Characteristics of Visual Cognition in Patients with Anoxic Encephalopathy: An Eye-tracking Study (2024)

ABSTRACT

Objectives : Despite the frequent occurrence of visual cognitive impairment after anoxic encephalopathy, only a few studies have analyzed gaze movements following encephalopathy. This study determined the visual cognitive characteristics of patients with anoxic encephalopathy using an eye-tracking system.

Methods : This study included ten patients with anoxic encephalopathy and ten age/sex-matched controls. Factors for anoxic encephalopathy onset and brain imaging findings were extracted from medical records. An eye-tracking system was used to track eye movements during three visual search tasks (pop-out, serial search, and saliency) in patient and healthy control groups. The average target search time, number of saccades, and number of fixations to salient stimuli were compared.

Results : Stagnant blood flow, observed in six of ten patients, was the most common cause of disease onset, four of whom exhibited hypoperfusion in bilateral occipital or parietal lobes on single-photon emission computed tomography. The patient group required longer search times during all visual search tasks and a higher number of saccades during pop-out and serial search tasks. However, no significant difference was observed between the two groups for the number of fixations to salient stimuli during the saliency task.

Conclusions : Following anoxic encephalopathy, bottom–up (pop-out task) and top–down (serial search task) gaze control were considered impaired because of extensive parieto-occipital lobe damage after blood flow stagnation. Patients exhibited reduced top–down function for finding targets (serial search task) but relatively retain inhibitory function for salient stimuli (saliency task). Gaze analysis can be used to reveal the clinical characteristics of anoxic encephalopathy.

INTRODUCTION

Anoxic encephalopathy is a general term for functional brain impairment caused by transient disruption of oxygen and glucose supply to the central nervous system. Causes of anoxic encephalopathy include cardiopulmonary arrest, respiratory failure, drowning, severe anemia, carbon monoxide poisoning, and other factors.¹⁾ The sequelae of anoxic encephalopathy include memory impairment, attention impairment, executive function impairment, and personality and behavioral changes.²^,³⁾ All of these conditions require careful evaluation and observation in clinical practice. Furthermore, anoxic encephalopathy can lead to visual cognitive impairment,²^,⁴^,⁵^,⁶⁾ manifesting as various symptoms, including Balint’s syndrome, simultanagnosia, and visual agnosia.²^,⁷^,⁸^,⁹^,¹⁰⁾ However, brain imaging studies often fail to detect focal lesions associated with visual cognitive impairment in patients with anoxic encephalopathy. In addition, neuropsychological tests lack objectivity because of the difficulties in determining the type of eye movements and whether the participant is visually perceiving the objects. Considering these clinical experiences, there may be a need to visualize eye movements following anoxic encephalopathy and objectively assess visual cognitive function.

Gaze analysis has previously been conducted for various brain diseases, including Alzheimer’s disease,¹¹⁾ Parkinson’s disease,¹²⁾ spinocerebellar degeneration,¹³⁾ posterior cortical atrophy,¹⁴^,¹⁵⁾ and brain tumors.¹⁶⁾ These studies primarily considered the control function of the human gaze, such as gaze movements [including gaze fixation and saccade responses (movements between two fixations)], bottom–up function (a response wherein the gaze is drawn to a prominent object), and top–down function (a response wherein the gaze is actively searching for an object).¹³^,¹⁵^,¹⁷^,¹⁸^,¹⁹⁾ Prolonged gazing time and reduced saccade amplitude have been reported in patients with Alzheimer’s disease,¹¹⁾ reduced saccade frequency and narrow gazing area in those with Parkinson’s disease,¹²⁾ reduced saccade amplitude and prolonged target gazing time in those with posterior cortical atrophy (sticky fixation, i.e., the inability to disengage from gazing at a target),¹⁴⁾ and eye movements such as gaze and gazing point movements¹⁶⁾ in those with brain tumors. In addition, gaze movement analysis in patients suffering from spinocerebellar degeneration and posterior cortical atrophy with visual agnosia has confirmed impaired top–down gaze control.¹³^,¹⁵⁾ However, to the best of our knowledge, no studies have attempted to visualize gaze movements using an eye-tracking system in patients with anoxic encephalopathy, a neurological disorder presenting with visual cognitive impairment.

Therefore, the current study analyzed gaze and saccade responses using an eye-tracking system to clarify the characteristics of visual cognitive function in patients with anoxic encephalopathy, focusing particularly on bottom–up and top–down gaze control. In addition, we analyzed the relationship between eye movements, such as gazing and saccade responses, based on brain imaging findings and neuropsychological tests. Herein, we hypothesized that patients with anoxic encephalopathy show prolonged search time with increased gazing and saccade responses regardless of the presence of abnormalities on brain imaging or neuropsychological tests. Patients with anoxic encephalopathy have poorer social outcomes than those with other brain injury disorders.⁵⁾ The results of the present study may help our understanding of anoxic encephalopathy and develop supportive measures to assist the reintegration of affected patients into society.

MATERIAL AND METHODS

Participants

The following inclusion criteria were used for the patient group: (1) patients with anoxic encephalopathy who visited a base hospital for higher brain dysfunction in f*ckui Prefecture during April 2021, and (2) patients who visited the hospital at least 1 year after the onset of anoxic encephalopathy. The following exclusion criteria were used: (1) patients who could not understand verbal instructions, and (2) patients who had obvious vision and visual field abnormalities that interfered with their daily life. For the healthy control group, age/sex-matched healthy participants were recruited. After recruiting individuals from April 2021 to March 2022 based on the abovementioned criteria, ten patients with anoxic encephalopathy and ten age/sex-matched healthy participants were included in this study (aged 20–72 years). The characteristics of patients with anoxic encephalopathy and healthy control participants are shown in Table 1. The mean ages (standard deviation) of patients with anoxic encephalopathy and healthy participants were 56.7 (16.6) and 56.8 (14.2) years, respectively. The median (interquartile range) time since the onset of anoxic encephalopathy among affected patients was 94.0 (78.0–186.3) months. Factors affecting the onset of anoxic encephalopathy were classified based on Fitzgerald’s classification.¹⁾ The most common factor was blood flow stagnation (BFS), which was observed in six of ten patients.

Table 1. Participant characteristics

Characteristic	Patients (n=10)	Healthy controls (n=10)	P value
Sex (male/female)	9/1	9/1	1.00
Age, years	56.7 (16.6)	56.8 (14.2)	0.99
Time since onset, months	94.0 (78.0–186.3)
Factor in onset
BFS (CA)	6
Anemia, carbon monoxide poisoning	1
Anoxia, asphyxia caused by hanging	1
Mixed mode of injury
Pulmonary embolus + BFS (CA)	1
Anoxic (drowning) + BFS (CA)	1

Data given as number, mean (standard deviation) or median (interquartile range, IQR). Intergroup comparison: Sex, chi-squared test; Age, Student's t-test. CA, cardiac arrest.

Brain imaging findings were extracted from the medical records of the included patients. Both groups underwent neuropsychological tests, including the Mini-Mental State Examination (MMSE), Japanese version of the Trail Making Test (TMT-J; Parts A and B), Visual Perception Test for Agnosia [VPTA; naming pictures (No. 8) and describing pictures of situations (No. 16)], Rey–Osterrieth Complex Figure Test (ROCFT; copying), and reaching tests (central and peripheral vision). The test results were used to assess generalized cognitive, attention, and visual cognitive functions. Furthermore, the participants were asked about the presence of visual perceptual abnormalities in their daily life.

Written informed consent was obtained from all participants. This study was approved by the Ethical Review Committee of Nittazuka Medical Welfare Center (approval no. 2021–16).

Visual Stimuli

Visual stimuli were presented on a 60-cm monitor (1920 × 1080 pixels) synchronized with an eye-tracking system. Based on previous research,²⁰^,²¹^,²²⁾ three types of visual search tasks were used for both groups (Fig. 1a–c): (1) pop-out task (searching for 1 open circle among 30 closed circles as distractors), (2) serial search task (searching for 1 closed circle among 30 open circles as distractors), and (3) saliency task (searching for 1 closed circle from 30 open circles as distractors while suppressing the response to a red open circle as a salient stimulus). The saliency task was developed for the current study. The positions of the targets and order in which the tasks were presented were randomized using a random number table.

Fig. 1.

Visual search tasks. (a) Pop-out task; (b) Serial search task; (c) Saliency task; (d) Flow of tasks. First, the target to be detected is displayed at the center of the monitor, on which the subject fixates [1]. After confirming fixation at the center of the monitor (1000 ms), the screen switches to another screen with the target and distractors, after which the subject performs a visual search from the center of the monitor to find the target among the distractors and fixates on it [2]. When the preset fixating condition (1000 ms) is satisfied, the screen switches to display the next target [3].

Based on the basic task flow shown in Fig. 1d, a practice session was first conducted to enhance understanding of the task. During the practice session, the participants were asked to search for a target among 10 distractors (three different visual search tasks, three times each). After the practice session, the participants proceeded to the main experiment, wherein they were asked to search for a target among 30 distractors (three visual search tasks, three times each). An additional experiment was conducted using the same method as the main experiment to compensate for missing data within the main experiment. Missing data indicated the failure of the eye-tracking system to adequately detect the participant’s eye movements during a visual search task.

Experimental Setup and Protocol

Eye movements during the task were measured using a screen-based eye tracker (Tobii Pro Spectrum 600 Hz Ver. 2.2.3; Tobii, Stockholm, Sweden; Fig. 2) and analyzed via biometric measurements and analysis software (Tobii Pro Lab Ver. 1.171; Tobii; Fig. 2). The distance between the screen and the patient’s eyes was ~60 cm. Nine calibration points were used, and angular velocities of greater than and less than 30°/s were defined as fixations and saccades, respectively. Subsequently, the visual search time and number of saccades until detection of the target were determined (Fig. 3a). An area of interest (AOI) was set to extract the number of fixations and saccades (Fig. 3a). Repeated fixations to the target were confirmed by extracting the number of visits to the target using the AOI (Fig. 3b). During the saliency task, an AOI was also set for salient stimuli, and the influence of the salient stimuli was extracted based on the number of fixations to the salient stimulus itself and stimuli around it (Fig. 3c). Accordingly, an AOI of 150 × 150 (width × height; similar in size to the target) was set to extract visual search time and number of saccades, whereas an AOI of 300 × 300 (width × height) was set to extract the number of visits to the target as well as the number of fixations to the salient stimulus (Fig. 3a–c).

Fig. 2.

Experimental setup of the eye tracker system. [1] Monitor displaying tasks. [2] Device used to measure eye movement (Tobii Pro Spectrum). [3] Device used to analyze the measured eye movement (Tobii Pro Lab). The height of the table is adjusted according to the participant’s viewpoint using an adjustable elevating table (right).

Fig. 3.

Gaze analysis. (a) Example analysis results from Tobii Pro Lab. [1] Fixation: the area where the line of sight is fixed. [2] Saccade: the movement of the line of sight from one gazing point to the next. [3] Area of interest (AOI): the location set for extracting the fixations and saccades recognized in the circled area. (b) Inner AOI: the AOI set for extracting fixations or saccades to the target. Outer AOI: the AOI set for extracting the number of visits near the target (number of visits). Refixation response represents a response in which fixation occurs within the outer AOI, including when the gaze returns to the target after leaving it. (c) Examples of gazing at a salient stimulus. To confirm the influence of salient stimuli during the saliency task, we set an AOI for the saliency stimuli and extracted the number of fixations within that AOI.

Statistical Analysis

Intergroup comparisons in participant background, neuropsychological tests, and gaze analysis parameters were conducted using the Chi-squared test, Student’s t-test, and Mann–Whitney U test according to data characteristics. For intertask comparisons in gaze analysis parameters both in the patient and healthy control groups, we applied Friedman’s test followed by a multiple comparison test (Holm method) when a main effect was found between the three types of tasks. The r values for the Mann–Whitney U test and multiple comparison in Friedman’s test (Wilcoxon signed-rank test) were calculated as effect size (ES) indices.²³⁾ Spearman’s rank correlation coefficient was applied to determine the relationship between gaze analysis parameters and neuropsychological tests. All statistical analyses were conducted using EZR version 1.55 (Saitama Medical Center, Jichi Medical University Hospital, Omiya, Japan) with a significance level of 5%.

RESULTS

Brain Imaging Findings

Among the ten patients in the patient group, nine underwent head computed tomography (CT) or magnetic resonance imaging (MRI), whereas six underwent cerebral perfusion single-photon emission computed tomography (SPECT). Head CT and MRI revealed brain atrophy in eight of the nine patients and basal ganglia lesions (high MRI signal in bilateral globus pallidus) in one patient. In the six patients who underwent cerebral perfusion SPECT, cerebral hypoperfusion was observed in various regions, including the frontal lobe (one patient), frontoparietal lobe (one patient), occipital lobes (two patients), occipital–parietal–temporal lobe (one patient), and thalamus/cerebellum (one patient), all bilaterally (Table 2).

Table 2. Brain imaging findings of patient group

Patient age, years	Factor in onset	Brain or cranial CT and MRI		Hypoperfusion in brain SPECT
Patient age, years	Factor in onset	Brain atrophy	Basal ganglia lesion	Frontal lobe	Parietal lobe	Temporal lobe	Occipital lobe	Thalamus	Cerebellum
20	BFS	+	−	−	−	−	+	−	−
55	BFS	+	−	−	+	+	+	−	−
63	BFS	+	−	+	−	−	−	−	−
64	BFS	+	−	−	−	−	+	−	−
67	BFS	+	−	Not examined
76	BFS	+	−	+	+	−	−	−	−
59	Anemia	−	+	Not examined
72	Anoxia	+	−	Not examined
40	Mixed mode	+	−	−	−	−	−	+	+
51	Mixed mode	Not examined		Not examined

Plus sign: finding was present; minus sign: finding was absent. For hypoperfusion in brain SPECT, all areas of decreased blood flow were observed bilaterally.

Neuropsychological Tests, Reaching Tests (Central and Peripheral Vision), and Visual Perceptual Abnormalities in Daily Life

Based on neuropsychological test results, the patient group exhibited a significantly poorer performance in MMSE, TMT-J Part A, TMT-J Part B, VPTA (No. 16), and ROCFT than the healthy control group (P=0.001, 0.002, 0.001, 0.035, and 0.030, respectively). No significant differences were observed in the performance of VPTA (No. 8) and reaching tests (central and peripheral vision). In addition, no participants reported any visual perception abnormalities in daily life (Table 3).

Table 3. Neuropsychological testing, central and peripheral vision reaching tests, and abnormalities of visual perceptual in daily life

Test	Patients (n=10)	Healthy controls (n=10)	P value
MMSE	26 (23–28)	30 (29–30)	0.001
TMT-J
Part A, s	114 (73–155)	38 (27–42)	0.002
Part B, s	220 (128–287)	55 (47–57)	0.001
VPTA
Picture naming (No.8)	0 (0–0)	0 (0–0)	0.168
Describing pictures of situations (No.16)	0 (0–2)	0 (0–0)	0.035
ROCFT	31 (22–36)	36 (36–36)	0.030
Reaching tests
Central vision (success)	10	10	1.00
Peripheral vision (success)	10	10	1.00
Abnormalities of visual perception in daily life (present/absent)	0/10	0/10	1.00

Data given as median (IQR) or number. Intergroup comparison used Mann–Whitney U test for MMSE, TMT-J, VPTA, ROCFT; chi-squared test for Reaching tests, Abnormalities of visual perception in daily life.

Gaze Analysis

The visual search times were significantly longer in the patient group than in the healthy control group for all tasks (pop-out: P=0.004, ES=−0.651; serial search: P=0.007, ES=−0.592; and saliency: P=0.019, ES=−0.524; Table 4). Intertask comparisons revealed significant differences in both patient and healthy control groups. Furthermore, multiple comparisons revealed that the patient group required significantly more visual search time for the saliency task than for the pop-out task (P=0.018, ES=0.839), whereas the healthy control group required significantly more visual search time for serial search and saliency tasks than for the pop-out task (serial search: P=0.006, ES=0.886; saliency: P=0.006, ES=0.886; Table 5).

Table 4 . Comparison of gaze analysis results between patients and healthy controls

	Task	Patients (n=10)	Healthy controls (n=10)	ES	P value
Visual search times, s	Pop-out	2.3 (1.5–4.7)	1.1 (0.7–1.3)	−0.651	0.004
	Serial search	5.2 (4.3–6.1)	3.2 (2.1–4.6)	−0.592	0.007
	Saliency	8.2 (4.6–12.8)	3.9 (2.1–4.4)	−0.524	0.019
Number of saccades	Pop-out	8.7 (7.2–10.3)	3.3 (3.3–4.9)	−0.781	<0.001
	Serial search	17.3 (14.5–20.1)	11.3 (8.0–12.3)	−0.457	0.045
	Saliency	26.0 (11.7–36.9)	12.0 (7.9–14.2)	−0.423	0.064
Number of visits	Pop-out	0.2 (0.0–0.6)	0.0 (0.0–0.0)	−0.359	0.119
	Serial search	0.5 (0.3–1.5)	0.3 (0.0–0.7)	−0.312	0.175
	Saliency	0.8 (0.3–1.8)	0.3 (0.3–0.6)	−0.347	0.131
Number of fixations to the saliency stimulus	Saliency	0.7 (0.3–1.3)	0.7 (0.3–0.9)	−0.087	0.73

Data given as median (IQR). Intergroup comparisons used Mann–Whitney U test.

Table 5. Comparison of gaze analysis task results between patients and healthy controls

	Pop-out	Serial search	Saliency	P value	Pop vs. Serial		Pop vs. Saliency		Serial vs. Saliency
	Pop-out	Serial search	Saliency	P value	P value	ES	P value	ES	P value	ES
Patients (n=10)
Visual search times, s	2.3 (1.5–4.7)	5.2 (4.3–6.1)	8.2 (4.6–12.8)	0.008	0.098	0.629	0.018	0.839	0.49	0.242
Number of saccades	8.7 (7.2–10.3)	17.3 (14.5–20.1)	26.0 (11.7–36.9)	0.025	0.039	0.725	0.029	0.790	0.38	0.306
Number of visits	0.2 (0.0–0.6)	0.5 (0.3–1.5)	0.8 (0.3–1.8)	0.042	0.074	0.682	0.062	0.713	0.73	0.091
Healthy controls (n=10)
Visual search times, s	1.1 (0.7–1.3)	3.2 (2.1–4.6)	3.9 (2.1–4.4)	<0.001	0.006	0.886	0.006	0.886	0.70	−0.145
Number of saccades	3.3 (3.3–4.9)	11.3 (8.0–12.3)	12.0 (7.9–14.2)	<0.001	0.006	0.886	0.006	0.886	1.00	0.000
Number of visits	0.0 (0.0–0.0)	0.3 (0.0–0.7)	0.3 (0.3–0.6)	0.014	0.100	0.651	0.036	0.822	0.93	−0.028

Data given as median (IQR). Multiple comparisons performed using Friedman’s test and Holm method.

The patient group required a significantly higher number of saccades for target detection than the healthy control group during pop-out and serial search tasks (pop-out: P <0.001, ES=−0.781; serial search: P=0.045, ES=−0.457; Table 4). Intertask comparisons revealed significant differences both in the patient and healthy control groups. Furthermore, based on multiple comparisons, the patient group required a significantly higher number of saccades during serial search and saliency tasks than during the pop-out task (serial search: P=0.039, ES=0.725; saliency: P=0.029, ES=0.790; Table 5). Similarly, the healthy control group required a significantly higher number of saccades during serial search and saliency tasks than during the pop-out task (serial search: P=0.006, ES=0.886; saliency: P=0.006, ES=0.886; Table 5).

For all tasks, no significant difference was noted in the number of visits between the two groups. Intertask comparisons revealed significant differences in the number of visits in both the patient and healthy control groups. Furthermore, multiple comparisons revealed no significant difference between the three tasks in the patient group; however, the healthy control group showed a significantly higher number of visits during the saliency task than during the pop-out task (P=0.036, ES=0.822; Table 5).

In each group, the visual search times, number of saccades, and number of visits were not significantly different between serial search and saliency tasks (Table 5). There was no significant difference in the number of fixations (within the AOI set) to salient stimuli (red open circle) between the two groups (Table 4).

Relationship between Gaze Analysis Parameters and Neuropsychological Test Results

In the patient group, MMSE showed a significant moderate-to-strong negative correlation with visual search times (pop-out, r=−0.79; serial search, r=−0.81; saliency, r=−0.67). TMT-J Part A showed significant strong positive correlations with visual search times (pop-out and saliency tasks; r=0.78 and 0.83, respectively) and significant moderate positive correlations with the number of saccades (pop-out and saliency tasks; r=0.66 and 0.70, respectively). A significant moderate positive correlation (r=0.76) was observed between VPTA (No. 16) and visual search times (serial search task) (Table 6).

Table 6. Correlation coefficients between gaze analysis parameters and neuropsychological tests

	Task	MMSE	TMT-J		VPTA		ROCFT
	Task	MMSE	A	B	No. 8	No. 16	ROCFT
Patients (n=10)
Visual search time	Pop-out	−0.79^*	0.78^*	0.51	0.16	0.25	0.01
	Serial search	−0.81^*	0.56	0.51	0.01	0.76^*	−0.31
	Saliency	−0.67^*	0.83^*	0.34	0.28	0.62	−0.24
Number of saccades	Pop-out	−0.57	0.66	0.39	0.06	0.13	0.05
	Serial search	−0.62	0.50	0.18	−0.03	0.03	0.34
	Saliency	−0.40	0.70^*	0.08	0.34	0.05	0.11
Healthy controls (n=10)
Visual search time	Pop-out	0.11	0.29	0.28	NA	NA	0.09
	Serial search	−0.31	0.80^*	0.42	NA	NA	0.11
	Saliency	−0.80^*	0.33	0.37	NA	NA	−0.51
Number of saccades	Pop-out	0	−0.03	0.38	NA	NA	0.31
	Serial search	−0.49	0.41	0.40	NA	NA	−0.06
	Saliency	−0.73^*	0.08	0.23	NA	NA	−0.37

^*P<0.05; NA, not available.

In the healthy control group, MMSE showed a significant strong negative correlation with visual search times (saliency task; r=−0.80) and a significant moderate negative correlation with the number of saccades (saliency task; r=−0.73). A significant strong positive correlation was noted between TMT-J Part A and visual search times (serial search task; r=0.80) (Table 6).

DISCUSSION

Based on the analysis of eye movements during visual search tasks in patients with anoxic encephalopathy, the current study highlights the following findings. First, compared with healthy participants, patients with anoxic encephalopathy required significantly longer visual search times for all tasks and a significantly higher number of saccades for pop-out and serial search tasks. Second, no significant difference was observed between the two groups for the number of fixations to the salient stimulus during the saliency task. Third, in both groups, the saliency task or both serial search and saliency tasks required significantly longer visual search times and a significantly higher number of saccades than the pop-out task. However, within each group, there was no significant difference in gaze data for the serial search and saliency tasks. Fourth, compared with healthy participants, patients with anoxic encephalopathy exhibited significantly poorer performance in MMSE, TMT-J Part A, TMT-J Part B, VPTA (No. 16), and ROCFT. In addition, compared with the healthy control group, significant correlations were observed between MMSE and visual search times (for all tasks) and between TMT-J Part A and visual search times/number of saccades (for pop-out and saliency tasks) in the patient group.

Visual Cognitive Functional Characteristics of Patients with Anoxic Encephalopathy

Compared with healthy participants, patients with anoxic encephalopathy required significantly more visual search time for all tasks and a significantly higher number of saccades during pop-out and serial search tasks. Previous studies have confirmed the relationship between the visual search task used in this study and gaze control function. In gaze analysis, the pop-out task reflects the bottom–up function (i.e., a response wherein the gaze is drawn to a salient part), whereas the serial search task reflects top–down function (i.e., a response wherein the gaze is actively searching for a specific target).¹³^,²⁰^,²¹^,²²⁾ In the field of neurology, gaze analysis studies have revealed that patients with Alzheimer’s disease, spinocerebellar degeneration, and visual agnosia caused by posterior cortical atrophy exhibit impaired top–down function but have preserved bottom–up function.¹¹^,¹³^,¹⁵⁾ In the current study, patients with anoxic encephalopathy showed significantly longer visual search times and greater number of saccades during the pop-out and serial search tasks than healthy participants, suggesting that bottom–up and top–down functions were impaired.

Therefore, the current study focused on the areas of reduced brain function following anoxic encephalopathy as a factor influencing the reduction of both top–down and bottom–up functions. Six of the ten patients with anoxic encephalopathy included in the current study suffered from BFS following cardiopulmonary arrest caused by myocardial infarction or ventricular fibrillation. Among these six patients, four of the five patients who underwent cerebral perfusion SPECT exhibited reduced blood flow in bilateral occipital or parietal lobes. Parieto-occipital lobe injury reportedly occurs following cardiac surgery and BFS because of hypotension.²⁴^,²⁵⁾ In the present study, patients with anoxic encephalopathy often experienced BFS, which may have caused extensive parieto-occipital lobe damage. From the perspective of visual information processing, bottom–up function is localized to the inferior parietal lobe, whereas and top–down function is localized to the superior parietal lobe,²⁶⁾ both of which may have been impaired because of extensive parieto-occipital lobe damage. The current study also included two patients with anoxic encephalopathy who had reduced blood flow to bilateral frontal lobes. The frontal lobes are responsible for executive function.²⁷⁾ A previous gaze analysis study reported that older participants with impaired executive function have difficulty selecting both functions of gaze control appropriately according to the characteristics of visual search tasks.²⁸⁾ It is possible that patients with anoxic encephalopathy in the present study not only suffered direct damage to both gaze control functions because of parieto-occipital lobe damage but also that frontal lobe damage prevented them from flexibly selecting between gaze functions.

In real-life settings, both bottom–up function, wherein the gaze is drawn to the most prominent feature, and top–down function, wherein the gaze is actively directed to a specific target, are often mixed.¹⁹⁾ In such situations where both functions are mixed, top–down function has been suggested to predominate.²⁹⁾ Conversely, a reported case of visual agnosia caused by posterior cortical atrophy indicated that in such situations, the participant becomes dependent on bottom–up cues and cannot perform top–down visual searches.¹⁵⁾ Several cases of visual agnosia have been reported in patients with anoxic encephalopathy caused by carbon monoxide poisoning.⁷^,⁸^,⁹^,¹⁰⁾ In the current study, a saliency task was performed to analyze gaze in situations wherein both top–down and bottom–up functions were mixed. Our results revealed that patients with anoxic encephalopathy required significantly longer visual search times than healthy participants. However, no significant difference was noted in the number of fixations to the salient stimuli during the saliency task between the two groups, suggesting that the prolonged visual search times could not be attributed to the influence of salient stimuli. In other words, we believe that our results were not in line with those reported previously.¹⁵⁾ The same could be assumed considering the lack of significant differences in all gaze analysis parameters between the serial search and saliency tasks in the patient and healthy control groups. In terms of gaze control and attentional function, top–down function has two described functions: (1) to direct gaze and attention toward a target object¹⁵^,¹⁹⁾ and (2) to inhibit unimportant objects.³⁰^,³¹⁾ It was suggested that patients with anoxic encephalopathy exhibit retained top–down function for inhibiting salient stimuli but impaired top–down function for locating objects.

In the current study, only the patient with anoxic encephalopathy caused by carbon monoxide poisoning had lesions in both basal ganglia (globus pallidus), and the patient did not have visual agnosia. This suggests differences in the areas of reduced brain function between cases of anoxic encephalopathy caused by carbon monoxide poisoning, as reported previously,⁷^,⁸^,⁹^,¹⁰⁾ and those caused by BFS, which accounted for the majority of patients with anoxic encephalopathy in the current study as well as differences in visual cognitive functional characteristics. In addition, unlike anoxic encephalopathy caused by ischemia, which accounted for many of the participants included in this study, anoxic encephalopathy caused by oxygen deprivation has been reported to cause irreversible neuronal damage in a short time,³²⁾ suggesting greater severity. Studies have shown that anoxic encephalopathy caused by ischemia causes simultanagnogia³³^,³⁴⁾ and that anoxic encephalopathy caused by oxygen deprivation, such as carbon monoxide poisoning, causes apperceptive visual agnosia, the most severe type of visual agnosia, in which participants cannot recognize objects.⁷^,⁸^,⁹^,¹⁰^,³⁵⁾ This suggests that the type and severity of visual cognitive impairment may differ depending on the factors causing anoxic encephalopathy. Therefore, in clinical practice, visual cognitive function should be examined closely, accounting for factors that cause anoxic encephalopathy and areas of reduced brain function following anoxic encephalopathy.

In the pop-out task, the target stimulus (open circle) appeared to stand out among the interfering stimuli (closed circles), thereby facilitating the search. Conversely, in the serial search task, the target stimulus (closed circle) did not appear to stand out among the interfering stimuli (open circles) and required more attention and exploration for detection, making the search more challenging.²¹^,²²⁾ This change in visual search times after swapping the target and interfering stimuli has been described as “search asymmetry.”²⁰^,³⁶⁾ The saliency task used in the current study was developed based on the serial search task. Herein, the saliency task or both serial search and saliency tasks required significantly more visual search time and a significantly higher number of saccades than the pop-out task. This can be considered search asymmetry. Our results suggest that not only healthy participants with intact bottom–up and top–down functions but also patients with anoxic encephalopathy, in whom both bottom–up and top–down gaze control are impaired, can use the bottom–up rather than top–down function to search for objects with shorter visual search times and fewer saccades. This finding could contribute to environmental adjustments, such as making objects stand out.

Relationship between Gaze Analysis Parameters and Neuropsychological Test Results

Our findings revealed that patients with anoxic encephalopathy exhibited significantly lower MMSE scores than healthy participants, which were correlated with visual search times in all three tasks. The MMSE has been used to elucidate general cognitive function.³⁷⁾ We cannot rule out the influence of general cognitive decline on bottom–up and top–down gaze control in patients with anoxic encephalopathy, which may have interfered with visual searching. In addition, patients with anoxic encephalopathy displayed significantly poorer performance in TMT-J Parts A and B than healthy participants. Furthermore, only TMT-J Part A correlated with search time and number of saccades in the pop-out and saliency tasks. TMT-J Part A is similar to the visual search task used in this study because it measures visual search ability and information processing speed³⁸⁾; moreover, it is considered to be correlated with gaze analysis parameters in patients with anoxic encephalopathy. However, TMT-J Part B is considered to measure mental flexibility and executive function,³⁹^,⁴⁰^,⁴¹⁾ which are unlikely to be reflected in the visual search task used in the current study.

Limitations

This study had several notable limitations. First, most anoxic encephalopathy cases included in the present study were caused by BFS. Considering that the visual cognitive characteristics of patients with anoxic encephalopathy cannot be excluded from influencing the etiopathogenic factor, further analyses on more cases with other factors are needed. Second, the present study revealed no significant difference in the number of fixations to salient stimuli between the two groups during the saliency task, suggesting that salient stimuli failed to influence visual searching in patients with anoxic encephalopathy. However, previous research⁴²⁾ has shown that the appearance of red-colored interfering stimuli (salient stimuli) unrelated to the target stimuli decreases search efficiency. Hence, the method of presenting salient stimuli should also be considered. Third, patients with anoxic encephalopathy in this study had lower general cognitive function than healthy participants, which may have affected gaze analysis parameters. In Alzheimer’s disease, which causes generalized cognitive decline, the top–down function of gaze control is disrupted, whereas the bottom–up function is preserved.¹¹⁾ However, patients with anoxic encephalopathy in the present study differed from those with Alzheimer’s disease in that both functions were disrupted. Although we were unable to establish a different disease population in this study, future studies should examine patients in different disease groups, such as Alzheimer’s disease or diffuse axonal injury, who have extensive brain damage similar to that in patients with anoxic encephalopathy.

CONCLUSION

This study examined the visual cognitive characteristics of patients with anoxic encephalopathy using an eye-tracking system. The results revealed that patients with anoxic encephalopathy exhibited deficits in both bottom–up and top–down functions of gaze control. Therefore, gaze analysis may be a useful tool for elucidating the clinical characteristics of anoxic encephalopathy.

ACKNOWLEDGMENTS

The authors thank Tobii Technology for their advice on use of the eye tracker equipment. The authors thank Enago (www.enago.jp) for the English language review of a draft version of the manuscript.

CONFLICTS OF INTEREST

The authors declare no conflict of interest.

REFERENCES

1.FitzGerald A, Aditya H, Prior A, McNeill E, Pentland B: Anoxic brain injury: clinical patterns and functional outcomes. A study of 93 cases. Brain Inj 2010;24:1311–1323. PMID:20722503, DOI:10.3109/02699052.2010.506864
2.Caine D, Watson JD: Neuropsychological and neuropathological sequelae of cerebral anoxia: a critical review. J Int Neuropsychol Soc 2000;6:86–99. PMID:10761372, DOI:10.1017/S1355617700611116
3.Urakami Y: Rehabilitation of anoxic encephalic patients: role of medical rehabilitation [in Japanese]. J Clin Rehabil 2013;22:693–699.
4.Wilson BA: Cognitive functioning of adult survivors of cerebral hypoxia. Brain Inj 1996;10:863–874. PMID:8939305, DOI:10.1080/026990596123846
5.Urakami Y: Rehabilitation of anoxic encephalopathy patients: epidemiology, pathology, symptoms, and prognosis [in Japanese]. J Clin Rehabil 2013;22:580–586.
6.Brownlee NN, Wilson FC, Curran DB, Lyttle N, McCann JP: Neurocognitive outcomes in adults following cerebral hypoxia: a systematic literature review. NeuroRehabilitation 2020;47:83–97. PMID:32716324, DOI:10.3233/NRE-203135
7.Benson DF, Greenberg JP: Visual form agnosia. A specific defect in visual discrimination. Arch Neurol 1969;20:82–89. PMID:4303441, DOI:10.1001/archneur.1969.00480070092010
8.Campion J, Latto R: Apperceptive agnosia due to carbon monoxide poisoning. An interpretation based on critical band masking from disseminated lesions. Behav Brain Res 1985;15:227–240. PMID:4005031, DOI:10.1016/0166-4328(85)90177-9
9.Milner AD, Perrett DI, Johnston RS, Benson PJ, Jordan TR, Heeley DW, Bettucci D, Mortara F, Mutani R, Terazzi E, Davidson DL: Perception and action in ‘visual form agnosia’. Brain 1991;114:405–428. PMID:2004249, DOI:10.1093/brain/114.1.405
10.Sparr SA, Jay M, Drislane FW, Venna N: A historic case of visual agnosia revisited after 40 years. Brain 1991;114:789–800. PMID:2043949, DOI:10.1093/brain/114.2.789
11.Mosimann UP, Felblinger J, Ballinari P, Hess CW, Müri RM: Visual exploration behaviour during clock reading in Alzheimer’s disease. Brain 2004;127:431–438. PMID:14691059, DOI:10.1093/brain/awh051
12.Matsumoto H, Terao Y, Furubayashi T, Yugeta A, f*ckuda H, Emoto M, Hanajima R, Ugawa Y: Small saccades restrict visual scanning area in Parkinson’s disease. Mov Disord 2011;26:1619–1626. PMID:21449014, DOI:10.1002/mds.23683
13.Matsuda S, Matsumoto H, Furubayashi T, f*ckuda H, Emoto M, Hanajima R, Tsuji S, Ugawa Y, Terao Y: Top-down but not bottom-up visual scanning is affected in hereditary pure cerebellar ataxia. PLoS One 2014;9:e116181. PMID:25545148, DOI:10.1371/journal.pone.0116181
14.Shakespeare TJ, Kaski D, Yong KX, Paterson RW, Slattery CF, Ryan NS, Schott JM, Crutch SJ: Abnormalities of fixation, saccade and pursuit in posterior cortical atrophy. Brain 2015;138:1976–1991. PMID:25895507, DOI:10.1093/brain/awv103
15.Foulsham T, Barton JJ, Kingstone A, Dewhurst R, Underwood G: Fixation and saliency during search of natural scenes: the case of visual agnosia. Neuropsychologia 2009;47:1994–2003. PMID:19428433, DOI:10.1016/j.neuropsychologia.2009.03.013
16.Tabata A, Tanimukai H, Ueda K, Yamawaki R, Murai T: Utility of an eye-tracking assessment in understanding the postoperative symptoms of a brain tumor patient with Bálintʼs syndrome: a case study [in Japanese]. High Brain Funct Res 2020;40:227–234. DOI:10.2496/hbfr.40.227
17.Henderson J: Human gaze control during real-world scene perception. Trends Cogn Sci 2003;7:498–504. PMID:14585447, DOI:10.1016/j.tics.2003.09.006
18.Henderson JM, Brockmole JR, Castelhano MS, Mack M: Visual saliency does not account for eye movements during visual search in real-world scenes. In: Van Gompel RPG, Fischer MH, Murray WS, Hill RL, editors. Eye movements: a window on mind and brain. 2007; pp. 537–562. https://doi.org/10.1016/B978-008044980-7/50027-6.
19.Terao Y: Studying visual perception by eye tracking in neurological disorders [in Japanese]. Neurol Med 2015;83:23–28.
20.Treisman A, Souther J: Search asymmetry: a diagnostic for preattentive processing of separable features. J Exp Psychol Gen 1985;114:285–310. PMID:3161978, DOI:10.1037/0096-3445.114.3.285
21.Maekawa T, Anderson SJ, de Brecht M, Yamagishi N: The effect of mood state on visual search times for detecting a target in noise: an application of smartphone technology. PLoS One 2018;13:e0195865. PMID:29664952, DOI:10.1371/journal.pone.0195865
22.Sakai M, Yonami R, Murakami J, Mitsugami I, Yamagishi N: The effect of mood state on visual search times under VR environment [in Japanese]. J Virtual Real Soc Jpn 2021;26:52–61. DOI:10.18974/tvrsj.26.1_52
23.Field A: Discovering statistics using IBM SPSS Statistics, 3rd ed. London: Sage; 2009; pp. 56–698.
24.Malone M, Prior P, Scholtz CL: Brain damage after cardiopulmonary by-pass: correlations between neurophysiological and neuropathological findings. J Neurol Neurosurg Psychiatry 1981;44:924–931. PMID:7310410, DOI:10.1136/jnnp.44.10.924
25.Barbut D, Grassineau D, Lis E, Heier L, Hartman GS, Isom OW: Posterior distribution of infarcts in strokes related to cardiac operations. Ann Thorac Surg 1998;65:1656–1659. PMID:9647076, DOI:10.1016/S0003-4975(98)00272-0
26.Shomstein S, Lee J, Behrmann M: Top-down and bottom-up attentional guidance: investigating the role of the dorsal and ventral parietal cortices. Exp Brain Res 2010;206:197–208. PMID:20571784, DOI:10.1007/s00221-010-2326-z
27.Gilbert SJ, Burgess PW: Executive function. Curr Biol 2008;18:R110–R114. PMID:18269902, DOI:10.1016/j.cub.2007.12.014
28.Kawagoe T: Dependence on frontal lobe function in behavior in old age [in Japanese]. Dissertation, Kumamoto University. 2016. https://irdb.nii.ac.jp/01242/0003289312.
29.Chen X, Zelinsky GJ: Real-world visual search is dominated by top-down guidance. Vision Res 2006;46:4118–4133. PMID:17005231, DOI:10.1016/j.visres.2006.08.008
30.Bundesen C: A theory of visual attention. Psychol Rev 1990;97:523–547. PMID:2247540, DOI:10.1037/0033-295X.97.4.523
31.Wolfe JM: Guided Search 2.0 a revised model of visual search. Psychon Bull Rev 1994;1:202–238. PMID:24203471, DOI:10.3758/BF03200774
32.Omata N, Murata T, Fujibayashi Y, Waki A, Sadato N, Yoshimoto M, Wada Y, Yonekura Y: Hypoxic but not ischemic neurotoxicity of free radicals revealed by dynamic changes in glucose metabolism of fresh rat brain slices on positron autoradiography. J Cereb Blood Flow Metab 2000;20:350–358. PMID:10698073, DOI:10.1097/00004647-200002000-00017
33.Rizzo M, Robin DA: Simultanagnosia: a defect of sustained attention yields insights on visual information processing. Neurology 1990;40:447–455. PMID:2314587, DOI:10.1212/WNL.40.3_Part_1.447
34.Wang R, Hopping G, Al Othman B, Tabba S, Kini A, Lee AG: Simultanagnosia as the presenting symptom in neuro-ophthalmology. Proc Bayl Univ Med Cent 2020;33:105–107. PMID:32063790, DOI:10.1080/08998280.2019.1656013
35.Hirayama K, Suzuki Y, Kinai M: Apperceptive visual agnosia [in Japanese]. In: Hirayama K, editor. Understanding and medical examination higher brain dysfunction, 1st ed. Tokyo: Cyugaiigakusya; 2017. pp. 76–79.
36.Treisman AM, Gelade G: A feature-integration theory of attention. Cogn Psychol 1980;12:97–136. PMID:7351125, DOI:10.1016/0010-0285(80)90005-5
37.Folstein MF, Folstein SE, McHugh PR: “Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975;12:189–198. PMID:1202204, DOI:10.1016/0022-3956(75)90026-6
38.Sánchez-Cubillo I, Periáñez JA, Adrover-Roig D, Rodríguez-Sánchez JM, Ríos-Lago M, Tirapu J, Barceló F: Construct validity of the Trail Making Test: role of task-switching, working memory, inhibition/interference control, and visuomotor abilities. J Int Neuropsychol Soc 2009;15:438–450. PMID:19402930, DOI:10.1017/S1355617709090626
39.Arbuthnott K, Frank J: Trail Making Test, part B as a measure of executive control: validation using a set-switching paradigm. J Clin Exp Neuropsychol 2000;22:518–528. PMID:10923061, DOI:10.1076/1380-3395(200008)22:4;1-0;FT518
40.Kortte KB, Horner MD, Windham WK: The Trail Making Test, part B: cognitive flexibility or ability to maintain set? Appl Neuropsychol 2002;9:106–109. PMID:12214820, DOI:10.1207/S15324826AN0902_5
41.Tombaugh T: Trail Making Test A and B: normative data stratified by age and education. Arch Clin Neuropsychol 2004;19:203–214. PMID:15010086, DOI:10.1016/S0887-6177(03)00039-8
42.Theeuwes J: Perceptual selectivity for color and form. Percept Psychophys 1992;51:599–606. PMID:1620571, DOI:10.3758/BF03211656