Frontal Functional Connectivity Impairment in Early Stages of Alzheimer’s Disease

Many neurophysiological markers such as neuronal oscillations, cortical neuronal synchronization and long-range neuronal coupling have been suggested to permit the identication of early cognitive impairments in Alzheimer's disease (AD). However, it is still unclear whether alterations in long-range Functional Connectivity (FC) constitute part of early mechanisms in the initial stages of AD. Eighteen participants (69-88 years old), classied as early AD and matching healthy controls, were evaluated while performing a virtual spatial navigation task. We combined electroencephalography (EEG) and eye movement recordings during the performance of a virtual version of the Morris Water Maze (VMWM), where participants had to nd a submerged, invisible platform. The groups were compared in their navigation performance with different metrics of brain activity.

reported as one of the rst signs in AD, showing the role of the neurodegenerative process in the hippocampus, a key part of the brain, in solving spatial navigation challenges [8][9][10][11]. The hippocampus and parietal, prefrontal, and occipital regions constitute the core network for navigation even in the prodromal and preclinical stages. However, an interesting association was the activation in young and elderly controls, but not in AD and MCI subjects, of several frontal lobe areas related to action planning sequences, including executive functioning, organization, error monitoring, and global response decision making [12][13][14]. A putative effective method of controlling disease progression is an early diagnosis and an appropriate management strategy that starts from the beginning of the cognitive decline. For this reason, one of the main challenges for developing treatment methods in all neurodegenerative diseases, especially AD, is to understand the pathological brain mechanisms underlying cognitive impairment [15,16]. In this work we evaluate whether long-range Functional Connectivity (FC) alterations constitute part of the early mechanisms at the initial stages of Alzheimer's disease.

Participants
A total of thirty-eight individuals were recorded, 22 elderly cognitively healthy controls and 16 early AD (eAD), where the mean age was (70.32 (+/-7.79); 6 male) and (77.06 (+/-6.97); 8 male) years for the control and eAD group, respectively. The exclusion criteria for both groups were evidence of nondegenerative dementia (e.g., in ammatory, metabolic, or vascular etiology); dementia or mild cognitive impairment of doubtful origin; severe medical conditions that limited their ability to participate in the study (e.g., uncompensated diabetes, severe hypertension). The Scienti c and Ethical Committee approved all procedures involving participants of the Clinical Hospital of the University of Chile Protocol number: 26/2015. In addition, all participants signed an informed consent previously approved by this Ethics Committee.

Neuropsychological testing
Cognitive functions were examined with the Clinical Dementia Rating scale (CDR) [17] and CDR Sum-of-Boxes (CDR-SOB) [18], a numeric scale used to quantify the severity of dementia (e.g., its stages). Also, the Montreal Cognitive Assessment (MoCA) [19], the Montreal Cognitive Assessment Memory Index Score (MoCA-MIS) [20], and the Minimental State Examination (MMSE) were used as a rapid screening instrument for mild cognitive dysfunction. A complete clinical evaluation, including full neuropsychological tests, was performed by a neurologist from the Department of Neurology of the Clinical Hospital of the Universidad de Chile, who was blind to the performance of the subjects in the navigation task. Participants were classi ed in cognitively unimpaired (CDR0) and early AD (eAD) based on their cognitive status.
The study involved 18 volunteers: 9 participants (eAD group) and 9 participants (control group) aged between 61-88 years. The eAD group consisted of patients with cognitive impairment with apparent memory de cits, with a global CDR 0.5, a CDR-SOB ≥ 1.5, and a MoCA-MIS ≤ 10, with two or fewer words out of 5 of the MoCA tests recalled spontaneously. Also, eAD participants had a very mild loss of instrumental activities of daily living. Control participants were submitted to the same neurological and neuropsychological evaluations as the eAD group. The demographic data of the participants are shown in Table 1. Spatial navigation paradigms: Virtual Morris Water Navigation Task Spatial exploration was assessed by the VMWM task in both eAD and control participants. The task was executed through an open-use program with support provided by its author [10]. The virtual environment of the VMWM task simulates the traditional MWM, which consists of a circular pool located inside a room with visual cues on its four walls. The task is presented on a computer screen that the participant must navigate through the buttons of a keyboard to nd a platform hidden under the water surface. The setup enables controlling a series of parameters to optimize the sensitivity of the task, related to characteristics of the visual cues and the location of the hidden platform, the maximum duration, and the number of trials. After nishing the task, the program stores in text les the information related to the route traveled, the latency to nd the platform, and the relative percentages of the length of the path spent in each quadrant of the virtual pool.
All participants performed a well-established navigation task which was divided into three stages. Training (Stage I): this task involved nding the platform hidden under the water in a room furnished with visual clues on each of its walls. The platform became visible after one minute of exploration if the participant did not nd it. Therefore, the participant needed to reach the platform to nish the trial. Before the end of the trial, the participant had 2 seconds to rotate in place to explore its position in the environment visually. This stage comprised four repetitions of the trial. The platform was kept in the same hidden position inside the pool, and the participant always started each repetition from a different position. Task I (Stage II): The participant had to nd the platform hidden based on different visual cues to those presented in the training session as in the previous stage. The task included twenty trials divided into four groups. A two-minute break was considered among each group of trials. Similar to the training session, the platform was always kept in the same position inside the pool. The participant always started each repetition from a different position inside the pool. Task II (Stage III): in an equivalent room but including different visual cues, the participant had to select one of two visible platforms; only one represented the correct choice. A note on the screen indicated whether the platform chosen was correct. This task was used to control each participant's visual and psychomotor functioning, making it possible to rule out any deterioration in these parameters as plausible reasons for unsatisfactory performance in task I.

EEG recordings and eye-tracking
During the behavioral navigation task, the continuous acquisition of the electroencephalographic activity of each participant was carried out beneath adequate recording conditions. An EEG system of 32 + 8 channels was employed (8 external channels to measure electro-ocular activity and referential mastoid recording; BioSemi®). The acquired analog signal was ltered between 0 (real DC) and 1000 Hz before its conversion, sampled at 2048 Hz, and digitally converted with a precision of 24 bits. The recording system used a speci c reference system (CRS-DRL) that allows unlimited data storage. After electrodes were located, the participant was installed in a suitable chair in front of the monitor where the VMWM task was displayed. The head was positioned on chin support to minimize movement during the task and allow optimal detection and recording of eye movements. Eye-tracking was performed using an Eyelink® 1000 system, which digitizes and stores eye-tracking data in a binary le convertible to text. The bi-dimensional position of the pupils was obtained at a frequency of 500 Hz. The system also automatically registers the occurrence of blinks, xations, and ocular saccades based on user-de ned initial setting parameters.

Data Analysis
Henceforth, all data analyses made in this study, including behavioral parameters of space navigation in the VMWM task, electroencephalographic signals, and eye-tracking data, were performed with MATLAB® (The MathWorks, Inc.). First, for behavioral data, the text les generated by the program after nishing the task were imported and analyzed using custom algorithms designed for these purposes, employing general functions of the basic software package. Second, the binary .bdf les generated were directly preprocessed using the Fieldtrip open-source toolbox for EEG signals. The eye-tracking binary data were initially converted to ASCII text les using the Eyelink® executable EDF2ASC.exe. Finally, the EYE-EEG MATLAB toolbox was used as a plugin of the EEGLAB package to import, visualize and verify the detected eye-tracking events and synchronize with the EEG signals [21].
EEG signals were visually inspected by a quali ed clinical neurophysiologist to judge the collected signals' quality and reject abnormal recordings (i.e., epileptiform activity and abnormal basal rhythms).
Then the segments of data presenting non-extractable artifacts were manually marked for the exclusion of the analysis. O ine lter settings were bandpass lter at 1-40 Hz (Butterworth, FIR). Next, artifact elimination was applied using ICA decomposition ( eldtrip toolbox) over the whole continuous record. Finally, noise components were identi ed in a semi-automated way, utilizing algorithms executed in the EYE-EEG package to detect ocular movement-related components.
Continuous EEG signals were segmented using the ocular tracking signal synchronized with the ltered EEG signal, between − 1000 and 3000 ms around the ocular xations. To ensure that the data did not signi cantly bias, we evaluated the rst 30 seconds of the records because the time of each participant was unsteady over the trials. Fixation-related epochs were consequently subjected to analyses in the time and frequency domains. First, the time-frequency decomposition was computed on the EEG epochs tapered by a sliding Hanning window, using the same xed window length seconds for all frequency ranges of 1 and 40 Hz in steps of 1 Hz implemented at the time between − 750 and 1500 ms in steps of 10 ms. Second, we implement an analysis method multitaper based on multiplication in the frequency domain, obtaining nally output power-spectra ( eldtrip toolbox). First, for each time, frequency, and electrode, was calculated as the modulus of the mean across trials. Then, for each frequency, values of power-spectra were transformed to Z scores, normalizing by the corresponding mean and standard deviation of prestimulus time between − 750 and − 450 ms. Finally, these normalized values were compared for each time, frequency, and electrode between experimental groups.

Eye-tracking analysis
Fixations and saccades were automatically identi ed based on the velocity (30°/s) and acceleration threshold (8000°/s2). Saccades longer than 5 ms and smaller than 100 ms and xations between 50 and 800 ms were picked for further processing. In addition, blinks were de ned as the absence of pupil data.

Statistics
Demographics and neuropsychological performance were divided into continuous variables expressed as mean and standard deviation (SD), while the categorical variables were expressed as frequencies (%). Wilcoxon rank-sum test was used for age, education, CDR-SOB, MoCA, MoCA-MIS, and MMSE comparison between groups. A chi-square test was used for gender comparison (Fisher's exact test). Differences in ocular behavior as the frequency of ocular movements ( xations and saccades) were measured using a nonparametric Wilcoxon rank-sum test between the control and eAD group.
Additionally, we obtained a heat map of the probability of differences and a map of signi cant differences between the exploration performed by both groups. Finally, we conducted a rank-sum test as a statistical approach pixel by pixel on the image, performed at the 0.05 signi cance level.
Statistical tests based on permutations were applied to evaluate the differences between groups of the oscillatory activity, such as the time-frequency power spectral decomposition and coherence. The Montecarlo method was considered an estimator of the permutation's signi cance probabilities and critical values (two-tailed, alpha: p < 0.01, cluster correction, cluster-alpha: p < 0.05). Given many corrections applied to the number of electrodes to be compared between groups, we used cluster as a correction method that solves the Multiple Comparison Problem (MCP) [22,23]. Moreover, we calculated effect sizes (Cohen's d) as the difference of the means of groups (control and eAD) divided by the weighted pooled standard deviations of the groups. Cohen's d effect size of 0.2 to 0.3 could be a "small" effect, around 0.5 a "medium" effect, and 0.8 to in nity, a "big" effect (a Cohen's d greater than 1 as possible). We estimated signi cant differences intragroup for the baseline in a beta-band coherence region of interest (15)(16)(17)(18)(19)(20) Hz and between 0-300 ms) using a statistical threshold criterion of effect size (> 0.5) and one-sample t-test (< 0.05). This threshold was applied to a chi-square test for intragroup ratios. The data were treated by MATLAB toolbox and Fieldtrip toolbox.

Results
The detailed demographic and neuropsychological assessment information is presented in Table 1. Nine patients diagnosed with eAD, and nine matched control participants were recruited for this study. We found no signi cant differences in the age distributions between the eAD (76,7 ± 6,2 years old) and the control groups (71,2 ± 8,5) years old); Wilcoxon Rank sum test, (p = 0.2138). Sex subgroups were the same in both groups (2 males

Ocular behavior and visual exploring strategies
We rst obtained a graphical representation of the visual eye xations distributions per group, represented through heat maps. Then, a Gaussian lter was applied for each heat map seeking to normalize the location and duration of the eye xations. The eye xations in the control group were more focused in the central region (Fig. 1A). While the eAD group, the visual exploration was much more homogeneous in the image and without a clear preference on the Region of Interest that could guide the task's solution (Fig. 1B). We decided to calculate the absolute differences between groups with this approach as an example of visual exploration. Although both groups visually explored the center of the image in contrast to the periphery, we found the main differences in the intermediate image region. By comparing the absolute value of difference for both the control and the group with eAD, we observed that the control group had a high preference to x the central area very close to a position on the water in the pool (the region where the platform could be located). In contrast, the eAD group focused on the upper midline without a de ned focus in the target platform.
Additionally, we performed a comparison pixel-by-pixel of the visual exploration of the groups. We used level curves for the heat maps of the eye xation with the normalized values. Furthermore, we applied a rank-sum test for all pixels of the heat map to obtain a probability of differences in the exploration with a color scale between 0 and 1, where the values closer to 1 (red color) were more likely to be different. This result highlights the lower central region, which could correspond to an area close to the hidden platform.
The segment with statistical differences shown in red (Wilcoxon rank-sum test, p < 0.05). (Fig. 1C and  1D). To con rm that the visual exploration behavior of participants with eAD differed from that of controls and that their xation targets were located more in the lateral, top, or bottom positions of the middle line, whereas controls xated more on the straight-ahead position, we analyzed the distribution.
Eye xation frequency distribution revealed a signi cant difference between control and eAD groups (KS test; D = 0.11, p = 0.0014). The participants with eAD had more eye xations over the middle line and less bottom relative to controls (Fig. 2). These results con rm that both groups differ behaviorally in their visual exploration, re ecting their ability to process visual information.

Time-frequency analysis
Power spectra were performed to identify differences in a particular frequency band and use this range for subsequent functional connectivity analysis. We applied multiple time-frequency transformation (TF) windows in the frequency domain. The TF decomposition was achieved by generating epochs in the signal associated with the eye xation and taking a baseline that considered − 750 to -450 ms before the event. Two examples of TF maps are shown for each group (Fig. 3). Moreover, we calculated the average per group for the Oz channel, where both the Control group and the eAD group showed a marked predominance of activity in the low-frequency spectrum, in particular, theta and alpha bands, but also an activity relevant in the beta band that went only recognized in the control group ( Fig. 4A and 4B).
A two-tailed non-parametric cluster-based permutation test with an alpha error level of 0.01 for the Oz channel was performed to test for differences in the power between the groups. This test was conducted on the TF data from − 250 ms to 600 ms, which means that we include what happens eventually with the previous saccade, and after the origin of the xation (adequate window to involve the visual event and its early cognitive processing) and this for a frequency range that considers low frequencies 2 Hz to 25 Hz. A Monte Carlo estimate of the permutation p-value was computed by randomly permuting condition labels (N = 1000). We observed signi cant differences in the beta band between 15Hz to 18 Hz for the rst 250 ms post-xation and the same effect between 17 Hz to 20 Hz for the 250 to 600 ms post-xation (Fig.  4C). Additionally, the power difference (control -eAD) is shown in a topographic map for the beta frequency band, within the range of 15 Hz to 20 Hz in the time interval between 0 and 200 ms (Fig. 4D).

Functional connectivity analysis
Coherence can be considered a measure of normalized linear synchrony of different EEG series in the frequency domain or a functional substrate of activity patterns with high relative power, such as an ERP.
Here, coherence analysis was applied to pairwise EEG channel subgroups for the eAD and control groups in the beta frequency band. The analysis of the frequency spectrum to Cluster-based permutation tests mentioned above had signi cant group differences. In particular, the activity of this component was associated with visual processing at the time when we expect to get connectivity differences between both groups concerning the visual information processing.
The mean coherence between 14 channels of eAD and the control group in the beta frequency band (15 Hz to 20 Hz) is shown in (Figs. 5A and 5B). The mean coherence distributions of the two groups were symmetrical, indicating the presence of fast bidirectional transmission of information between brain areas. These pairwise electrodes were principally related to the frontoparietal activity, whose functions are widely associated with working memory and, in particular, relevant in visual working memory and visual attention. The diagonal of the coherence matrix was all = 1 because the EEG signal could achieve complete synchrony with itself. The coherence matrices of the two groups were complex and initially very similar. Then, however, we calculated the connectivity matrices with baseline correction from − 750 to -450 ms. In the control group (Fig. 5C), there were more areas of high (red color) values of coherence than that of the eAD group (Fig. 5D). While in the eAD group, the low values of coherence were distributed in areas like frontal and frontoparietal areas for the beta band. The matrix of P values returned by a onesample t-test between groups (Fig. 5E) showed the signi cant difference uncorrected on coherence.
To know the impact of these results, the effect size of the coherence spectrum was estimated. The values considered the Frontoparietal-occipital axis, where the biggest differences between groups were appreciated. The effect size was obtained by using Cohen's d. This value was calculated by subtracting the mean from the eAD group minus the mean of the Control group and dividing by the standard deviation of the population to which both groups belong. These values could be interpreted as small (Cohen's d = 0.2 to 0.3), medium (Cohen's d = 0.5 to 0.7) and large (Cohen's d > 0.8) values. We proposed to consider the medium and large values in the matrix (Fig. 5F).
Additionally, only the large values of Cohen's d were used in a topographic map highlighting which areas could impact the participants' behavioral performance (Fig. 6A). This map of the effect size showed the regions that varied the most between groups and could account for the early behavioral differences of the eAD participants for the strategies used in the virtual navigation task. We found the more prominent changes in coherence were essentially frontoparietal and suggest that lower activation or synchronization in prefrontal cortices may be the cause of worse spatial navigation.
Then, we created a binary connectivity matrix. Here, the threshold is set as a combination of values calculated previously in the beta frequency band (15 Hz to 20 Hz). Firstly, we considered an effect size (> 0.5); secondly, when the P-value matrix for one-sample t-test is (< 0.05). The color scale represents in black signi cant coherence differences from the baseline time (-750 to -450 ms). At the same time, the white color depicts no differences from baseline time for the activity of the pairwise comparison. The statistical analysis results of the intragroup coherence for the control connectivity matrix show many pairwise comparisons with signi cant differences (Fig. 6B). The eAD matrix shows only one electrode pair with signi cant brain activity changes (Fig. 6C). Finally, we show the topographic distribution of electrode pairs with signi cant coherence differences from baseline time by groups, using as a threshold the p < 0.05 and Cohen's d > 0.5 (Fig. 6D). Based on these results, we suggest that the long and shortdistance synchrony and Intra and Interhemispheric connections that present a higher degree of differences to the coherence belong to the frontoparietal region of the brain. This could be the consequence of less effective planning of navigation routes since this region is critical for executive functions, including planning, organization, error monitoring, and decision making.

Discussion
Despite the vast progress of recent years in the study of AD, the principal mechanisms involved in cognitive de cits in the disease are still unclear. This study compared ocular behavior and the functional connectivity in control vs. eAD subjects in a sensitive early detection test of cognitive impairment [24][25][26][27][28]. Our results showed a signi cant difference in visual exploration between the groups; however, the xing area percentage did not show signi cant differences. Additionally, early differences were observed in the beta band activity (15 Hz to 20 Hz) that has been related to the execution of cognitive processes. Finally, the coherence between different cortical regions, signi cant effect differences were observed between groups in the prefrontal region in the beta band with consequences for less effective planning of navigation strategies.

Very early Alzheimer's disease and ocular behavior impairments
Patients with a diagnosis of eAD performed signi cantly different exploration strategies than the control group to reach the visual keys in the task. These ndings rea rm the notion observed in other studies that identi ed increased instability in xation, enhanced latency of voluntary exits, the direction of microsaccades, greater anti-saccadic errors, and decreased correction of this anti-saccadic error in AD participants [29][30]. The strategies of visual exploration based on the distribution of the xations were analyzed employing heat maps. They were described as an early manifestation of AD alterations of executive functions and different visual parameters such as visuospatial skills, processing, and selective visual attention [31]. We showed that the xations in the control group were preferably concentrated on the objects being relevant for the task (visual keys and possible platform location), unlike the eAD group, where the visual exploration was more homogeneous in the image and without a clear preference on regions of interest that could guide to solve the task. Furthermore, the participants with eAD had more eye xations over the middle line and less bottom relative to controls. These results could suggest that visual exploration strategies of the eAD participants can be associated with attention de ciencies and the involvement of prefrontal networks and visual attention WM [32].
Early electrophysiological changes and role of prefrontal Beta band coherence in eAD.
Our electrophysiological results showed differences in the application of the spatial navigation task between the groups. The spectral power analysis through the time-frequency maps, presenting differences in activity in the theta frequency ranges, alpha, and signi cantly in the beta band (15 Hz to 20 Hz), con rmed by the analysis of Cluster-based permutations tests. These ndings are consistent with a large number of studies that reported increases in the theta band and a decrease in the alpha and beta band of patients with AD [33]. Other studies even attributed the reduction in the alpha band and reduced beta power in the parietal and occipital regions as a differentiating factor between normal aging, MCI, and AD [34][35][36][37][38][39][40][41]. Particularly, the beta band has been related to memory processes, and their decrease in the eAD group could be associated with di culties in spatial navigation when the reactivation of memories is necessary to nd the platform by visual cues [42]. Beta oscillations have been described in many studies of perceptual, cognitive, and motor processes. Although a single term could not entirely explain its function, the more accepted descriptions ascribed a role of the beta band in interneural communication of inhibitory networks and high executive demands [43,44], while synchronization might be involved in sensory processing [45]. Our results showed a reduction in the beta band functional connectivity in the eAD group in prefrontal cortices. Both alpha and beta frequencies are associated with many functions that represent top-down in uences. Still, particularly beta oscillations are crucial to long-distance communication between cortical regions, with the function of maintaining a constant update of brain status [42].
Additionally, the prefrontal cortex could be involved in associative learning in navigation and changing strategies and engaged in processes to search for speci c objectives such as selecting the best route or trajectories in space navigation processes [46]. Our results, as well as previous studies, showed that selective visual attention is sensitive in eAD. Besides, disconnections between brain regions were described to play an essential role in cognitive impairment in AD, with reduced synchrony as a marker [16, [47][48][49].
Some studies have proposed that activating the right prefrontal cortex (PFC) during spatial WM tasks was described in young adults. In contrast, older adults presented bilateral activation of PFC as a compensatory response to cognitive impairment. Likewise, during particular memory tasks (such as coding complex visual scenes), healthy older adults also showed bilateral hyperactivation of the frontal cortex. Furthermore, during the MCI to AD progression, it has been proposed that a disintegration of the compensatory networks occurs due to the observed lack of activity of the lateral regions of the PFC, the precuneus, and posterior parietal cortex. These regions are all involved in the executive function compared to controls 50,51 . These ndings are consistent with our proposed that the symptoms observed in patients with eAD are closely related to a loss of functional connectivity, re ected by physiological changes and an attenuation of the electrical activity [48,49,52,53].
As we supposed, spatial navigation impairments can be associated with early mechanisms of cognitive deterioration in the progression to AD. This task joints the occipital, parietal, and frontal cortices functions. We already know that the navigation abilities depend on the occipital cortices for the early processing of visual information and the parietal and hippocampal cortices on the generation of the cognitive map and egocentric/allocentric navigation strategies. At the same time, the frontal cortex is relevant for its participation in the decision, development, and planning of actions [12]. A study by Coughlan showed that morphological changes (in particular the accumulation of Aβ in the cerebral cortex) follow a typical course of progression, which coincides with the structures involved in spatial navigation. Therefore, it seems that these structural changes are not isolated and that the functional alterations are later.
Furthermore, these changes occur in regions that appear to be involved in the networks necessary to establish functional connectivity in navigation tasks [13]. In brief, it is essential to understand the proposed mechanism of coordinate participation of all these mentioned cortices for better performance.
We suggest an operational model whose functionality depends on many structures, where each area's integrity matters but the generated functional connectivity. Thus, the spatial navigation impairments are consistent with the progress of the loss of cognitive skills and less functional connectivity in very early AD [8-11, 15, 16].

Limitations
Our work provides new evidence for a dysfunctional neural pattern in visuospatial processing, higher order attention and executive function in early AD patients, but also reveals some limitations that should be mentioned. First, this study is limited by the statistical approach as it is necessary to apply many corrections for the number of electrodes that were compared between the study groups. To solve the problem of multiple comparisons (MCP), the cluster-based permutation test was used as a correction method in the time-frequency analysis and in the data coherence analysis [22,23]. In addition, the timing of trials was variable within subjects and between groups. However, this random temporal variability was corrected using the rst 30 seconds of the recordings, as each participant's timing was unstable across trials, avoiding systematic bias.

Conclusions
This study showed early electrophysiological alterations in the eAD group in the prefrontal cortex. These changes could play a critical role in executive functions, planning, organization, error monitoring, and decision-making for spatial navigation and predictive motor planning [11,14,54]. Thus, we can conclude that the study of long-range functional connectivity in early stages could help predict AD progression.
Finally, this study contributes to a better understanding of the mechanisms underlying cognitive impairment in AD and opens the door to better therapeutic management by identifying critical periods of change. This could provide the possibility of intervening in the system with the generation of more effective therapies when changes are still reversible and not when neurodegenerative processes have already been installed.     Coherence Beta (15-20 Hz) connectivity matrices from 0 to 300 ms. Absolute Coherence matrix for the Control a) and the eAD b) groups. Connectivity matrices with baseline correction from -750 to -450 ms c) Coherence matrix for the control group, d) Coherence matrix for the eAD group. The color scale represents low (blue) and high (red) coherence between each pair of electrodes. e) Matrix of P-value of the two groups obtained by t-test, the highest value with red color, and f) Matrix of differences between groups for Page 23/24 an effect size > 0.5 Cohen's d statistic (medium-large effect size). Thus, the color scale represents in black (dark color) medium differences of effect size and in white (light color) marked differences of effect size for the pair of electrodes. In white, there is no difference for baseline period for the activity of the electrode pair, using as a