leave-one-bin-out cross-validation SVM model to discriminate animal from non-animal. At each time-point, the decoding accuracy is the average of 10,000 repetitions of the classification procedure described above. We repeated the entire process of each individual separately.

2.9 Multivariate pattern analysis—Pairwise decoding

At each time point, we measured the accuracy of an SVM classifier in discriminating pairs of stimuli using leave-one-out cross-validation. This leads to a symmetric 32×32 representational dissimilarity matrix (RDM) at each time point, representing the pairwise dissimilarities of stimuli in the off-diagonal elements (Fig 3). We repeated the entire process for each individual to create an RDM at each time point.

Fig 3 — We trained an SVM to discriminate pairs of stimuli using their EEG responses at time-point t. This pairwise classification of stimuli by cross-validation SVM model leads to a 32 × 32 RDM at each time-point t.

2.10 Multivariate pattern analysis—HC vs. MCI classification

We characterized the activation pattern at each time-point as a 63×32 matrix, with each column being the responses of the EEG channels to a stimulus, averaged over all repetition trials (Fig 4). We applied 10,000 bootstrap resampling (without replacement) of participants and trained a leave-one-out cross-validation SVM model to discriminate the EEG activation pattern of HC from that of MCI.

Fig 4 — At each time-point, the pattern of EEG activation is a 63x32 matrix with columns being the EEG responses of channels to the 32 stimuli. We applied 10,000 bootstrap resamplings (without replacement) of participants and each time trained a leave-one-out cross-validation SVM model to discriminate HC from MCI based on their EEG activation patterns.

2.11 Multidimensional Scaling (MDS)

Multidimensional Scaling (MDS) is a non-linear dimension reduction algorithm. It rearranges the data points in a p-dimensional space until their pairwise distance is consistent with a given dissimilarity matrix. Here, we used MDS to visualize the stimuli on a 2D plane based on their pairwise dissimilarity in RDMs.

2.12 Statistical analysis

To avoid any assumption about the observed distributions, we only used non-parametric statistical tests. They are capable of testing a null hypothesis without any prior assumptions about the nature of the distribution:

2.12.1 Bootstrap test

Bootstrapping is a strategy to estimate different statistics over an unknown distribution. It consists of a resampling (with or without replacement) procedure and applying a target function. The result is a bootstrap distribution that can be used for statistical inference purposes. We used 10,000 bootstrap resampling of participants without replacement from each group (HC and MCI) and computed a p-value as follows:

p - v a l u e = \frac{n u m b e r o f e l e m e n t s l o w e r (o r h i g h e r) t h a n b a s e l i n e + 1}{n u m b e r o f b o o t s t r a p r e s a m p l i n g (10000) + 1}

2.12.2 Permutation test

The permutation test consists of randomly relabeling the samples from two populations to form a null distribution. It computes a p-value by testing a target statistic against the null hypothesis:

p - v a l u e = \frac{n u m b e r o f m e m b e r s f r o m t h e n u l l d i s t i b u t i o n l o w e r (o r h i g h e r) t h a n t h e t a r g e t + 1}{n u m b e r o f p e r m u t a t i o n s (10000) + 1}

2.12.3 Rank-sum

Rank-sum (also known as Wilcoxon–Mann–Whitney test) tests the null hypothesis that the data in x and y are sampled from continuous distributions with equal medians, against the alternative that they are not [56]. We used rank-sum to compute the p-value when comparing HC and MCI medians of animal vs. non-animal decoding amplitude, ICA speed and accuracy and, mean of ERP responses.

3 Results

3.1 A reduction in the speed and accuracy of animacy processing in MCI patients

We compared the neural speed and accuracy of animal/non-animal discrimination between HC and MCI. To this end, for each group, we computed the time at which animal images can best be discriminated from non-animals based on their EEG responses. This time-point is referred to as the peak of animal/non-animal decoding. MCI patients showed a median delay of 39 ms (95% CI = [8, 111], SE = 34 ms) in processing the animacy information in comparison to healthy individuals (p-value = 0.0001; 10000 bootstrap resampling of participants, Fig 5A). Additionally, in this decoding peak time-point, the neural accuracy of animal detection was significantly lower in MCI patients compared to healthy controls (p-value = 0.018; rank-sum, Fig 5B).

Fig 5 — a) Median of animal vs. non-animal decoding accuracy peak time-points. 10000 bootstrap resampling without replacement of participants was applied to measure the differences between HC and MCI. The difference of medians of HC and MCI peaks is 39 ms (p-value = 0.0001; One-sided bootstrap test) b) Box-plot of animal vs. non-animal EEG decoding peak amplitude. The difference of HC and MCI amplitude medians is 5% (p-value = 0.018; rank-sum). c) Mean EEG RDM of participants at the time of animal vs. non-animal decoding peak. No significant difference was observed between the RDMs (permutation test). d) Mean ERP of participants at the time of animal vs. non-animal decoding peak. The black dots indicate significant activation in channels of the specified region (FDR-corrected at 0.05 sign-rank). No significant difference between the ERP of HC and MCI was observed (FDR-Corrected at 0.05 rank-sum). e) Box-plot of the ICA test accuracy (results of the behavioral ICA, taken on iPad). A significant difference is observed between HC and MCI (p-value = 0.037; rank-sum). f) Box-plot of the ICA test speed (results of the behavioral ICA, taken on iPad). A significant difference is observed between HC and MCI (p-value = 0.025; rank-sum). Fp: prefrontal; AFL: left anterior frontal; AFR: right anterior frontal; FL left frontal; Fz: midline frontal; FR: right frontal; FCL: left fronto-central; FCR: right fronto-central; TL: left temporal; CL: left central; Cz: midline central; CR: right central; TR: right temporal; TPL: left temporo-parietal; CPL: left centro-parietal; CPR: right centro-parietal; TPR: right temporo-parietal; PL: left parietal; PR: right parietal; POL: left parieto-occipital; Pz: midline parietal; POR: right parieto-occipital; O: occipital.

The representational dissimilarity method is a way to represent patterns of brain information processing [57]. To study the neural representation underlying animacy categorization, we compared the representational dissimilarity matrices (RDM) and ERP responses (Fig 5C and 5D) of the two groups at the time of each individual’s animal vs. non-animal decoding peak.

The HC/MCI RDM in the peak animacy decoding time-point represents the pattern of EEG responses at a time-point in which the brain representation of animal images is best separated from that of non-animals (Fig 5C). While the peak animacy time-point was delayed for the MCI group, the MCI RDM was not significantly different from that of the HC RDM in its peak, suggesting that it took more time for the MCI patients to converge to a pattern similar to that of the HC, which is used for discriminating animals from non-animals.

We also looked at the ERP responses at these peak time-points between HC and MCI: channels in the right and left parietals, right temporal-parietal, midline parietal, and midline frontal lobes were significantly activated in both groups. HC showed significant activation levels in the right parieto-occipital, right temporal, and central lobes (midline central, right and left central-parietals, right central, and fronto-central); however, these were absent from the MCI brain activation map. On the other hand, only MCI patients showed significant activation on the left parieto-occipital lobe.

Analyzing the behavioral data of participants while taking the ICA test (on an iPad outside the EEG), we found the results to be consistent with the findings from the EEG data: the speed and accuracy of animacy detection, as measured by the ICA test, were also significantly deteriorated in patients with MCI (p-value = 0.025 and p-value = 0.037 respectively) (Fig 5F and 5G).

We also examined the channel-specific animacy decoding time courses to investigate whether there is a significant delay in the peak and/or the onset of animacy processing in MCI patients at the level of EEG channels (Fig 6). We found that there is a significant delay in the animacy decoding peak time-point of MCI patients in the right parietal lobe compared to HC (p-value = 0.0013, Fig 6). Additionally, the significance of animal vs. non-animal decoding started later in the MCI patients in EEG channels of left fronto-central, midline frontal, midline central, and left parieto-occipital lobes (Fig 6).

Fig 6 — Horizontal color dots indicate time-points with significant decoding accuracy in the corresponding group (green for HC and orange for MCI). In regions specified with red rectangles MCI’s peak animacy decoding time-point is significantly later than that of HC; in regions specified with blue rectangles MCI’s onset of animacy decoding significance is significantly later than that of HC (FDR-corrected at 0.05; bootstrap test with 10,000 resampling of participants on MCI minus HC time-points).

3.2 Comparing patterns of visual information processing in HC and MCI

We compared patterns of visual information processing in HC and MCI using their RDMs over time. The maximum difference between HC and MCI RDMs was observed at t = 224 ms after the stimulus onset (scaled Euclidean distance, d = 0.53 [out of 1], p-value = 0.012; Fig 7A). At this time-point (t = 224 ms), an SVM classifier (leave-one-out cross-validation) could significantly discriminate between HC and MCI patterns of visual information (represented by their RDMs) with an accuracy of 70.4% (p-value = 0.0036, 10,000 bootstrap sampling of participants).

We also looked at the univariate differences between HC and MCI at the same time-point (i.e., t = 224 ms); there was no significant difference between the two groups in their ERP responses (Fig 7B). This demonstrates that the response patterns carry valuable information above and beyond what is captured in the neural activity’s mean responses.

Furthermore, examining the HC and MCI RDMs, we highlighted RDM cells that were significantly different between the two groups (Fig 7A). Each RDM cell represents the difference between the EEG patterns of the two stimuli at a given time-point. To provide a more intuitive understating of the differences in patterns, we used multidimensional scaling (MDS) to visualize the RDMs on a 2D surface. Fig 7C illustrates the stimuli with a significant difference between HC and MCI, and Fig 7D visualizes all the stimuli.

3.3 Temporal dynamics of animacy categorization

To study the temporal dynamics of animacy categorization in HC and MCI, we compared the mean response (i.e., ERP) as well as EEG activation patterns (i.e., 63 channels × 32 stimuli matrices) between HC and MCI over time. For each EEG channel group, we measured the ERP differences between HC and MCI over time: midline frontal, right fronto-central, and right temporal regions showed a significant difference (Fig 8A). We also performed a classification over the EEG activation patterns to see if HC and MCI can be discriminated based on their epoched EEG responses. HC and MCI could be discriminated based on their EEG activation patterns in the left frontal, midline frontal, left parietal, and right central parietal lobes (Fig 8B).

Looking at the EEG data (Fig 8), we found that HC and MCI could be discriminated starting from 375 ms in the left parietal (PL) and from 495 ms in the left frontal (FL) both to 515 ms after the stimulus onset only based on their patterns of activity, but not the ERPs. Additionally, the pattern of activity in centro-parietal (CPR) could discriminate HC from MCI in almost every time-point after t = 655ms. On the other hand, the ERP responses showed a significant difference between HC and MCI in the right fronto-central (FCR) at around t = 405 ms and in the right temporal (TR) from 155 ms to 174 ms and 365 ms to 545 ms. At the same time points, the two groups could not be separated based on their activation patterns. The midline frontal (Fz) was the only region that could differentiate HC and MCI based on both the ERP responses and the activation patterns at around t = 405 ms after the stimulus onset.

In the previous subsection, we demonstrated that at t = 224 ms, the difference between HC and MCI in the EEG response patterns (captured by RDM) was at its maximum, while the level of activity (captured by ERP) remains unchanged. Here, we identified five groups of channels whose EEG data could discriminate between HC and MCI, either based on the activation patterns or the ERP responses, but not both. This is consistent with the reported results in the previous section (3.2), demonstrating that EEG activation patterns could be different even though there might be no difference at the level of ERP.

4 Discussion

In this study, we investigated the temporal dynamics of animacy visual processing in patients with MCI and argued that the speed of processing animacy information is a potential biomarker for detecting MCI. The proposed rapid visual categorization task in the ICA test is more challenging than the typical memory tasks, making it more sensitive to less severe brain deteriorations. The ICA is particularly suited for population-wide screening of cognitive impairment to help identify patients with MCI and mild AD (MiAD) and not designed for cognitive assessment in more severe stages of the disease, such as moderate to severe AD. Early detection of cognitive impairment is becoming increasingly important, particularly in the context of disease-modifying therapies for early stages of AD, such as Aducanumab–which has recently received FDA approval.

Previous resting-state and task-based EEG studies have focused on univariate changes (e.g., ERP, frequency bands, connectivity) in patients with MCI and individuals with mild to moderate AD [5–7, 15–17, 58]. Here, using a rapid visual categorization task and applying multivariate pattern analysis, we looked beyond the univariate changes and studied the categorical representation of animacy information in the brain of old healthy individuals and patients with MCI. We demonstrated that patients with MCI could be discriminated from HC based on their pattern of animacy representation. Furthermore, we identified regions in which either the mean EEG responses or the pattern of brain activity show significant differences between HC and MCI.

Having a closer look at the ERP of different regions, we observed a decrease in the P300 amplitude of the MCI patients. This decrease is significant in the electrodes of the midline frontal, right fronto-central, and right temporal regions between 250 ms to 500 ms after the stimulus onset. This observation is in line with the reported result in [13], especially with the P300 of the midline frontal electrodes, and highlights the importance of this signal in the detection of MCI.

4.1 Task differences in EEG studies of HC and MCI

Consistent with previous reports in resting-state EEG studies [59] and studies with a visual memory task [17], we observed univariate differences between HC and MCI in the temporal and the fronto-central electrodes.

In contrast, we did not find any significant difference in ERP responses of HC vs. MCI in centro-parietal and parietal electrodes–which is reported previously in an EEG study with a visual working memory task [16]. We also observed no univariate difference in the frontal and occipital electrodes, while previous resting-state EEG studies have reported differences between HC and MCI in these regions [59, 60]. On the other hand, we found that MCI patients could be discriminated from HC based on their ERP responses of the midline frontal region electrodes. These differences could potentially be explained by the difference in the tasks used for each of these studies (i.e., visual working memory and resting-state vs. rapid visual categorization).

In addition to the previous univariate findings, here we revealed multivariate differences between HC and MCI in their patterns of EEG responses: midline frontal, left frontal, left parietal, and right centro-parietal electrodes showed significant multivariate differences between HC and MCI. Furthermore, the categorical representation of animacy information of the right parietal electrode emerged later in the MCI patients compared to that of HC. Also, in comparison with HC, the MCI patients had significantly longer onset latencies of animacy information in the left fronto-central, midline frontal, midline central, and left parieto-occipital electrodes.

4.2 What do differences in the pattern of activation mean?

The overall changes in the pattern of EEG responses happen in the regions that are involved in visual processing and motor movement. These regions are engaged during the ICA task’s execution, and their activation is captured through frontal and parietal electrodes.

Neurons of the parietal cortex integrate sensory inputs (visual, auditory, etc.) through motor control regions to execute movements [61–63]. Visuomotor skills that are known to be resolved in regions of the parietal cortex are heavily involved in the ICA task, as the task requires the participant to use visual information to categorize presented images with a movement of both hands.

In the case of frontal regions, univariate changes in the electrodes of this lobe have previously been shown in EEG studies [17, 59]. Additionally, the Posterior-Anterior Shift in Aging (PASA) suggests that the brain of individuals with age-related changes tends to engage other networks in the frontal region to compensate for the decline of processing information in posterior areas [64]. Consistent with PASA, we observed that the pattern of information processing is altered in both frontal and parietal electrodes.

4.3 Neural speed of information processing in MCI patients

Rapid recognition of animate objects is a fundamental ability of human visual cognition. Previous M/EEG studies have investigated the temporal neural dynamics of animacy processing in young, healthy individuals. Using slightly different visual tasks and stimuli, studies have shown that the onset and peak of animacy decoding emerge between 66 ms to 157 ms [35]; or from 80 ms to 240 ms [36] after the stimulus onset. Here, we showed that in old healthy individuals, the onset, and the peak of animacy decoding emerge between 131 ms (SE = 30) and 434 ms (SE = 30) after the stimulus onset. This result indicates that normal aging causes a reduction in the animacy information processing speed (IPS). Compared to the old healthy individuals, animacy IPS was further delayed in MCI patients (onset of animacy decoding: 196 ms±16, peak animacy decoding: 473 ms±34 after the stimulus onset). These findings confirm a significant decrease in the speed of neural information processing in patients with MCI and are consistent with previous behavioral studies showing a decline in visual IPS in MCI patients compared to HC [44, 47]. Together, these results suggest the IPS as a potential biomarker for the detection of MCI patients. However, other complementary biomarkers should be employed for the diagnosis of MCI due to AD.

Some of the study limitations include the relatively small number of patients recruited. Additionally, the study was not a longitudinal study to determine if the MCI patients will convert to AD. In future longitudinal studies, we aim to investigate how well the current results generalize to cohorts of larger MCI/AD patients.

5 Conclusion

We showed that the speed of processing animacy information is decreased in MCI patients compared to healthy individuals. This suggests the use of the ICA test, which is based on the categorization of animal and non-animal images, as a digital biomarker for detecting cognitive impairment in MCI patients.

Furthermore, we showed that in addition to univariate changes, the brains of MCI patients and HC individuals are different in the pattern of representing animacy information.

Supporting information

S1 Fig. EEG electrodes layout.

We used a 64-channel g.tec product at a sampling rate of 1200 Hz for EEG data acquisition (S1 Fig). The reference electrode (#33) was placed on the participant’s right ear. Fp: prefrontal; AFL: left anterior frontal; AFR: right anterior frontal; FL left frontal; Fz: midline frontal; FR: right frontal; FCL: left fronto-central; FCR: right fronto-central; TL: left temporal; CL: left central; Cz: midline central; CR: right central; TR: right temporal; TPL: left temporo-parietal; CPL: left centro-parietal; CPR: right centro-parietal; TPR: right temporo-parietal; PL: left parietal; PR: right parietal; POL: left parieto-occipital; Pz: midline parietal; POR: right parieto-occipital; O: occipital.

(TIF)

Click here for additional data file.^{(217.5KB, tif)}

S1 Table. Performance of participants on paper tests (MoCA and ACE-R) and ICA test (administered on iPad).

The absent subjects were removed from this study either due to their status of disease (AD or mild AD) or that their status changed during the development of the study.

(PDF)

Click here for additional data file.^{(59.7KB, pdf)}

Acknowledgments

We thank the National Brain Mapping Laboratory (NBML), where all the EEG data acquisitions were done. Authors are also grateful to Mohammad Mohaghar for proofreading the manuscript.

Data Availability

The data for the behavioral tests (MoCA, ACE-R, and ICA) can be found in the supplementary materials (S1 Table). The EEG dataset related to the findings in the presented manuscript is available at RepOD (https://doi.org/10.18150/DEQMGF).

Funding Statement

This work was partly supported by the Iranian Cognitive Sciences, and Technologies Council’s (COGC) grant (#4873) awarded to SKR. Cognetivity ltd. provided support in the form of salaries for authors SKR, CK, MHM. The funders did not have any additional role in the study design, data collection, analysis, decision to publish, or preparation of the manuscript.

References

1.Petersen R. C., Caracciolo B., Brayne C., Gauthier S., Jelic V., and Fratiglioni L., “Mild cognitive impairment: a concept in evolution,” J Intern Med, vol. 275, no. 3, pp. 214–228, Mar. 2014, doi: 10.1111/joim.12190 [DOI] [PMC free article] [PubMed] [Google Scholar]
2.“2018 Alzheimer’s disease facts and figures,” Alzheimer’s & Dementia: The Journal of the Alzheimer’s Association, vol. 14, no. 3, pp. 367–429, Mar. 2018, doi: 10.1016/j.jalz.2018.02.001 [DOI] [Google Scholar]
3.Ward A., Tardiff S., Dye C., and Arrighi H. M., “Rate of conversion from prodromal Alzheimer’s disease to Alzheimer’s dementia: a systematic review of the literature,” Dement Geriatr Cogn Dis Extra, vol. 3, no. 1, pp. 320–332, 2013, doi: 10.1159/000354370 [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Dunne R. A., et al. , “Mild cognitive impairment: the Manchester consensus,” Age Ageing, Nov. 2020, doi: 10.1093/ageing/afaa228 [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Vecchio F., et al. , “Human brain networks in cognitive decline: a graph theoretical analysis of cortical connectivity from EEG data,” J. Alzheimers Dis., vol. 41, no. 1, pp. 113–127, 2014, doi: 10.3233/JAD-132087 [DOI] [PubMed] [Google Scholar]
6.Babiloni C., et al. , “Brain neural synchronization and functional coupling in Alzheimer’s disease as revealed by resting state EEG rhythms,” International Journal of Psychophysiology, vol. 103, pp. 88–102, May 2016, doi: 10.1016/j.ijpsycho.2015.02.008 [DOI] [PubMed] [Google Scholar]
7.Pijnenburg Y. a. L., Made Y. v d, van Cappellen van Walsum A. M., Knol D. L., Scheltens P., and Stam C. J., “EEG synchronization likelihood in mild cognitive impairment and Alzheimer’s disease during a working memory task,” Clin Neurophysiol, vol. 115, no. 6, pp. 1332–1339, Jun. 2004, doi: 10.1016/j.clinph.2003.12.029 [DOI] [PubMed] [Google Scholar]
8.Moretti D. V., Frisoni G. B., Binetti G., and Zanetti O., “Anatomical Substrate and Scalp EEG Markers are Correlated in Subjects with Cognitive Impairment and Alzheimer’s Disease,” Front. Psychiatry, vol. 1, 2011, doi: 10.3389/fpsyt.2010.00152 [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Poil S.-S., De Haan W., van der Flier W. M., Mansvelder H. D., Scheltens P., and Linkenkaer-Hansen K., “Integrative EEG biomarkers predict progression to Alzheimer’s disease at the MCI stage,” Front. Aging Neurosci., vol. 5, 2013, doi: 10.3389/fnagi.2013.00058 [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Babiloni C., et al. , “Abnormalities of cortical neural synchronization mechanisms in patients with dementia due to Alzheimer’s and Lewy body diseases: an EEG study,” Neurobiology of Aging, vol. 55, pp. 143–158, Jul. 2017, doi: 10.1016/j.neurobiolaging.2017.03.030 [DOI] [PubMed] [Google Scholar]
11.Vecchio F., et al. , “‘Small World’ architecture in brain connectivity and hippocampal volume in Alzheimer’s disease: a study via graph theory from EEG data,” Brain Imaging and Behavior, vol. 11, no. 2, pp. 473–485, Apr. 2017, doi: 10.1007/s11682-016-9528-3 [DOI] [PubMed] [Google Scholar]
12.Cassani R., Estarellas M., San-Martin R., Fraga F. J., and Falk T. H., “Systematic Review on Resting-State EEG for Alzheimer’s Disease Diagnosis and Progression Assessment,” Dis Markers, vol. 2018, p. 5174815, 2018, doi: 10.1155/2018/5174815 [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Hedges D., Janis R., Mickelson S., Keith C., Bennett D., and Brown B. L., “P300 Amplitude in Alzheimer’s Disease: A Meta-Analysis and Meta-Regression,” Clin EEG Neurosci, vol. 47, no. 1, pp. 48–55, Jan. 2016, doi: 10.1177/1550059414550567 [DOI] [PubMed] [Google Scholar]
14.Dauwels J., et al. , “Slowing and Loss of Complexity in Alzheimer’s EEG: Two Sides of the Same Coin?,” Int J Alzheimers Dis, vol. 2011, p. 539621, Apr. 2011, doi: 10.4061/2011/539621 [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Missonnier P., et al. , “Working memory load-related electroencephalographic parameters can differentiate progressive from stable mild cognitive impairment,” Neuroscience, vol. 150, no. 2, pp. 346–356, Dec. 2007, doi: 10.1016/j.neuroscience.2007.09.009 [DOI] [PubMed] [Google Scholar]
16.Fraga F. J., Mamani G. Q., Johns E., Tavares G., Falk T. H., and Phillips N. A., “Early diagnosis of mild cognitive impairment and Alzheimer’s with event-related potentials and event-related desynchronization in N-back working memory tasks,” Comput Methods Programs Biomed, vol. 164, pp. 1–13, Oct. 2018, doi: 10.1016/j.cmpb.2018.06.011 [DOI] [PubMed] [Google Scholar]
17.Fraga F. J., Ferreira L. A., Falk T. H., Johns E., and Phillips N. D., “Event-related synchronisation responses to N-back memory tasks discriminate between healthy ageing, mild cognitive impairment, and mild Alzheimer’s disease,” in 2017. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Mar. 2017, pp. 964–968. doi: 10.1109/ICASSP.2017.7952299 [DOI] [Google Scholar]
18.Han Y., Wang K., Jia J., and Wu W., “Changes of EEG Spectra and Functional Connectivity during an Object-Location Memory Task in Alzheimer’s Disease,” Front Behav Neurosci, vol. 11, May 2017, doi: 10.3389/fnbeh.2017.00107 [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Paik J.-S., et al. , “Low vision and the risk of dementia: a nationwide population-based cohort study,” Scientific Reports, vol. 10, no. 1, Art. no. 1, Jun. 2020, doi: 10.1038/s41598-020-66002-z [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Cogan D. G., “Visuospatial dysgnosia,” Am. J. Ophthalmol., vol. 88, no. 3 Pt 1, pp. 361–368, Sep. 1979, doi: 10.1016/0002-9394(79)90634-2 [DOI] [PubMed] [Google Scholar]
21.Jackson G. R. and Owsley C., “Visual dysfunction, neurodegenerative diseases, and aging,” Neurologic Clinics, vol. 21, no. 3, pp. 709–728, Aug. 2003, doi: 10.1016/s0733-8619(02)00107-x [DOI] [PubMed] [Google Scholar]
22.Tzekov R. and Mullan M., “Vision function abnormalities in Alzheimer disease,” Surv Ophthalmol, vol. 59, no. 4, pp. 414–433, Aug. 2014, doi: 10.1016/j.survophthal.2013.10.002 [DOI] [PubMed] [Google Scholar]
23.Sadun A. A., Borchert M., DeVita E., Hinton D. R., and Bassi C. J., “Assessment of Visual Impairment in Patients With Alzheimer’s Disease,” American Journal of Ophthalmology, vol. 104, no. 2, pp. 113–120, Aug. 1987, doi: 10.1016/0002-9394(87)90001-8 [DOI] [PubMed] [Google Scholar]
24.Cogan D. G., “Visual Disturbances With Focal Progressive Dementing Disease,” American Journal of Ophthalmology, vol. 100, no. 1, pp. 68–72, Jul. 1985, doi: 10.1016/s0002-9394(14)74985-2 [DOI] [PubMed] [Google Scholar]
25.van Deursen J. A., Vuurman E. F. P. M., Verhey F. R. J., van Kranen-Mastenbroek V. H. J. M., and Riedel W. J., “Increased EEG gamma band activity in Alzheimer’s disease and mild cognitive impairment,” J Neural Transm, vol. 115, no. 9, pp. 1301–1311, Sep. 2008, doi: 10.1007/s00702-008-0083-y [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Briels C. T., Eertink J. J., Stam C. J., van der Flier W. M., Scheltens P., and Gouw A. A., “Profound regional spectral, connectivity and network changes reflect visual deficits in Posterior Cortical Atrophy: an EEG study,” Neurobiology of Aging, Aug. 2020, doi: 10.1016/j.neurobiolaging.2020.07.029 [DOI] [PubMed] [Google Scholar]
27.Kiani R., Esteky H., Mirpour K., and Tanaka K., “Object Category Structure in Response Patterns of Neuronal Population in Monkey Inferior Temporal Cortex,” Journal of Neurophysiology, vol. 97, no. 6, pp. 4296–4309, Jun. 2007, doi: 10.1152/jn.00024.2007 [DOI] [PubMed] [Google Scholar]
28.Kriegeskorte N., et al. , “Matching Categorical Object Representations in Inferior Temporal Cortex of Man and Monkey,” Neuron, vol. 60, no. 6, pp. 1126–1141, Dec. 2008, doi: 10.1016/j.neuron.2008.10.043 [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Connolly A. C., et al. , J. Neurosci., vol. 32, no. 8, pp. 2608–2618, Feb. 2012, doi: 10.1523/JNEUROSCI.5547-11.2012 [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Khaligh-Razavi S.-M. and Kriegeskorte N., “Deep Supervised, but Not Unsupervised, Models May Explain IT Cortical Representation,” PLOS Computational Biology, vol. 10, no. 11, p. e1003915, Nov. 2014, doi: 10.1371/journal.pcbi.1003915 [DOI] [PMC free article] [PubMed] [Google Scholar]
31.Khaligh-Razavi S.-M., Henriksson L., Kay K., and Kriegeskorte N., “Fixed versus mixed RSA: Explaining visual representations by fixed and mixed feature sets from shallow and deep computational models,” Journal of Mathematical Psychology, vol. 76, pp. 184–197, Feb. 2017, doi: 10.1016/j.jmp.2016.10.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
32.Naselaris T., Stansbury D. E., and Gallant J. L., “Cortical representation of animate and inanimate objects in complex natural scenes,” Journal of Physiology-Paris, vol. 106, no. 5, pp. 239–249, Sep. 2012, doi: 10.1016/j.jphysparis.2012.02.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Bacon-Macé N., Macé M. J.-M., Fabre-Thorpe M., and Thorpe S. J., “The time course of visual processing: Backward masking and natural scene categorisation,” Vision Research, vol. 45, no. 11, pp. 1459–1469, May 2005, doi: 10.1016/j.visres.2005.01.004 [DOI] [PubMed] [Google Scholar]
34.Thorpe S. J., “The Speed of Categorization in the Human Visual System,” Neuron, vol. 62, no. 2, pp. 168–170, Apr. 2009, doi: 10.1016/j.neuron.2009.04.012 [DOI] [PubMed] [Google Scholar]
35.Cichy R. M., Pantazis D., and Oliva A., “Resolving human object recognition in space and time,” Nat Neurosci, vol. 17, no. 3, pp. 455–462, Mar. 2014, doi: 10.1038/nn.3635 [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Carlson T., Tovar D. A., Alink A., and Kriegeskorte N., “Representational dynamics of object vision: The first 1000 ms,” Journal of Vision, vol. 13, no. 10, pp. 1–1, Aug. 2013, doi: 10.1167/13.10.1 [DOI] [PubMed] [Google Scholar]
37.Khaligh-Razavi S.-M., Cichy R. M., Pantazis D., and Oliva A., “Tracking the Spatiotemporal Neural Dynamics of Real-world Object Size and Animacy in the Human Brain,” Journal of Cognitive Neuroscience, vol. 30, pp. 1559–1576, 2018, doi: 10.1162/jocn_a_01290 [DOI] [PubMed] [Google Scholar]
38.Braak H. and Braak E., “Neuropathological stageing of Alzheimer-related changes,” Acta Neuropathol, vol. 82, no. 4, pp. 239–259, 1991, doi: 10.1007/BF00308809 [DOI] [PubMed] [Google Scholar]
39.Braak H. and Braak E., “Frequency of stages of Alzheimer-related lesions in different age categories,” Neurobiol Aging, vol. 18, no. 4, pp. 351–357, Aug. 1997, doi: 10.1016/s0197-4580(97)00056-0 [DOI] [PubMed] [Google Scholar]
40.Delacourte A., et al. , “The biochemical pathway of neurofibrillary degeneration in aging and Alzheimer’s disease,” Neurology, vol. 52, no. 6, pp. 1158–1165, Apr. 1999, doi: 10.1212/wnl.52.6.1158 [DOI] [PubMed] [Google Scholar]
41.Braak H., Alafuzoff I., Arzberger T., Kretzschmar H., and Del Tredici K., “Staging of Alzheimer disease-associated neurofibrillary pathology using paraffin sections and immunocytochemistry,” Acta Neuropathol, vol. 112, no. 4, pp. 389–404, 2006, doi: 10.1007/s00401-006-0127-z [DOI] [PMC free article] [PubMed] [Google Scholar]
42.Johnson K. A., et al. , “Tau positron emission tomographic imaging in aging and early Alzheimer disease,” Ann Neurol, vol. 79, no. 1, pp. 110–119, Jan. 2016, doi: 10.1002/ana.24546 [DOI] [PMC free article] [PubMed] [Google Scholar]
43.Villemagne V. L., Okamura N., and Rowe C. C., “Untangling tau imaging,” Alzheimer’s & Dementia: Diagnosis, Assessment & Disease Monitoring, vol. 4, no. 1, pp. 39–42, 2016, doi: 10.1016/j.dadm.2016.05.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
44.Khaligh-Razavi S.-M., et al. , “Integrated Cognitive Assessment: Speed and Accuracy of Visual Processing as a Reliable Proxy to Cognitive Performance,” Sci Rep, vol. 9, Jan. 2019, doi: 10.1038/s41598-018-37709-x [DOI] [PMC free article] [PubMed] [Google Scholar]
45.Kalafatis C. and Khaligh Razavi S.-M., “F3-01-03: the integrated cognitive assessment—employing artificial intelligence for the detection of cognitive impairment,” Alzheimer’s & Dementia, vol. 15, no. 7S_Part_16, pp. P864–P864, Jul. 2019, doi: 10.1016/j.jalz.2019.06.4598 [DOI] [Google Scholar]
46.Kalafatis C., et al. , “P4‐207: EMPLOYING ARTIFICIAL INTELLIGENCE IN THE DEVELOPMENT OF A SELF‐ADMINISTERED, COMPUTERISED COGNITIVE ASSESSMENT FOR THE ASSESSMENT OF NEURODEGENERATION,” Alzheimer’s & Dementia, vol. 15, pp. P1355–P1356, Jul. 2019, doi: 10.1016/j.jalz.2019.06.3870 [DOI] [Google Scholar]
47.Ruiz-Rizzo A. L., et al. , “Simultaneous object perception deficits are related to reduced visual processing speed in amnestic mild cognitive impairment,” Neurobiology of Aging, vol. 55, pp. 132–142, Jul. 2017, doi: 10.1016/j.neurobiolaging.2017.03.029 [DOI] [PubMed] [Google Scholar]
48.Costa S. L., Genova H. M., DeLuca J., and Chiaravalloti N. D., “Information processing speed in multiple sclerosis: Past, present, and future,” Mult Scler, vol. 23, no. 6, pp. 772–789, May 2017, doi: 10.1177/1352458516645869 [DOI] [PubMed] [Google Scholar]
49.DeLuca J., Chelune G. J., Tulsky D. S., Lengenfelder J., and Chiaravalloti N. D., “Is Speed of Processing or Working Memory the Primary Information Processing Deficit in Multiple Sclerosis?,” Journal of Clinical and Experimental Neuropsychology, vol. 26, no. 4, pp. 550–562, Jun. 2004, doi: 10.1080/13803390490496641 [DOI] [PubMed] [Google Scholar]
50.Khaligh-Razavi S. M. and Habibi S., “System for assessing mental health disorder,” UK Intellect. Prop. Off, 2013. [Google Scholar]
51.Khaligh-Razavi S.-M., Sadeghi M., Khanbagi M., Kalafatis C., and Nabavi S. M., “A self-administered, artificial intelligence (AI) platform for cognitive assessment in multiple sclerosis (MS),” BMC Neurology, vol. 20, no. 1, p. 193, May 2020, doi: 10.1186/s12883-020-01736-x [DOI] [PMC free article] [PubMed] [Google Scholar]
52.Nasreddine Z. S., et al. , “The Montreal Cognitive Assessment, MoCA: a brief screening tool for mild cognitive impairment,” J Am Geriatr Soc, vol. 53, no. 4, pp. 695–699, Apr. 2005, doi: 10.1111/j.1532-5415.2005.53221.x [DOI] [PubMed] [Google Scholar]
53.Mioshi E., Dawson K., Mitchell J., Arnold R., and Hodges J. R., “The Addenbrooke’s Cognitive Examination Revised (ACE-R): a brief cognitive test battery for dementia screening,” Int J Geriatr Psychiatry, vol. 21, no. 11, pp. 1078–1085, Nov. 2006, doi: 10.1002/gps.1610 [DOI] [PubMed] [Google Scholar]
54.Tadel F., Baillet S., Mosher J. C., Pantazis D., and Leahy R. M., “Brainstorm: A User-Friendly Application for MEG/EEG Analysis,” Computational Intelligence and Neuroscience, 2011. https://www.hindawi.com/journals/cin/2011/879716/ (accessed Jul. 28, 2019). doi: 10.1155/2011/879716 [DOI] [PMC free article] [PubMed] [Google Scholar]
55.Chang C.-C. and Lin C.-J., “LIBSVM: A library for support vector machines,” ACM Trans. Intell. Syst. Technol., vol. 2, no. 3, p. 27:1–27:27, May 2011, doi: 10.1145/1961189.1961199 [DOI] [Google Scholar]
56.Nonparametric Statistical Inference. Boca Raton, 2010. [Google Scholar]
57.Kriegeskorte N. and Kievit R. A., “Representational geometry: integrating cognition, computation, and the brain,” Trends Cogn Sci, vol. 17, no. 8, pp. 401–412, Aug. 2013, doi: 10.1016/j.tics.2013.06.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
58.Missonnier P., et al. , “Early event-related potential changes during working memory activation predict rapid decline in mild cognitive impairment,” J. Gerontol. A Biol. Sci. Med. Sci., vol. 60, no. 5, pp. 660–666, May 2005, doi: 10.1093/gerona/60.5.660 [DOI] [PubMed] [Google Scholar]
59.Babiloni C., et al. , “Sources of cortical rhythms change as a function of cognitive impairment in pathological aging: a multicenter study,” Clinical Neurophysiology: Official Journal of the International Federation of Clinical Neurophysiology, vol. 117, no. 2, pp. 252–268, Feb. 2006, doi: 10.1016/j.clinph.2005.09.019 [DOI] [PubMed] [Google Scholar]
60.Choi J., et al. , “Resting-state prefrontal EEG biomarkers in correlation with MMSE scores in elderly individuals,” Scientific Reports, vol. 9, no. 1, Art. no. 1, Jul. 2019, doi: 10.1038/s41598-019-46789-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
61.Culham J. C. and Valyear K. F., “Human parietal cortex in action,” Curr Opin Neurobiol, vol. 16, no. 2, pp. 205–212, Apr. 2006, doi: 10.1016/j.conb.2006.03.005 [DOI] [PubMed] [Google Scholar]
62.Fogassi L. and Luppino G., “Motor functions of the parietal lobe,” Curr Opin Neurobiol, vol. 15, no. 6, pp. 626–631, Dec. 2005, doi: 10.1016/j.conb.2005.10.015 [DOI] [PubMed] [Google Scholar]
63.Fogassi L., Ferrari P. F., Gesierich B., Rozzi S., Chersi F., and Rizzolatti G., “Parietal lobe: from action organization to intention understanding,” Science, vol. 308, no. 5722, pp. 662–667, Apr. 2005, doi: 10.1126/science.1106138 [DOI] [PubMed] [Google Scholar]
64.Davis S. W., Dennis N. A., Daselaar S. M., Fleck M. S., and Cabeza R., “Que PASA? The posterior-anterior shift in aging,” Cereb Cortex, vol. 18, no. 5, pp. 1201–1209, May 2008, doi: 10.1093/cercor/bhm155 [DOI] [PMC free article] [PubMed] [Google Scholar]

PLoS One. doi: 10.1371/journal.pone.0264058.r001

Decision Letter 0

Stavros I Dimitriadis

6 May 2021

PONE-D-21-04261

Temporal dynamics of animacy categorization in the brain of patients with mild cognitive impairment

PLOS ONE

Dear Dr. Karimi,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

I have received the comments from two reviewers. One of the two reviewers raised an issue with the small sample size of patients group.

In my opinion, you can estimate the power of your analysis and also you can add Cohen's d statistic while

you can avoid strong use of significance.

A paragraph with limitation should be added in the discussion part.

If you decided to revise according to reviewers' comments, you should update the manuscript and prepare

a response letter to every comment.

Please submit your revised manuscript by Jun 19 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Stavros I. Dimitriadis

Academic Editor

PLOS ONE

Additional Editor Comments:

Dear authors,

I have received the comments from two reviewers. One of the two reviewers raised an issue with the small sample

size of patients group.

In my opinion, you can estimate the power of your analysis and also you can add Cohen's d statistic while

you can avoid strong use of significance.

A paragraph with limitation should be added in the discussion part.

If you decided to revise according to reviewers' comments, you should revise the manuscript and prepare

a response letter to every comment.

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Thank you for stating the following in the Competing Interests section:

"SKR and CK serve as the CSO and CMO at Cognetivity ltd., and MM as the lead data scientist. Other authors declared no competing interests."

We note that one or more of the authors are employed by a commercial company: Cognetivity ltd.

2.1. Please provide an amended Funding Statement declaring this commercial affiliation, as well as a statement regarding the Role of Funders in your study. If the funding organization did not play a role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript and only provided financial support in the form of authors' salaries and/or research materials, please review your statements relating to the author contributions, and ensure you have specifically and accurately indicated the role(s) that these authors had in your study. You can update author roles in the Author Contributions section of the online submission form.

Please also include the following statement within your amended Funding Statement.

“The funder provided support in the form of salaries for authors [insert relevant initials], but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The specific roles of these authors are articulated in the ‘author contributions’ section.”

If your commercial affiliation did play a role in your study, please state and explain this role within your updated Funding Statement.

2.2. Please also provide an updated Competing Interests Statement declaring this commercial affiliation along with any other relevant declarations relating to employment, consultancy, patents, products in development, or marketed products, etc.

Within your Competing Interests Statement, please confirm that this commercial affiliation does not alter your adherence to all PLOS ONE policies on sharing data and materials by including the following statement: "This does not alter our adherence to PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors http://journals.plos.org/plosone/s/competing-interests) . If this adherence statement is not accurate and there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared.

Please include both an updated Funding Statement and Competing Interests Statement in your cover letter. We will change the online submission form on your behalf.

Please know it is PLOS ONE policy for corresponding authors to declare, on behalf of all authors, all potential competing interests for the purposes of transparency. PLOS defines a competing interest as anything that interferes with, or could reasonably be perceived as interfering with, the full and objective presentation, peer review, editorial decision-making, or publication of research or non-research articles submitted to one of the journals. Competing interests can be financial or non-financial, professional, or personal. Competing interests can arise in relationship to an organization or another person. Please follow this link to our website for more details on competing interests: http://journals.plos.org/plosone/s/competing-interests

3. We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For more information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially sensitive information, data are owned by a third-party organization, etc.) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: I Don't Know

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Overall:

The manuscript analyzes the ERP response during the integrated cognitive assessment to find differences between healthy controls and patients with mild cognitive impairment.

Results suggest a slower response in the MCI population.

Recommend:

Major comments:

* In Section 1, relevant works on the use of EEG for AD and MCI assessment are missing, e.g.:

- Regarding resting state EEG: "Systematic Review on Resting-State EEG for Alzheimer’s Disease Diagnosis and Progression Assessment Resting-state EEG" (Cassani, 2018), and

- Regarding ERPs: "P300 Amplitude in Alzheimer’s Disease: A Meta-Analysis and Meta-Regression" (Hedges, 2016)

- Regarding slower EEG signals: "Slowing and Loss of Complexity in Alzheimer's EEG: Two Sides of the Same Coin?" (Dauwels, 2011)

* The MoCA and ACE-R tests are described in Sections 2.2 and 2.3 respectively. In Section 2.4, it is stated the participants completed these tests, however the scores are not presented nor used.

* In Section 2.5, describe the electrode layout that was used, as well as the location of the reference electrode.

* In Section 2.7, describe the kernel used in the SVM classifier, as well as the hyperparameters.

* Provide a Conclusion section at the end of the manuscript

* Discuss in how the findings about the ERP amplitude are in line with the results provided in (Hedges, 2016)

Minor comments:

* Pay attention to the capitalization of words, e.g., "Congestive Cardiac Failure" and "Diabetes Mellitus" should not be capitalized.

* Remove extra space in "80 ±20" and all the instances in the manuscript.

* Use always space between quantity and units (e.g., "66 ms" instead of "66ms")

Reviewer #2: The manuscript presents a novel study investigating animacy categorization in MCI patients vs healthy older adults using univariate and multivariate pattern analysis. Results demonstrated decreased speed of processing in MCI patients and differences in activation patterns across a range of electrodes. While the study is well presented, I have a number of concerns around reliability and generalizability of the results, which prevent me from recommending the manuscript from publication.

The statistical analyses appear to have been performed in a technically sound way. However, the small sample size significantly limits the reliability and generalizability of the classification findings. Several studies have highlighted the dangers of small sample sizes when using brain data for classification (e.g. Hosseini et al., 2020). Based on this limitation, it is difficult to interpret the findings (or their significance).

MOCA and ACE-R results should be presented in the manuscript.

The authors should include information on the number of EEG trials rejected at each stage of processing across each condition and for each group. Without this information, it is difficult to interpret the reliability of findings.

The manuscript is presented in an intelligible fashion. However, the Introduction would benefit from more detailed description of previous literature on visual processing in MCI and why this is important to explore (beyond the fact that differences between MCI and HCs have yet to be investigated). Typically, the final paragraph of the Introduction would state the study hypotheses – not the main findings. I suggest the authors revise this accordingly.

Similarly, the Discussion requires a more detailed interpretation of the study findings, especially with regard to the overall pattern of differences found, e.g. why were there group differences in midline frontal, left frontal, left parietal and right centro-parietal electrodes? How does this relate to previous findings?

In my opinion, the authors’ conclusion that the results “suggest the IPS as a potential biomarker for the early detection of AD” is overstated. As mentioned in the Introduction, not all cases of MCI will progress to AD. This fact, coupled with the small patient sample (n=18), significantly limit generalizability to AD.

The authors state that the ICA task is “more challenging than the typical memory tasks, making it more sensitive to less severe brain deteriorations”. This is also at odds with the suggestion of IPS as a potential biomarker for AD, as the task may not be suitable for AD patients. The authors should comment on this in their Discussion.

As far as I can tell, the data underlying the findings have not been made available.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

PLoS One. 2022 Feb 23;17(2):e0264058. doi: 10.1371/journal.pone.0264058.r002

Author response to Decision Letter 0

3 Aug 2021

We would like to thank the Editor for pointing out constructive suggestions. Following these suggestions, based on a power of 0.8, we estimated the sample size to be 22 (for details, please see our response to comment 1 of Reviewer #2). We have also added a paragraph to the Discussion explaining the study limitations (page 18, line 35 to page 19, line 2).

We would also like to thank the reviewers for pointing out thoughtful comments and great suggestions. We’ve accommodated the comments, and below you can see point-by-point responses to these comments in blue.

Reviewer #1: Overall:

The manuscript analyzes the ERP response during the integrated cognitive assessment to find differences between healthy controls and patients with mild cognitive impairment.

Results suggest a slower response in the MCI population.

Recommend:

Major comments:

1. In Section 1, relevant works on the use of EEG for AD and MCI assessment are missing, e.g.:

- Regarding resting state EEG: "Systematic Review on Resting-State EEG for Alzheimer’s Disease Diagnosis and Progression Assessment Resting-state EEG" (Cassani, 2018), and

- Regarding ERPs: "P300 Amplitude in Alzheimer’s Disease: A Meta-Analysis and Meta-Regression" (Hedges, 2016)

- Regarding slower EEG signals: "Slowing and Loss of Complexity in Alzheimer's EEG: Two Sides of the Same Coin?" (Dauwels, 2011)

We have added all the above literature to our introduction (page 2, line 9).

2. The MoCA and ACE-R tests are described in Sections 2.2 and 2.3 respectively. In Section 2.4, it is stated the participants completed these tests, however the scores are not presented nor used.

The test results were employed by the consultant neurologist to help with the diagnosis. We added this explanation to the Methods (page 4, line 7, and page 4, line 13). Additionally, the results of both MoCA and ACE are now included in the supplementary materials (Table S1).

3. In Section 2.5, describe the electrode layout that was used, as well as the location of the reference electrode.

This is now added to the supplementary materials (Figure S1).

4. In Section 2.7, describe the kernel used in the SVM classifier, as well as the hyperparameters.

We added the following subsection to the Methods (section 2.7; page 7 lines 6 to 10), explaining the details of the SVM classifier:

“We used a linear support vector machine (SVM) classifier throughout the analyses to decode neural data. The hyperparameters were as follows: The cost/regularization parameter (C) and the weight of classes were all set to 1. All the classifications were done using the LIBSVM software implementation.“

5. Provide a Conclusion section at the end of the manuscript

We added a section for the conclusion (page 19, Line 4 to 10).

6. Discuss in how the findings about the ERP amplitude are in line with the results provided in (Hedges, 2016)

We added the following to the discussion (page 17, lines 7 to 12), comparing the findings of the current study with that of Hedges, 2016:

“Having a closer look at the ERP of different regions, we observed a decrease in the P300 amplitude of the patients with MCI. This decrease is significant in the electrodes of the midline frontal, right fronto-central, and right temporal regions between 250 ms to 500 ms after the stimulus onset. This observation is in line with the reported result in (Hedges, 2016), especially with the P300 of the midline frontal electrodes, and highlights the importance of this signal in the detection of MCI.”

Minor comments:

* Pay attention to the capitalization of words, e.g., "Congestive Cardiac Failure" and "Diabetes Mellitus" should not be capitalized.

Done

* Remove extra space in "80 ±20" and all the instances in the manuscript.

Done

* Use always space between quantity and units (e.g., "66 ms" instead of "66ms")

Done

1. The statistical analyses appear to have been performed in a technically sound way. However, the small sample size significantly limits the reliability and generalizability of the classification findings. Several studies have highlighted the dangers of small sample sizes when using brain data for classification (e.g. Hosseini et al., 2020). Based on this limitation, it is difficult to interpret the findings (or their significance).

We understand that the number of participants might raise a concern. However, these findings are supported by our previous behavioral results from MCI patients, which indicate a significant decrease in the speed and accuracy of categorizing animacy information (SM Khaligh-Razavi, 2019, Scientific Report , with 448 participants, and MH Modarres, 2021, bioRxiv (accepted at Front Psychiatry) , with 230 participants). The current study was designed for taking a closer look at the underlying neural response of MCI patients and healthy individuals while doing a task that already showed their difference in speed and accuracy at the behavioral level.

Furthermore, the number of individuals who participated in our study is comparable to the recent related EEG publications on MCI/AD. For reference here is a list of such publications:

1. Zhao et al., 2020, Effects of creative expression program on the event-related potential and task reaction time of elderly with mild cognitive impairment, International journal of nursing sciences

Control MCI: 18, MCI: 18

2. Massa et al., 2020, Utility of quantitative EEG in early Lewy body disease, Parkinsonism & related disorders

HC: 24, MCI-LBD: 12, MCI-AD: 11

3. Briels et al., 2020, Profound regional spectral, connectivity, and network changes reflect visual deficits in posterior cortical atrophy: an EEG study, Neurobiology of Aging

HC: 29, PCA (Posterior Cortical Atrophy)-AD: 29

4. Sharma et al., 2019, EEG and cognitive biomarkers based mild cognitive impairment diagnosis, IRBM

HC: 13, MCI: 16, dementia: 15

5. Fraga et al. 2018, Early diagnosis of mild cognitive impairment and Alzheimer’s with event-related potentials and event-related desynchronization in N-back working memory tasks, Computer methods and programs in biomedicine

HC: 27, MCI: 21, AD: 15

6. Simons et al., 2018, Fuzzy Entropy Analysis of the Electroencephalogram in Patients with Alzheimer’s Disease: Is the Method Superior to Sample Entropy?, Entropy

HC: 11, AD: 11

7. Azami et al., 2017, Univariate and Multivariate Generalized Multiscale Entropy to Characterise EEG Signals in Alzheimer’s Disease, Entropy

HC: 11, AD: 11

8. Han et al., 2017, Changes of EEG Spectra and Functional Connectivity during an Object-Location Memory Task in Alzheimer’s Disease, Frontiers in behavioral neuroscience

HC: 19, mild AD: 20

9. Fraga et al. 2017, Event-related synchronisation responses to N-back memory tasks discriminate between healthy ageing, mild cognitive impairment, and mild Alzheimer's disease, IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

HC: 27, MCI: 21

Following the editor’s suggestion, we also estimated the sample size needed for the desired power of 0.8 (when comparing the decoding peaks of HC vs. MCI). For μ1 = 434 (median of HC peak time points), μ2 = 473 (median of MCI peak time points), σ = 32.28 (pooled standard deviation of HC and MCI peak time points), α = 0.001 (type I error rate), the sample size of each group was estimated as 22, which is also consistent with the mentioned studies.

2. MOCA and ACE-R results should be presented in the manuscript.

We added these data as well as the data for the ICA test to the Supplementary materials (Table S1 of supplementary materials).

3. The authors should include information on the number of EEG trials rejected at each stage of processing across each condition and for each group. Without this information, it is difficult to interpret the reliability of findings.

We included resting trials in between the image trials (i.e., almost every 70 seconds, they were given 10 seconds to rest their eyes, blink, and swallow). Participants were instructed to only blink (or swallow) during these trials to prevent contamination of EEG signals with the eye-blink (and swallowing) artifacts. These trials were excluded from further EEG analyses. Because of such a design, we did not have to reject any of the image trials. Other potential artifacts were removed with Independent Component Analysis. We added this description to the methods (section 2.5, page 6, lines 18 to 24).

4. The manuscript is presented in an intelligible fashion. However, the Introduction would benefit from more detailed description of previous literature on visual processing in MCI and why this is important to explore (beyond the fact that differences between MCI and HCs have yet to be investigated). Typically, the final paragraph of the Introduction would state the study hypotheses – not the main findings. I suggest the authors revise this accordingly.

We updated the introduction to include more information about the importance of the visual system and the ICA task with regard to cognitive impairment in MCI/AD. (See page 2, line 27-29 and line 33 of page 2 to line 7 of page 3).

There are various styles in scientific writing. We favor that of Mensh & Kording, Plos Comp (2017) ; this suggests the inclusion of the main findings’ summary. Following the reviewer’s suggestion, we also added a few sentences on study hypotheses.

5. Similarly, the Discussion requires a more detailed interpretation of the study findings, especially with regard to the overall pattern of differences found, e.g. why were there group differences in midline frontal, left frontal, left parietal and right centro-parietal electrodes? How does this relate to previous findings?

We added the sub-section (4.2) to the Discussion on how these changes relate to previous findings (page 18, line 1 to line 16).

6. In my opinion, the authors’ conclusion that the results “suggest the IPS as a potential biomarker for the early detection of AD” is overstated. As mentioned in the Introduction, not all cases of MCI will progress to AD. This fact, coupled with the small patient sample (n=18), significantly limit generalizability to AD.

We have modified the discussion to mention the study limitations (page 18 line 35 to page 19, line 2).

It is worth noting (as also added to the discussion; page 18 lines 32 to 34) that the purpose of this study was to detect mild cognitive impairment. We do not necessarily claim that all MCI cases are due to AD or will convert to AD. There is clinical value in detecting MCI patients via large-scale population screening using digital biomarkers, such as ICA. The next step is then to use more sophisticated tools, such as PET, CSF, etc to identify those who are more likely to develop AD or have an amyloid burden.

7. The authors state that the ICA task is “more challenging than the typical memory tasks, making it more sensitive to less severe brain deteriorations”. This is also at odds with the suggestion of IPS as a potential biomarker for AD, as the task may not be suitable for AD patients. The authors should comment on this in their Discussion.

We added this to the discussion (page 16 lines 28 to 33):

The ICA is particularly suited for population-wide screening of cognitive impairment to help identify patients with MCI and mild AD (MiAD) and not designed for cognitive assessment in more severe stages of the disease, such as moderate to severe AD. Early detection of cognitive impairment is becoming increasingly important, particularly in the context of disease-modifying therapies for early stages of AD, such as Aducanumab –which has recently received FDA approval.

8. As far as I can tell, the data underlying the findings have not been made available.

Attachment

Submitted filename: Response to Reviewer.pdf

Click here for additional data file.^{(311.4KB, pdf)}

PLoS One. doi: 10.1371/journal.pone.0264058.r003

Decision Letter 1

Stavros I Dimitriadis

3 Feb 2022

Temporal dynamics of animacy categorization in the brain of patients with mild cognitive impairment

PONE-D-21-04261R1

Dear Dr. Karimi,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Stavros I. Dimitriadis

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

The authors responded properly to the reviewers' comments.

I have reviewed personally the draft and I have no further comments to add.

I recommend the acceptance of the manuscript.

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

Reviewer #2: Yes

**********

6. Review Comments to the Author

Reviewer #2: All comments have been addressed. I have no further suggestions for revisions.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #2: No

PLoS One. doi: 10.1371/journal.pone.0264058.r004

Acceptance letter

Stavros I Dimitriadis

9 Feb 2022

PONE-D-21-04261R1

Temporal dynamics of animacy categorization in the brain of patients with mild cognitive impairment

Dear Dr. Karimi:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Stavros I. Dimitriadis

Academic Editor

PLOS ONE

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

S1 Fig. EEG electrodes layout.

(TIF)

Click here for additional data file.^{(217.5KB, tif)}

S1 Table. Performance of participants on paper tests (MoCA and ACE-R) and ICA test (administered on iPad).

The absent subjects were removed from this study either due to their status of disease (AD or mild AD) or that their status changed during the development of the study.

(PDF)

Click here for additional data file.^{(59.7KB, pdf)}

Attachment

Submitted filename: Response to Reviewer.pdf

Click here for additional data file.^{(311.4KB, pdf)}

Data Availability Statement

[pone.0264058.ref001] 1.Petersen R. C., Caracciolo B., Brayne C., Gauthier S., Jelic V., and Fratiglioni L., “Mild cognitive impairment: a concept in evolution,” J Intern Med, vol. 275, no. 3, pp. 214–228, Mar. 2014, doi: 10.1111/joim.12190 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0264058.ref002] 2.“2018 Alzheimer’s disease facts and figures,” Alzheimer’s & Dementia: The Journal of the Alzheimer’s Association, vol. 14, no. 3, pp. 367–429, Mar. 2018, doi: 10.1016/j.jalz.2018.02.001 [DOI] [Google Scholar]

[pone.0264058.ref003] 3.Ward A., Tardiff S., Dye C., and Arrighi H. M., “Rate of conversion from prodromal Alzheimer’s disease to Alzheimer’s dementia: a systematic review of the literature,” Dement Geriatr Cogn Dis Extra, vol. 3, no. 1, pp. 320–332, 2013, doi: 10.1159/000354370 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0264058.ref004] 4.Dunne R. A., et al. , “Mild cognitive impairment: the Manchester consensus,” Age Ageing, Nov. 2020, doi: 10.1093/ageing/afaa228 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0264058.ref005] 5.Vecchio F., et al. , “Human brain networks in cognitive decline: a graph theoretical analysis of cortical connectivity from EEG data,” J. Alzheimers Dis., vol. 41, no. 1, pp. 113–127, 2014, doi: 10.3233/JAD-132087 [DOI] [PubMed] [Google Scholar]

[pone.0264058.ref006] 6.Babiloni C., et al. , “Brain neural synchronization and functional coupling in Alzheimer’s disease as revealed by resting state EEG rhythms,” International Journal of Psychophysiology, vol. 103, pp. 88–102, May 2016, doi: 10.1016/j.ijpsycho.2015.02.008 [DOI] [PubMed] [Google Scholar]

[pone.0264058.ref007] 7.Pijnenburg Y. a. L., Made Y. v d, van Cappellen van Walsum A. M., Knol D. L., Scheltens P., and Stam C. J., “EEG synchronization likelihood in mild cognitive impairment and Alzheimer’s disease during a working memory task,” Clin Neurophysiol, vol. 115, no. 6, pp. 1332–1339, Jun. 2004, doi: 10.1016/j.clinph.2003.12.029 [DOI] [PubMed] [Google Scholar]

[pone.0264058.ref008] 8.Moretti D. V., Frisoni G. B., Binetti G., and Zanetti O., “Anatomical Substrate and Scalp EEG Markers are Correlated in Subjects with Cognitive Impairment and Alzheimer’s Disease,” Front. Psychiatry, vol. 1, 2011, doi: 10.3389/fpsyt.2010.00152 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0264058.ref009] 9.Poil S.-S., De Haan W., van der Flier W. M., Mansvelder H. D., Scheltens P., and Linkenkaer-Hansen K., “Integrative EEG biomarkers predict progression to Alzheimer’s disease at the MCI stage,” Front. Aging Neurosci., vol. 5, 2013, doi: 10.3389/fnagi.2013.00058 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0264058.ref010] 10.Babiloni C., et al. , “Abnormalities of cortical neural synchronization mechanisms in patients with dementia due to Alzheimer’s and Lewy body diseases: an EEG study,” Neurobiology of Aging, vol. 55, pp. 143–158, Jul. 2017, doi: 10.1016/j.neurobiolaging.2017.03.030 [DOI] [PubMed] [Google Scholar]

[pone.0264058.ref011] 11.Vecchio F., et al. , “‘Small World’ architecture in brain connectivity and hippocampal volume in Alzheimer’s disease: a study via graph theory from EEG data,” Brain Imaging and Behavior, vol. 11, no. 2, pp. 473–485, Apr. 2017, doi: 10.1007/s11682-016-9528-3 [DOI] [PubMed] [Google Scholar]

[pone.0264058.ref012] 12.Cassani R., Estarellas M., San-Martin R., Fraga F. J., and Falk T. H., “Systematic Review on Resting-State EEG for Alzheimer’s Disease Diagnosis and Progression Assessment,” Dis Markers, vol. 2018, p. 5174815, 2018, doi: 10.1155/2018/5174815 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0264058.ref013] 13.Hedges D., Janis R., Mickelson S., Keith C., Bennett D., and Brown B. L., “P300 Amplitude in Alzheimer’s Disease: A Meta-Analysis and Meta-Regression,” Clin EEG Neurosci, vol. 47, no. 1, pp. 48–55, Jan. 2016, doi: 10.1177/1550059414550567 [DOI] [PubMed] [Google Scholar]

[pone.0264058.ref014] 14.Dauwels J., et al. , “Slowing and Loss of Complexity in Alzheimer’s EEG: Two Sides of the Same Coin?,” Int J Alzheimers Dis, vol. 2011, p. 539621, Apr. 2011, doi: 10.4061/2011/539621 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0264058.ref015] 15.Missonnier P., et al. , “Working memory load-related electroencephalographic parameters can differentiate progressive from stable mild cognitive impairment,” Neuroscience, vol. 150, no. 2, pp. 346–356, Dec. 2007, doi: 10.1016/j.neuroscience.2007.09.009 [DOI] [PubMed] [Google Scholar]

[pone.0264058.ref016] 16.Fraga F. J., Mamani G. Q., Johns E., Tavares G., Falk T. H., and Phillips N. A., “Early diagnosis of mild cognitive impairment and Alzheimer’s with event-related potentials and event-related desynchronization in N-back working memory tasks,” Comput Methods Programs Biomed, vol. 164, pp. 1–13, Oct. 2018, doi: 10.1016/j.cmpb.2018.06.011 [DOI] [PubMed] [Google Scholar]

[pone.0264058.ref017] 17.Fraga F. J., Ferreira L. A., Falk T. H., Johns E., and Phillips N. D., “Event-related synchronisation responses to N-back memory tasks discriminate between healthy ageing, mild cognitive impairment, and mild Alzheimer’s disease,” in 2017. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Mar. 2017, pp. 964–968. doi: 10.1109/ICASSP.2017.7952299 [DOI] [Google Scholar]

[pone.0264058.ref018] 18.Han Y., Wang K., Jia J., and Wu W., “Changes of EEG Spectra and Functional Connectivity during an Object-Location Memory Task in Alzheimer’s Disease,” Front Behav Neurosci, vol. 11, May 2017, doi: 10.3389/fnbeh.2017.00107 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0264058.ref019] 19.Paik J.-S., et al. , “Low vision and the risk of dementia: a nationwide population-based cohort study,” Scientific Reports, vol. 10, no. 1, Art. no. 1, Jun. 2020, doi: 10.1038/s41598-020-66002-z [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0264058.ref020] 20.Cogan D. G., “Visuospatial dysgnosia,” Am. J. Ophthalmol., vol. 88, no. 3 Pt 1, pp. 361–368, Sep. 1979, doi: 10.1016/0002-9394(79)90634-2 [DOI] [PubMed] [Google Scholar]

[pone.0264058.ref021] 21.Jackson G. R. and Owsley C., “Visual dysfunction, neurodegenerative diseases, and aging,” Neurologic Clinics, vol. 21, no. 3, pp. 709–728, Aug. 2003, doi: 10.1016/s0733-8619(02)00107-x [DOI] [PubMed] [Google Scholar]

[pone.0264058.ref022] 22.Tzekov R. and Mullan M., “Vision function abnormalities in Alzheimer disease,” Surv Ophthalmol, vol. 59, no. 4, pp. 414–433, Aug. 2014, doi: 10.1016/j.survophthal.2013.10.002 [DOI] [PubMed] [Google Scholar]

[pone.0264058.ref023] 23.Sadun A. A., Borchert M., DeVita E., Hinton D. R., and Bassi C. J., “Assessment of Visual Impairment in Patients With Alzheimer’s Disease,” American Journal of Ophthalmology, vol. 104, no. 2, pp. 113–120, Aug. 1987, doi: 10.1016/0002-9394(87)90001-8 [DOI] [PubMed] [Google Scholar]

[pone.0264058.ref024] 24.Cogan D. G., “Visual Disturbances With Focal Progressive Dementing Disease,” American Journal of Ophthalmology, vol. 100, no. 1, pp. 68–72, Jul. 1985, doi: 10.1016/s0002-9394(14)74985-2 [DOI] [PubMed] [Google Scholar]

[pone.0264058.ref025] 25.van Deursen J. A., Vuurman E. F. P. M., Verhey F. R. J., van Kranen-Mastenbroek V. H. J. M., and Riedel W. J., “Increased EEG gamma band activity in Alzheimer’s disease and mild cognitive impairment,” J Neural Transm, vol. 115, no. 9, pp. 1301–1311, Sep. 2008, doi: 10.1007/s00702-008-0083-y [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0264058.ref026] 26.Briels C. T., Eertink J. J., Stam C. J., van der Flier W. M., Scheltens P., and Gouw A. A., “Profound regional spectral, connectivity and network changes reflect visual deficits in Posterior Cortical Atrophy: an EEG study,” Neurobiology of Aging, Aug. 2020, doi: 10.1016/j.neurobiolaging.2020.07.029 [DOI] [PubMed] [Google Scholar]

[pone.0264058.ref027] 27.Kiani R., Esteky H., Mirpour K., and Tanaka K., “Object Category Structure in Response Patterns of Neuronal Population in Monkey Inferior Temporal Cortex,” Journal of Neurophysiology, vol. 97, no. 6, pp. 4296–4309, Jun. 2007, doi: 10.1152/jn.00024.2007 [DOI] [PubMed] [Google Scholar]

[pone.0264058.ref028] 28.Kriegeskorte N., et al. , “Matching Categorical Object Representations in Inferior Temporal Cortex of Man and Monkey,” Neuron, vol. 60, no. 6, pp. 1126–1141, Dec. 2008, doi: 10.1016/j.neuron.2008.10.043 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0264058.ref029] 29.Connolly A. C., et al. , J. Neurosci., vol. 32, no. 8, pp. 2608–2618, Feb. 2012, doi: 10.1523/JNEUROSCI.5547-11.2012 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0264058.ref030] 30.Khaligh-Razavi S.-M. and Kriegeskorte N., “Deep Supervised, but Not Unsupervised, Models May Explain IT Cortical Representation,” PLOS Computational Biology, vol. 10, no. 11, p. e1003915, Nov. 2014, doi: 10.1371/journal.pcbi.1003915 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0264058.ref031] 31.Khaligh-Razavi S.-M., Henriksson L., Kay K., and Kriegeskorte N., “Fixed versus mixed RSA: Explaining visual representations by fixed and mixed feature sets from shallow and deep computational models,” Journal of Mathematical Psychology, vol. 76, pp. 184–197, Feb. 2017, doi: 10.1016/j.jmp.2016.10.007 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0264058.ref032] 32.Naselaris T., Stansbury D. E., and Gallant J. L., “Cortical representation of animate and inanimate objects in complex natural scenes,” Journal of Physiology-Paris, vol. 106, no. 5, pp. 239–249, Sep. 2012, doi: 10.1016/j.jphysparis.2012.02.001 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0264058.ref033] 33.Bacon-Macé N., Macé M. J.-M., Fabre-Thorpe M., and Thorpe S. J., “The time course of visual processing: Backward masking and natural scene categorisation,” Vision Research, vol. 45, no. 11, pp. 1459–1469, May 2005, doi: 10.1016/j.visres.2005.01.004 [DOI] [PubMed] [Google Scholar]

[pone.0264058.ref034] 34.Thorpe S. J., “The Speed of Categorization in the Human Visual System,” Neuron, vol. 62, no. 2, pp. 168–170, Apr. 2009, doi: 10.1016/j.neuron.2009.04.012 [DOI] [PubMed] [Google Scholar]

[pone.0264058.ref035] 35.Cichy R. M., Pantazis D., and Oliva A., “Resolving human object recognition in space and time,” Nat Neurosci, vol. 17, no. 3, pp. 455–462, Mar. 2014, doi: 10.1038/nn.3635 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0264058.ref036] 36.Carlson T., Tovar D. A., Alink A., and Kriegeskorte N., “Representational dynamics of object vision: The first 1000 ms,” Journal of Vision, vol. 13, no. 10, pp. 1–1, Aug. 2013, doi: 10.1167/13.10.1 [DOI] [PubMed] [Google Scholar]

[pone.0264058.ref037] 37.Khaligh-Razavi S.-M., Cichy R. M., Pantazis D., and Oliva A., “Tracking the Spatiotemporal Neural Dynamics of Real-world Object Size and Animacy in the Human Brain,” Journal of Cognitive Neuroscience, vol. 30, pp. 1559–1576, 2018, doi: 10.1162/jocn_a_01290 [DOI] [PubMed] [Google Scholar]

[pone.0264058.ref038] 38.Braak H. and Braak E., “Neuropathological stageing of Alzheimer-related changes,” Acta Neuropathol, vol. 82, no. 4, pp. 239–259, 1991, doi: 10.1007/BF00308809 [DOI] [PubMed] [Google Scholar]

[pone.0264058.ref039] 39.Braak H. and Braak E., “Frequency of stages of Alzheimer-related lesions in different age categories,” Neurobiol Aging, vol. 18, no. 4, pp. 351–357, Aug. 1997, doi: 10.1016/s0197-4580(97)00056-0 [DOI] [PubMed] [Google Scholar]

[pone.0264058.ref040] 40.Delacourte A., et al. , “The biochemical pathway of neurofibrillary degeneration in aging and Alzheimer’s disease,” Neurology, vol. 52, no. 6, pp. 1158–1165, Apr. 1999, doi: 10.1212/wnl.52.6.1158 [DOI] [PubMed] [Google Scholar]

[pone.0264058.ref041] 41.Braak H., Alafuzoff I., Arzberger T., Kretzschmar H., and Del Tredici K., “Staging of Alzheimer disease-associated neurofibrillary pathology using paraffin sections and immunocytochemistry,” Acta Neuropathol, vol. 112, no. 4, pp. 389–404, 2006, doi: 10.1007/s00401-006-0127-z [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0264058.ref042] 42.Johnson K. A., et al. , “Tau positron emission tomographic imaging in aging and early Alzheimer disease,” Ann Neurol, vol. 79, no. 1, pp. 110–119, Jan. 2016, doi: 10.1002/ana.24546 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0264058.ref043] 43.Villemagne V. L., Okamura N., and Rowe C. C., “Untangling tau imaging,” Alzheimer’s & Dementia: Diagnosis, Assessment & Disease Monitoring, vol. 4, no. 1, pp. 39–42, 2016, doi: 10.1016/j.dadm.2016.05.001 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0264058.ref044] 44.Khaligh-Razavi S.-M., et al. , “Integrated Cognitive Assessment: Speed and Accuracy of Visual Processing as a Reliable Proxy to Cognitive Performance,” Sci Rep, vol. 9, Jan. 2019, doi: 10.1038/s41598-018-37709-x [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0264058.ref045] 45.Kalafatis C. and Khaligh Razavi S.-M., “F3-01-03: the integrated cognitive assessment—employing artificial intelligence for the detection of cognitive impairment,” Alzheimer’s & Dementia, vol. 15, no. 7S_Part_16, pp. P864–P864, Jul. 2019, doi: 10.1016/j.jalz.2019.06.4598 [DOI] [Google Scholar]

[pone.0264058.ref046] 46.Kalafatis C., et al. , “P4‐207: EMPLOYING ARTIFICIAL INTELLIGENCE IN THE DEVELOPMENT OF A SELF‐ADMINISTERED, COMPUTERISED COGNITIVE ASSESSMENT FOR THE ASSESSMENT OF NEURODEGENERATION,” Alzheimer’s & Dementia, vol. 15, pp. P1355–P1356, Jul. 2019, doi: 10.1016/j.jalz.2019.06.3870 [DOI] [Google Scholar]

[pone.0264058.ref047] 47.Ruiz-Rizzo A. L., et al. , “Simultaneous object perception deficits are related to reduced visual processing speed in amnestic mild cognitive impairment,” Neurobiology of Aging, vol. 55, pp. 132–142, Jul. 2017, doi: 10.1016/j.neurobiolaging.2017.03.029 [DOI] [PubMed] [Google Scholar]

[pone.0264058.ref048] 48.Costa S. L., Genova H. M., DeLuca J., and Chiaravalloti N. D., “Information processing speed in multiple sclerosis: Past, present, and future,” Mult Scler, vol. 23, no. 6, pp. 772–789, May 2017, doi: 10.1177/1352458516645869 [DOI] [PubMed] [Google Scholar]

[pone.0264058.ref049] 49.DeLuca J., Chelune G. J., Tulsky D. S., Lengenfelder J., and Chiaravalloti N. D., “Is Speed of Processing or Working Memory the Primary Information Processing Deficit in Multiple Sclerosis?,” Journal of Clinical and Experimental Neuropsychology, vol. 26, no. 4, pp. 550–562, Jun. 2004, doi: 10.1080/13803390490496641 [DOI] [PubMed] [Google Scholar]

[pone.0264058.ref050] 50.Khaligh-Razavi S. M. and Habibi S., “System for assessing mental health disorder,” UK Intellect. Prop. Off, 2013. [Google Scholar]

[pone.0264058.ref051] 51.Khaligh-Razavi S.-M., Sadeghi M., Khanbagi M., Kalafatis C., and Nabavi S. M., “A self-administered, artificial intelligence (AI) platform for cognitive assessment in multiple sclerosis (MS),” BMC Neurology, vol. 20, no. 1, p. 193, May 2020, doi: 10.1186/s12883-020-01736-x [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0264058.ref052] 52.Nasreddine Z. S., et al. , “The Montreal Cognitive Assessment, MoCA: a brief screening tool for mild cognitive impairment,” J Am Geriatr Soc, vol. 53, no. 4, pp. 695–699, Apr. 2005, doi: 10.1111/j.1532-5415.2005.53221.x [DOI] [PubMed] [Google Scholar]

[pone.0264058.ref053] 53.Mioshi E., Dawson K., Mitchell J., Arnold R., and Hodges J. R., “The Addenbrooke’s Cognitive Examination Revised (ACE-R): a brief cognitive test battery for dementia screening,” Int J Geriatr Psychiatry, vol. 21, no. 11, pp. 1078–1085, Nov. 2006, doi: 10.1002/gps.1610 [DOI] [PubMed] [Google Scholar]

[pone.0264058.ref054] 54.Tadel F., Baillet S., Mosher J. C., Pantazis D., and Leahy R. M., “Brainstorm: A User-Friendly Application for MEG/EEG Analysis,” Computational Intelligence and Neuroscience, 2011. https://www.hindawi.com/journals/cin/2011/879716/ (accessed Jul. 28, 2019). doi: 10.1155/2011/879716 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0264058.ref055] 55.Chang C.-C. and Lin C.-J., “LIBSVM: A library for support vector machines,” ACM Trans. Intell. Syst. Technol., vol. 2, no. 3, p. 27:1–27:27, May 2011, doi: 10.1145/1961189.1961199 [DOI] [Google Scholar]

[pone.0264058.ref056] 56.Nonparametric Statistical Inference. Boca Raton, 2010. [Google Scholar]

[pone.0264058.ref057] 57.Kriegeskorte N. and Kievit R. A., “Representational geometry: integrating cognition, computation, and the brain,” Trends Cogn Sci, vol. 17, no. 8, pp. 401–412, Aug. 2013, doi: 10.1016/j.tics.2013.06.007 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0264058.ref058] 58.Missonnier P., et al. , “Early event-related potential changes during working memory activation predict rapid decline in mild cognitive impairment,” J. Gerontol. A Biol. Sci. Med. Sci., vol. 60, no. 5, pp. 660–666, May 2005, doi: 10.1093/gerona/60.5.660 [DOI] [PubMed] [Google Scholar]

[pone.0264058.ref059] 59.Babiloni C., et al. , “Sources of cortical rhythms change as a function of cognitive impairment in pathological aging: a multicenter study,” Clinical Neurophysiology: Official Journal of the International Federation of Clinical Neurophysiology, vol. 117, no. 2, pp. 252–268, Feb. 2006, doi: 10.1016/j.clinph.2005.09.019 [DOI] [PubMed] [Google Scholar]

[pone.0264058.ref060] 60.Choi J., et al. , “Resting-state prefrontal EEG biomarkers in correlation with MMSE scores in elderly individuals,” Scientific Reports, vol. 9, no. 1, Art. no. 1, Jul. 2019, doi: 10.1038/s41598-019-46789-2 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0264058.ref061] 61.Culham J. C. and Valyear K. F., “Human parietal cortex in action,” Curr Opin Neurobiol, vol. 16, no. 2, pp. 205–212, Apr. 2006, doi: 10.1016/j.conb.2006.03.005 [DOI] [PubMed] [Google Scholar]

[pone.0264058.ref062] 62.Fogassi L. and Luppino G., “Motor functions of the parietal lobe,” Curr Opin Neurobiol, vol. 15, no. 6, pp. 626–631, Dec. 2005, doi: 10.1016/j.conb.2005.10.015 [DOI] [PubMed] [Google Scholar]

[pone.0264058.ref063] 63.Fogassi L., Ferrari P. F., Gesierich B., Rozzi S., Chersi F., and Rizzolatti G., “Parietal lobe: from action organization to intention understanding,” Science, vol. 308, no. 5722, pp. 662–667, Apr. 2005, doi: 10.1126/science.1106138 [DOI] [PubMed] [Google Scholar]

[pone.0264058.ref064] 64.Davis S. W., Dennis N. A., Daselaar S. M., Fleck M. S., and Cabeza R., “Que PASA? The posterior-anterior shift in aging,” Cereb Cortex, vol. 18, no. 5, pp. 1201–1209, May 2008, doi: 10.1093/cercor/bhm155 [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Temporal dynamics of animacy categorization in the brain of patients with mild cognitive impairment

Hamed Karimi

Haniyeh Marefat

Mahdiyeh Khanbagi

Chris Kalafatis

Mohammad Hadi Modarres

Zahra Vahabi

Seyed-Mahdi Khaligh-Razavi

Roles

Abstract

1 Introduction

2 Methods

2.1 Integrated cognitive assessment (ICA) task

Fig 1. The EEG task.

2.2 Montreal cognitive assessment (MoCA)

2.3 Addenbrooke cognitive examination (ACE-R)

2.4 Subject recruitment

Table 1. Demographic information of participants.

2.5 EEG data acquisition and preprocessing

2.6 Univariate pattern analysis—Event-Related Potential (ERP)

2.7 Support vector classifier

2.8 Multivariate pattern analysis—Animal vs. non-animal decoding

Fig 2. Animal vs. non-animal decoding.

2.9 Multivariate pattern analysis—Pairwise decoding

Fig 3. Representational dissimilarity matrices (RDMs).

2.10 Multivariate pattern analysis—HC vs. MCI classification

Fig 4. HC vs. MCI classification over time based on EEG response patterns.

2.11 Multidimensional Scaling (MDS)

2.12 Statistical analysis

2.12.1 Bootstrap test

2.12.2 Permutation test

2.12.3 Rank-sum

3 Results

3.1 A reduction in the speed and accuracy of animacy processing in MCI patients

Fig 5.

Fig 6. Animal vs. non-animal decoding time course across groups of EEG channels.

3.2 Comparing patterns of visual information processing in HC and MCI

Fig 7.

3.3 Temporal dynamics of animacy categorization

Fig 8.

4 Discussion

4.1 Task differences in EEG studies of HC and MCI

4.2 What do differences in the pattern of activation mean?

4.3 Neural speed of information processing in MCI patients

5 Conclusion

Supporting information

Acknowledgments

Data Availability

Funding Statement

References

Decision Letter 0

Stavros I Dimitriadis

Roles

Author response to Decision Letter 0

Decision Letter 1

Stavros I Dimitriadis

Roles

Acceptance letter

Stavros I Dimitriadis

Roles

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases