Visual search in Alzheimer's disease and amnestic mild cognitive impairment: An eye‐tracking study

Abstract

INTRODUCTION

Visual search impairment is a potential cognitive marker for Alzheimer's disease (AD) and amnestic mild cognitive impairment (aMCI). The aim of this study is to compare eye movements during visual tracking in AD and aMCI patients versus healthy controls (HCs).

METHODS

A prospective cohort study included 32 AD and 37 aMCI patients, and 33 HCs. Each participant was asked to look at the target object in a visual stimulus containing one target and eight distractors, and eye movements were recorded with EyeLink 1000 Plus.

RESULTS

AD patients had fewer fixations and shorter target fixation duration than aMCI patients and HCs. Fixation durations were also shorter in aMCI patients compared to HCs. Also, AD patients were more fixated on distractors than HCs.

DISCUSSION

Our findings revealed that visual search is impaired in the early stages of AD and even aMCI, highlighting the importance of addressing visual processes in the Alzheimer's continuum.

Highlights

• AD patients looked to distractors more and longer than the target compared to aMCI patients and older healthy individuals.

• aMCI patients had an impaired visual search pattern compared to healthy controls, just like patients with AD.

• The visual search task differentiated AD and aMCI patients from healthy individuals without dementia.

1. INTRODUCTION

Progressive impairments in cognitive abilities occur in Alzheimer's disease (AD). ¹ In amnestic mild cognitive impairment (aMCI), cognitive impairments occur in memory and other cognitive functions less severely than in AD, and these patients continue their daily lives independently despite difficulties. ² Despite common cognitive impairments, studies addressing visual difficulties in AD and aMCI are relatively limited. ³ Visual information is projected from the primary visual cortex to the ventral or dorsal pathway for higher‐level visual processing. ⁴ These pathways were found to be affected in AD and aMCI. ⁵ In addition, structural and functional changes in the occipital and temporal lobes of the brain have been demonstrated in AD and aMCI. ⁶ , ⁷

Visual search is defined as performing an eye movement to find an object or target of interest in the visual environment. ⁸ Visual search requires distinguishing features and processing information in the visual field. ⁹ For an effective visual search, it is necessary to carry out functions such as recognizing objects and shifting attention with visual guidance. ¹⁰

Thanks to eye movement recording devices with high visual acuity, eye movements can be monitored instantly and with high resolution. The movements in which the eye, which places the object of interest in the center of the fovea, stays still and focuses on the object, are called fixations, and the rapid movements of the eye from one point to another are called saccades. Counts and durations of fixations are important to eye movement components of visual attention processes. ¹¹ , ¹² It has been shown that fixation durations increase when targets and distractors are similar in visual searches and therefore when objects need to be examined in detail. ¹³ , ¹⁴ It has also been shown that fixation durations tend to increase as the number of fixations increases. ¹⁵ It has been suggested that this is associated with an initially coarse visual search strategy and then a more refined search strategy during visual search. ¹⁶ , ¹⁷ Considering that eye tracking is not an invasive method, eye movements can be recorded practically, and this method provides objective results, it is suggested that oculomotor metrics may have the potential to be valid biomarkers in AD. ¹⁸

Previous studies have examined visual search using a task to find one or more targets, including a letter, symbol, or object, among distracting stimuli. ¹⁹ , ²⁰ , ²¹ , ²² , ²³ , ²⁴ , ²⁵ , ²⁶ , ²⁷ , ²⁸ , ²⁹ These tasks provided important information regarding the process of searching for the target in the visual field. However, the limitation of these previous studies was that visual search was evaluated with one or more objects rather than natural scenes reflecting daily life. These studies have shown that AD patients focus on non‐target areas, have difficulty finding the target, search for the target quickly, and spend more time on distractors. ¹⁹ , ²⁰ , ²¹ , ²² , ²³ , ²⁴ , ²⁵ , ²⁶ , ²⁷ , ²⁸ , ²⁹ Other examinations using natural scenes have revealed changes in visual search in AD patients compared with healthy controls (HCs). ³⁰ , ³¹ , ³² However, in one of these studies, increased and prolonged fixation on distractors in aMCI patients was demonstrated compared to controls, ³⁰ whereas in another study, visual search performance was not different between aMCI patients and controls. ³¹ In yet another study, an aMCI group was not included. ³² Despite the growing evidence mentioned above showing poor visual search performance in AD, there are not enough studies yet to reach a consensus on visual search performance in aMCI. ¹⁹ , ²⁰ , ²¹ , ²² , ²³ , ²⁴ , ²⁵ , ²⁶ , ²⁷ , ²⁸ , ²⁹ , ³⁰ , ³¹ , ³² This study aims to examine visual search with stimuli that reflect the real world in AD, aMCI, and cognitively normal older individuals. We aim to examine visual search in stimuli that reflect the real world in AD, aMCI, and cognitively normal older individuals. Consistent with this, we hypothesized that there may be visual search difficulties in AD and aMCI compared to older controls. We hope to demonstrate that these hypothetical visual search difficulties may be cognitive and behavioral markers that can distinguish AD and aMCI patients from older individuals.

2. METHODS

2.1. Participants

Thirty‐two Alzheimer's disease (AD) dementia, 37 amnestic mild cognitive impairment (aMCI), and 33 cognitively normal HCs participated in the study. Patients with AD were included in the study in line with the criteria of McKhann et al. ¹ evaluated by neurologists. Newly diagnosed AD patients without medical treatment and with a Clinical Dementia Rating score of 1 were included. aMCI patients were included in the study according to the criteria of Petersen et al. ³³ and Albert et al. ³⁴

RESEARCH IN CONTEXT

1. Systematic review: The authors reviewed the literature on visual search for Alzheimer's disease (AD) and mild cognitive impairment (MCI). While there are findings indicating inefficient visual search in AD, the evidence for visual search in MCI is still relatively unclear. We hypothesized that there might also be an alteration in visual search, as eye movement deficits are present in both MCI and AD.

2. Interpretation: Our findings provided evidence for changes in visual search performance in both AD and amnestic MCI (aMCI) patients, characterized by a tendency to look at distractors more and longer than the target. These results confirm past studies in patients with AD and provide accumulating evidence associated with visual search ability in aMCI.

3. Future directions: The identification and follow‐up of cognitive markers of AD are of great importance. The visual search task gives an idea about the difficulties experienced by patients in daily life. Whether visual search based on eye tracking are biomarkers in AD and MCI should be evaluated longitudinally with larger populations.

HCs consisted of individuals who were cognitively normal according to the neuropsychological tests and without any disease. Participants with neurological or psychiatric diseases or drug use were excluded from the study. In addition, individuals with eye disorders that could affect the study during the examinations performed by the clinician were excluded from the study.

2.2. Neuropsychological tests

The Mini‐Mental State Examination (MMSE) was applied in the general cognitive status assessment. ³⁵ , ³⁶ Episodic verbal and visual memory were evaluated with the Öktem Verbal Memory Processes Test and the Rey Complex Figures Test. ³⁷ , ³⁸ , ³⁹ , ⁴⁰ , ⁴¹ Executive functions were evaluated with the Stroop Test and the Wisconsin Card Sorting Test. ⁴² , ⁴³ , ⁴⁴ , ⁴⁵ The Digit Span and Trail Making tests were used for attention. ⁴⁶ , ⁴⁷ , ⁴⁸ The Benton Face Recognition and Judgement of Line Orientation tests were used for visual‐spatial skills. ⁴² , ⁴⁹ , ⁵⁰ , ⁵¹ These tests were performed in the application of inclusion and exclusion criteria to participants during the clinical evaluation.

2.3. Eye movements

2.3.1. Apparatus

An EyeLink 1000 Plus eye‐tracker was used for measuring eye movements in the visual search paradigm (SR Research, Mississauga, Ontario, Canada). The spatial resolution was 0.01^°, and the sample rate was 2000 Hertz. Response time was set to 4 ms and the screen refresh rate was set to 60 Hz. The screen width was 29^° x 17^°, and the screen resolution was 1920 × 1080 pixels.

A 9‐point calibration and validation were performed before the trials. Participants were seated in an adjustable chair 52 cm across from the screen. Eye movements were performed using a height‐adjustable head support (SR Research).

2.3.2. Stimuli

The color photographs used in this study were taken by the research team with a CANON EOS M50 camera. The resolution of the photos was 1920 × 1080 pixels.

The stimuli were nine square pieces of 5° each taken from each photograph (Figure 1). Each piece was adjusted to be 0.5^° apart. The fixation mark appears in the center of the screen before and between trials. Therefore, to adjust the difficulty of finding the target equally, the target stimulus was arranged to appear in all eight pieces except the central piece. The target was adjusted in such a way that it occurs two times in each of the eight pieces except for the central piece. Trials were randomly presented to the participants.

FIGURE 1.

The color photographs used in this study are shown on the left and the visual search paradigms are on the right. The paradigms, created by taking nine pieces from each photograph, contain only one piece with a car image.

2.3.3. Visual search task

Experimental Builder software (SR Research) created the visual search task. The task consists of two practice trials and 16 test trials. The target stimulus was the car picture. Participants were asked to look at the square piece with the car image.

The fixation mark (white cross) was shown for 1000 ms on a grey background (RGB 127, 127, 127, 255), followed by the stimuli for 3000 ms, and then a blank grey background for 2000 ms in both the practice trials and test trials. The instruction for the task was explained verbally and was also shown in writing on the screen (Figure 2).

FIGURE 2.

The fixation mark was presented for 1000 ms followed by the stimulus for 3000 ms in trials. A grey background was displayed for 2000 ms between trials.

2.3.4. Data analysis

Each trial was organized with one target and eight distractors. The piece with the car picture was determined as the target region of interest (ROI), and the eight pieces without the car picture were determined as the distractor ROI. The total fixation durations and the fixation counts were calculated for target and distractor ROIs. The fixation counts for the target ROI were calculated by averaging the fixations that occurred on the piece containing the target from all trials. For fixation counts for the distractor ROI, the fixations to non‐target pieces in all trials were similarly averaged.

Eye movement raw data were extracted from data viewer (SR Research) software and analyzed. Artifacts such as blinks have been filtered out.

2.4. Statistical analysis

A one‐way analysis of variance (ANOVA) was used for demographic and clinical variables including age, years of education, and MMSE scores of the groups. The chi‐square test was used to compare the sex frequency of the groups. Eye data including fixation counts and fixation durations for target ROI and distractor ROI were analyzed in SPSS software. A repeated measures ANOVA was performed. Two stimuli containing the target ROI and the distractor ROI were set as the within‐subject factor. The three groups, AD, aMCI, and HC, comprised the between‐subject factor. Bonferroni adjustment was used to correct the p‐value in post hoc tests.

3. RESULTS

3.1. Demographical features and neuropsychological tests

Age (F(2,99) = 2.322, p > 0.05), education (F(2,99) = 2.757, p > 0.05), and sex (χ²(2) = 2.672, p > 0.05) were not statistically different between groups (Table 1).

TABLE 1.

Demographical and clinical features of the participants.

	AD (n = 32)	aMCI (n = 37)	HC (n = 33)	p value
Education	9.06 ± 5.17	9.22 ± 5.22	11.24 ± 3.0	0.103
Age	72.81 ± 7.61	70.65 ± 6.47	68.85 ± 6.34	0.068
Sex (F/M)	15 /17	22/15	22/11	0.263
MMSE	22.03 ± 2.81	26.62 ± 1.32	29.09 ± 1.01	<0.001

Note: The values of mean and standard deviation for age, education, and MMSE scores are presented, and frequency (n) for gender is displayed.

Abbreviations: AD, Alzheimer's disease; aMCI, amnestic mild cognitive impairment; HC, healthy controls; MMSE, Mini‐Mental State Examination.

As expected, the MMMSE scores were statistically different in all groups (F(2,99) = 120.520, p < 0.001). Post hoc analysis with Bonferroni correction revealed that the MMSE scores were lower in the AD compared to the aMCI (p < 0.001) and HC (p < 0.001) groups. Similarly, the MMSE scores were lower in the aMCI than in the HC (p < 0.001) group.

3.2. Visual search

3.2.1. Total fixation durations

A repeated measures ANOVA was performed to analyze the effect of group (AD, aMCI, HC) and stimulus (target, distractor) on fixation durations (Table 2). This analysis revealed that there was a statistically significant interaction effect (F(2,99) = 22.722, p < 0.001, η _p ² = 0.317) between group and stimulus on fixation durations. The fixation durations of the AD group on the target were significantly shorter compared to the aMCI (p < 0.001) and HC (p < 0.001) groups. In addition, the fixation durations on the target were also significantly shorter in the aMCI compared to the HC (p = 0.003) group. The fixation durations of the AD, aMCI, and HC groups on the distractors were not significantly different (p > 0.05) (Figure 3).

TABLE 2.

Fixation durations and fixation counts of participants in visual searches.

					p‐value
Visual search parameter	ROI	AD	aMCI	HC	GROUP	ROI	Interaction	Post hoc
Fixation durations (ms)	Target	1461.007 ± 79.296	1952.293 ± 75.583	2313.682 ± 76.856	<0.001	<0.001	<0.001	HC–aMCI 0.003 a
	Distractor	1239.784 ± 35.453	1278.788 ± 32.451	1342.507 ± 34.362				HC–AD < 0.001 a
								aMCI–AD < 0.001 a
Fixation counts (n)	Target	2.90 ± 0.24	4.30 ± 0.22	4.79 ± 0.23	<0.001	0.001	<0.001	HC–AD < 0.001 a
	Distractor	2.12 ± 0.06	1.95 ± 0.06	1.76 ± 0.06				aMCI–AD < 0.001 a
								HC–AD 0.001 b

Note: Post hoc comparisons with Bonferroni corrections are presented for the interaction effect.

Abbreviations: AD, Alzheimer's disease; aMCI, amnestic mild cognitive impairment; HC, healthy controls; ROI, region of interest.

^a Indicates statistically significant pairwise comparisons for the target.

^b Indicates statistically significant pairwise comparisons for the distractor.

FIGURE 3.

The fixation counts (left) and fixation durations (right) on the target and distractors for the three groups. AD, Altzheimer's disease; aMCI, amnestic mild cognitive impairment; HC, health control.

There was a statistically significant main effect of group on fixation durations (F(2,99) = 27.310, p < 0.001, η _p ² = 0.358). Whatever the stimulus, the AD (1350.395 ± 46.466) group had shorter fixation durations compared to the aMCI (p < 0.001) and HC (p < 0.001) groups. Regardless of stimulus, the aMCI (1615.541 ± 42.532) patients also had shorter fixation durations than the HC (1828.095 ± 45.036, p = 0.003) group.

A statistically significant main effect of stimulus was found on fixation durations (F(2,99) = 194.262, p < 0.001, η _p ² = 0.665). Regardless of the group, fixation durations on the target (1908.994 ± 44.049) were significantly longer than on the distractor (1287.026 ± 19.694, p < 0.001).

Heat maps based on fixation durations of groups in selected trials are shown in Figure 4. Red areas indicate the highest fixation durations, and green areas show the shortest fixation durations.

FIGURE 4.

The frequency of fixation durations of the HC, aMCI, and AD groups in the visual search paradigm. Red indicates the areas with the longest fixation, and green shows the areas with the shortest fixation. The heatmap was created by combining the fixations of the groups in Data Viewer software. AD, Altzheimer's disease; aMCI, amnestic mild cognitive impairment; HC, health control.

3.2.2. Fixation counts

A significant interaction effect between groups was found on fixation counts (F(2,99) = 22.445, p < 0.001, η _p ² = 0.314) (Figure 3). The fixation counts on the target in AD patients were statistically significantly fewer than in the aMCI (p < 0.001) and HC (p < 0.001) groups. The fixation counts on the target were not statistically different between the aMCI and HC (p > 0.05) groups. The fixation counts on the distractor were significantly increased in the AD compared to the HC (p = 0.001) groups. The fixation counts on the distractor were not statistically different between the aMCI and HC groups, or between the aMCI and AD (p > 0.05) groups (Table 2).

A main effect of group on the fixation counts was revealed (F(2,99) = 10.534, p < 0.001, η _p ² = 0.177). Whatever the stimulus, AD patients (2.51 ± 0.12) had fewer fixation counts than the aMCI (3.13 ± 0.11, p = 0.002) and HC (3.27 ± 0.12, p < 0.001) groups. Regardless of the stimulus, there was no significant difference between the aMCI and HC groups on the fixation counts (p > 0.05).

A main effect of stimulus on the number of fixation counts was demonstrated (F(2,99) = 226.316, p < 0.001, η _p ² = 0.698). Bonferroni post hoc results showed that regardless of group, all participants fixed more on the target (4.0 ± 0.13) than the distractor (1.9 ± 0.03, p < 0.001).

4. DISCUSSION

Consistent with the hypothesis of our study, both AD and aMCI patients showed oculomotor changes reflecting impaired visual search compared to healthy individuals. The most important findings of our study were that the fixation durations on the target were significantly shorter in the AD compared to aMCI and HC groups, and the fixation durations on the target were also shorter in the aMCI compared to the HC group. Moreover, the number of fixations of AD patients on the target was significantly reduced compared to the aMCI and HC group, but increased on the distractor. The study by Pereira et al. ³⁰ examined the visual search task based on encoding and recognition in three trials. Abstract shapes were used in the first two trials, and colorful scenes were used in the last trial. Both AD and aMCI patients were shown to be more fixated on distractors for longer than the target. ³⁰ Although the visual search was examined during the visual‐spatial memory task in their study, unlike in our study, the findings of that study are consistent with our findings. A recent study examined the visual search performance of 18 AD patients and 18 controls by manipulating the target's template, semantic coherence, and the salience of the target and distractors. ³² It was reported that AD patients fixated on non‐target areas more and longer than the controls, and top‐down and bottom‐up guidance during visual search positively affected the search performance of patients. Top‐down and bottom‐up attention mechanisms by manipulation of the stimuli used in the study were well tested, but the study did not examine an aMCI group. ³² The most comparable study to the current visual search task is that of our previous study, Akkoyun et al. ³¹ That study presented a kitchen scene in everyday life, and the participants were asked to find relevant objects. In that study, although AD patients showed visual search changes compared to aMCI patients and healthy controls, no difference was found in the visual search performance of the aMCI and HC groups. In contrast to the results of our previous study, visual search performance between aMCI patients and older healthy individuals was found to be different in the current study. This may be because the visual search tasks were different. In our current study, there was only one target in each trial of the task, whereas in our previous study, the task included both one target and multiple targets. Altogether, these results provide evidence that visual search is impaired in the early stages of AD and even in aMCI, consistent with previous reports. ³⁰ , ³¹ , ³²

Regardless of the group, all participants looked at the target more and longer than the distractors. This seems to indicate that the target stimulus attracts more attention than the distractors. This may be related to the top‐down attention mechanism, which means directing attention to the target. However, the visual search pattern in the patient groups showed a trend in the opposite direction of this finding. The increased focus of AD patients on the distractor rather than the target is explained by excessive focus on non‐target things in the environment and difficulty in directing attention to the target. ¹⁹ , ²⁰ , ²⁵ In addition, as the number of distractors increases, the tendency to focus more on the distractors has been reported. ²⁵ , ²⁷ Consistent with our previous study, as more and longer fixations are made on distractors in AD and aMCI patients, it can naturally be expected that targeted fixations will be fewer and shorter. ³¹

Regardless of whether it was at a target or a distractor, AD patients looked at the screen less and for shorter periods of time than the aMCI and HC groups. Similarly, aMCI patients had shorter screen gazes compared to HCs. Increasing evidence of visual attention deficits points to impaired top‐down attentional control even in the early stages of AD and aMCI. ²⁶ , ⁵² , ⁵³ , ⁵⁴ , ⁵⁵ The process of attention to a visual stimulus is the directing of attention to specific areas of interest for priority and higher‐level processing. It is suggested that the visual search paradigm can be a useful method to evaluate the effectiveness of visual attention. ⁵⁶ In patients with mild to moderate AD, visual attention and visual search difficulties have been associated with bilateral parietal and right temporoparietal metabolism. ⁵⁷ In aMCI and AD, atrophy and loss of grey matter have been demonstrated in the visual cortex and brain regions responsible for higher‐level visual functions including the ventral tract (what) and the dorsal tract (where). ⁵ Moreover, Hao et al. ⁵⁸ found decreased activation in the parietal and left frontal lobes and increased activation in the right frontal lobes and the right occipitotemporal cortical regions during the visual search task in AD patients.

Our environment in everyday life is filled with distracting information from many modalities. ⁵⁶ Therefore, visual search using eye‐tracking is a popular behavioral method. ⁵⁹ Finding the target involves locating a feature which is highly salient among many stimuli, such as finding red one among the green distracting symbols, which automatically draws attention to the target. The visual search process continues until the target is found, and the higher the number of distractors the longer the visual search time. ²³ , ²⁴ , ⁶⁰ , ⁶¹ Patients with AD and aMCI and healthy individuals can find targets accurately regardless of the number of distractors. ²⁰ , ²⁷ , ⁶¹ However, in AD and aMCI, there are changes in eye movement, reaction time, and the process of dividing attention to the target and other elements during target finding. ²⁷ , ³¹ AD and aMCI patients have been reported as having a prolonged reaction time on the target in visual search. ¹⁹ , ²⁵ Consistent with our study, it has been shown that AD patients had more focus on non‐target areas and spend more time on distractors. ²³ , ²⁴ , ³⁰ , ³¹ , ³² These previous reports pointed out that AD patients have inefficient visual search. ²⁴ , ³⁰ , ⁵⁸ , ⁶¹

Accumulating evidence reports eye movement abnormalities in AD and MCI. ⁶² These studies have shown a prolongation of the latency of reflexive saccades towards a sudden stimulus and a decrease in the correct saccade rates. ⁶³ , ⁶⁴ , ⁶⁵ In addition, a decrease in correct saccade rates and an increase in uncorrected error rates were shown in antisaccade tasks performed in the opposite direction of the stimulus associated with inhibition and working memory in AD and aMCI. ⁶³ , ⁶⁶ , ⁶⁷ In a visual scanning task in which they were asked to freely view real‐world pictures, AD patients scanned less, and their visual scanning models narrowed compared to aMCI patients and healthy older subjects. ⁶⁸ Rösler et al. ²⁰ reported that patients with AD tend to focus on the center of the screen, which may be related to difficulty in diverting attention from the stimulus to peripheral stimuli. Consistent with this, in our previous study, individuals with AD increased the number of reflexive prosaccades toward stimuli with 5° of amplitude compared to the 10° of amplitude stimulus in a prosaccade task. ⁶⁴ In other words, individuals with AD can detect stimuli closer to the center faster and with higher accuracy. In the current study, target stimuli were adjusted so that they would not appear in the center due to the tendency of individuals with AD to focus on the center of the screen. All the above evidence points to oculomotor changes in AD and aMCI. These changes are associated with a disruption in the top‐down attentional mechanism. ⁵³

Alescio‐Lautier et al. ⁶⁹ suggested that impairment in visual attention in AD and aMCI develops early as an entity separate from memory decline, but visual‐spatial dysfunctions result from short‐term memory impairments. Therefore, it has been suggested that visual attention can be considered a cognitive marker with the potential to distinguish aMCI, which is highly likely to transform into AD.

It has been stated that visual search tasks using real‐world pictures and abstract shapes reflect visual search performance independent of the stimulus presented. ²⁵ , ²⁷ However, we still believe that presenting real‐world images in the visual search task and examining the visual scanning patterns on these stimuli at both individual and group levels can give an idea about the daily life difficulties of the patients. In addition, the use of these tasks to monitor visual impairments during disease progression in AD and aMCI can provide important information. Laboratory‐based behavioral studies can provide important information for future cognitive or behavioral interventions for patients.

One of the strengths of this study is the use of eye‐tracking technology, which is an objective and non‐invasive method. Another strength of this study is the use of real‐world pictures in the task, which can help with inferences about the visual difficulties that AD patients face in daily life. There was one target and eight distractors in the visual search paradigm in this study. The limitation of this study may be that multiple targets were not included in the paradigm. In future studies, the effect of increasing the number of targets on visual search can be examined.

5. CONCLUSION

This study revealed visual search changes in both AD and aMCI patients relative to healthy controls. AD patients looked more at the distractors, whereas both AD and aMCI patients looked less at the target. These results may indicate difficulty in detecting the visual target in AD and aMCI. Changes in visual search in AD and aMCI may negatively affect patients' daily life skills such as finding objects, finding a place, or cooking. Eye tracking, which can distinguish AD and aMCI patients from healthy individuals, may have the potential to be used as a cognitive biomarker. Eye tracking may provide early clues that can help identify AD/aMCI before traditional cognitive testing. Longitudinal studies with healthy individuals may reveal early visual changes in individuals likely to develop aMCI and AD.

CONFLICTS OF INTEREST

The authors report no conflicts of interest. Author disclosures are available in the supporting information.

CONSENT STATEMENT

Informed consent was obtained from all participants or their relatives in this study.

Supporting information

Supporting Information

Eraslan Boz H, Koçoğlu K, Akkoyun M, Tüfekci IY, Ekin M, Akdal G. Visual search in Alzheimer's disease and amnestic mild cognitive impairment: An eye‐tracking study. Alzheimer's Dement. 2024;20:759–768. 10.1002/alz.13478