Eye Movements: Clinical Disorders

What are the two ways in which eye movements can provide insights into clinical disorders?

• Characteristics of the eye movements themselves (fixation lengths, movement from stimuli, etc)

• The locations people look

Can be applied by inclusion in diagnostic criteria or treatment protocols

Different disorders have different characteristics

Think: People with ASD less likely to fixate on faces (diagnostic) VS attentional control training to correct attentional biases in GAD (treatment)?

What are the benefits of using eye movement tracking in clinical research?

Affords benefits such as:

• Easy to use

• Affordable

• Accessible

• Quick

• Non-invasive

Saccades

Overt, ballistic eye movements aimed at bringing a target into foveal vision

• Can be internally or externally guided

• Remember the fovea is a tiny spot int he centre of visual fields

• Saccadic eye movements work by capturing a broad range of visual information to be able to understand the whole environment and stimuli, rather than just fixating on that tiny proportion of the visual field

Think: Like when you can send a message or photo on text with that effect where you rub out the blurry overlay to see the whole thing underneath. In this case, your finger is working as a saccadic eye movement

External Saccades

Saccades that are guided by an instruction, task or salient cue

E.g. Participants are instructed to look at a visual stimulus as soon as it appears OR a red flashing light captures attention

Think: I have been summoned

Internal Saccades

Saccades that are executed in the absence of a visual stimulus/target or instruction

• Top down processes direct the saccades

• YOU are directing your eye movements to a spot

Think: I am searching for visual information because I want it

Pursuit

Eye movements that tracking stimuli that move relatively slowly and smoothly

• Smooth, non-ballistic movements

• Match the velocity of gaze to the velocity of the target

• Purpose is to keep a target in foveal vision

• System can be divided into initiation and maintenance phases

Think: Watching a plane fly across the sky

What does the ‘integrity of pursuit’ refer to?

The ease at which a pursuit eye movement matched the velocity of a moving target

• Essentially the mean velocity of eye movement divided by the mean target velocity

• Low gain scores indicate difficulty matching gaze velocity to target velocity (because it would suggest the target was much faster than the eye)

Think: If they can stay on track with the moving target, that is, keeping it in foveal vision, then the pursuit has a high integrity

Gain

A ratio of the velocity of a person’s pursuit eye movements to the velocity of the target itself

• Measures integrity of pursuit

• Mean eye velocity / mean target velocity

• 1 would be a perfect score, meaning the eye movements matched their velocity to keep the target in foveal vision (high integrity)

• Close to zero would mean their eye movements were much slower than the target, and they struggled to keep up (low integrity)

• If gain is low, people may engage in more compensatory saccades or intrusive saccades

Compensatory Saccades

Saccadic eye movements used to catch up to moving targets when integrity of pursuit/gain is low

Intrusive Saccades

Anticipatory saccades or pares of saccades that move away from the target and then back

Think: Doing something illegal and glancing back at the door in anticipation that someone will walk in

Two divisions of the pursuit system

• Initiation - picking up on the stimulus to be pursued

• Maintenance - matching and maintaining the velocity of the moving object with eyes

Therefore, any issues with pursuits could be attributed to a failure or deficiency in one of these.

Either you don’t catch the object to pursue, or you catch it but can’t keep up with it

State 4 metrics of fixations

• Location

• Duration

• Sequences of fixations

• Distance between fixations

Pupillary Dilation

The diameter of the pupil varies as a function of:

• Light

• Task-specific recruitment of cognitive resources

• General factors (arousal, stress, anxiety, etc)

• The onset of stimuli responses

They are sensitive to working memory load on tasks

• With each increase in working memory load, the pupil dilates linearly

How do we know that pupillary dilation is not a passive process?

They dilate to expected stimuli being absent

Think: There is something top-down and preparatory going on

How does cognition relate to pupillary dilation?

Increasing working memory load causes linear pupil dilation

Think: They trained people with locked-in syndrome to give dichotomous answers by doing mental arithmetic to dilate their pupils, as it increased working memory load.

Describe differences in eye movement behaviour for people with

Schizophrenia

• Reduced smooth pursuit gain

  • AKA low integrity of pursuit

  • They struggle to keep up with tracking moving objects and matching the velocity of their eye movements to the object

• The frequency of saccades during smooth pursuit is increased

  • Makes their ‘smooth’ pursuit less smooth (jerky eye movements rather than smooth)

• Increased anti-saccade error rates and latencies

  • I.e. looking at certain stimuli when you’re not supposed to, and taking longer to get to the target stimulus

• Increase in the frequency of catch-up saccades

  • Their pursuits lag behind so they often have to reposition their gaze back on the target

• Changes in saccadic dynamics, rates of intrusive saccades such as anticipatory saccades, increasing the mismatch between eye position and target position

  • Basically speaking more to the idea that they struggle to keep their gaze on the moving target because other eye movements redirect them

• Differences in fixation patterns of static scene viewing

Think: They have a hard time following the task at hand - WHAT’S COMING oh fuck I went too far WAIT WHERE DID IT GO

What is happening here?

This is a demonstration of a smooth pursuit of parents of a person with schizophrenia

• When the eye position matches the target position, smooth pursuit is occurring

• When the eye position dips below the target at the beginning, this is an anticipatory saccade (looking beyond the target)

• When the eye position goes above the target and then quickly returns, this is a catch up saccade (the lagged behind and need to quickly get back)

The presence of these intrusive saccades may indicate genetic predisposition even though they don’t have schizophrenia (parents can exhibit biomarkers of schizophrenia)

• Observed as an increase in the total number of eye movements because of these anticipatory and catch-up saccades

• This is observed in half of children with schizophrenic parents too

Think: Anticipatory quickly moves AWAY from the target (sharp change away) whereas catch ups quickly move back (sharp change towards)

Takeaway: Eye movements can be used to identify biomarkers which can inform diagnosis and sometimes even heritability factors for clinical disorders such as schizophrenia

Describe the difference between eye movement and smooth pursuits of typically developed (TD) 15 year olds VS schizophrenic 15 year olds as found by Jacobsen et al. (1996)

There are significant differences

• On the left, you can see that the TD child is far more accurate in smoothly following the target. Apart from maybe a tiny catch-up saccade when the target first changed direction, it is largely smooth

• On the right, the eye movements of the schizophrenic child is sporadic and they spend less time fixated on the target. There are significantly more intrusive saccades (anticipatory and catchup). They really can’t smoothly track the target.

Findings from this study contributed to existing support that there is significant continuity between adult and childhood onset forms of schizophrenia

Interpret this graph by Ross

Lower number of saccades = better

• typically developing (TD) children exhibit the lowest number of saccadic eye movements, which closely resembles TD adults

  • Demonstrates continuity (i.e. if eye movements are typical in childhood, it is likely they will be TD adults)

• Childhood onset schizophrenics display the highest number of saccadic eye movements

• The most important part of this graph is the resemblance between non-psychotic children of schizophrenic parents (Think: they have genetic predisposition and remember the diathesis stress hypothesis!) and adult-onset schizophrenics

  • Again, continuity → predisposed children’s eye movements may be biomarkers for schizophrenia that will onset later in life!

Means we need to keep an eye on children who are predisposed and are exhibiting these eye movement patterns, and perhaps put some extra supports and measures in place because they are showing the risk signs of developing later-onset schizophrenia!

Describe this study by Karatekin and Asarnow (1999)

They compared eye movements between schizophrenic children and typically developing children when asked to view a scene

• They were given:

  • global questions (i.e. what is happening in the picture?)

    • This required them to integrate all information given in the scene

  • Focal questions (I.e. how old do you think each person is?)

    • This required information from 2-3 places

  • Counting question (i.e. count the number of cups on the table)

The identified spatial areas of the pictures that were either relevant or irrelevant based on the task (ie. looking at the heads would be irrelevant when asked to count cups)

They then examined the % of regions (relevant or irrelevant) that were fixated on in each question

They did this because they hypothesised that the intrusive eye movements as seen in smooth pursuits may influence scene viewing in schizophrenics too

There were no elevated looks to irrelevant locations in any task in schizophrenics compared to TD children.

Therefore, whilst saccadic behaviours are affected by schizophrenia in smooth pursuits, scene viewing is unaffected

A child with schizophrenia is asked to describe what is happening in a picture. They are then asked to follow a moving target on a screen.

Describe what we can expect their performance to be like in each task, providing evidence as to why

Children with schizophrenia experience a greater number of intrusive saccadic eye movements, but this is only apparent in certain contexts. People with schizophrenia have difficulties specifically with smooth pursuits, which is a smooth and continuous eye movement used to track moving objects. Studies comparing TD children and schizophrenic children have demonstrated that whilst TD children can closely match the direction and velocity of targets, schizophrenic children exhibit a greater number of anticipatory and catch-up saccades and spend less time fixated on the object (Jacobsen et al., 1996; Ross et al., 1999). Therefore, their performance in the target tracking task would likely be poor.

However, these intrusive saccades do not appear to influence scene viewing. Karatekin and Asarnow (1999) compared eye movements and fixations between TD children and children with schizophrenia when asked to derive specific information from a visual scene. They found that schizophrenics did not look at task irrelevant information significantly longer than TD children, refuting their hypothesis that intrusive saccades may affect their performance. Therefore, schizophrenia does not appear to influence performance in scene viewing and they would be able to complete this task.

Describe Delerue et al.’s (2013) study

They conducted a sandwich making task with individuals with schizophrenia and healthy controls to see if there were any differences in gaze and fixation patterns in free viewing (looking at a kitchen bench with task relevant and irrelevant objects) vs active viewing (making the sandwich in that scene) in familiar VS unfamiliar tasks

• People with schizophrenia looked (fixations and gazes) at both task relevant and irrelevant objects significantly longer than healthy controls ONLY IN THE FREE VIEWING CONDITION FOR THE FAMILIAR TASK (sandwich making)

  • Think: They be staring at it all for ages when there is no task demand, but when they start doing the task, they don’t give a fuck about irrelevant objects anymore

• This was inverted when the task was unfamiliar: in the active viewing task, they fixated and gazed upon both relevant and irrelevant objects significantly more than healthy controls

  • May be because they were making less look-ahead fixations (cannot plan unfamiliar tasks)

  • There were no differences in fixations between HC and schizophrenics in free viewing unfamiliar tasks.

Takeaway: Schizophrenics be fixating. When the task is familiar they just want to take it all in when they’re sitting back and doing nothing, but when they do the task they lock in. When it is unfamiliar, they need to know EVERYTHING even if its irrelevant to the task which makes everything longer; even motor initiation of the task.

They be freaking out in unfamiliar tasks, which is proposed to be due to a planning deficit

Describe Dowiasch et al.’s (2016) Study

Compared eye movements of schizophrenics VS healthy controls in 4 natural tasks using a mobile eye tracker:

1. Fixating on stationary objects in a waiting room

2. Sitting in a hallway with free gaze

3. Walking down the hallway

4. Visually tracking targets on the floor while walking straight ahead

Results:

• When they had to fixate on predefined targets in a freely chosen, random order, the patient’s fixations were significantly shorter in duration, but occurred more frequently than healthy controls (think: that gif where she’s eating chips and looking side to side)

• During free gaze (i.e. sitting aimlessly in a hallway), this was inverse and they fixated for LONGER, but less frequently than HCs.

• However, the median fixation duration did not differ between the two groups

Takeaway: There are significant differences in fundamental oculomotor parameters between schizophrenic patients and HCs during NATURAL behaviour in a REAL environment

• Not just in lab settings!

• Also shows they might have some compensatory strategies in natural settings to overcome deficits observed in lab studies, but little is known about how

• Meaning, the bleak results from previous lab studies probably don’t translate into the real world (good news)

Describe differences in eye movements for people with ADHD

Context: There are fronto-striatal deficits that make it difficult for these individuals to inhibit reflexive responses, leading to hypotheses that people with ADHD may have some more crazy eye movements going on - not really the case!

The Jacobsen et al. (1996) study showed that whilst there are differences between ADHD and HC, the differences are not as significant as schizophrenics.

• Can do smooth pursuits fine (image)

• Struggles with saccadic inhibition (Munoz et al., 2003)

Think: They just can’t stop themselves from looking!

Who would do better in smooth pursuit tasks?

ADHD or schizophrenics

ADHD

Describe Munoz et al.’s (2003) study

Tested performance for people with ADHD in oculomotor tasks that require suppression of reflective or unwanted saccadic eye movements

• Measured eye movements for ADHD and HCs in pro- and anti-saccade tasks in 9-59yos

  • Pro saccade: they look at a central fixation point and instructed to look at a target to the side when it pops up (reflexive)

    • Think: Look at it!!! Woooo!!!!

  • Anti-saccade: they must suppress the saccade and look to the opposite side of the fixation point away from the target.

    • Think: MIND YOUR BUSINESS

Found that people with ADHD make significantly more errors than HCs in this task with more intrusive saccades (rate of saccades/sec)

Describe the results of Munoz et al.’s (2003) study

Pro-saccade task:

• People with ADHD had:

  • Long reaction times

  • Greater intra-subject variance

  • Saccades has reduced peak velocities and increased durations

Summary: When ADHDs were allowed and encouraged to look at the target, they were much slower with their eye movements and reactions, and their individual performance was inconsistence

Anti-saccade task:

• People with ADHD had:

  • Greater difficulty suppressing reflexive pro-saccades toward the target

  • Increased reaction times for correct anti-saccades

  • Greater intra-subject variance in a third task requiring prolonged fixation

Summary: It took them more effort and time to suppress the reflex to look at the target than HCs (they struggled to inhibit basically)

In general:

• ADHD had more intrusive saccades during periods they should have been fixating on the central area (naughty)

Takeaway: ADHD Px have reduced ability to suppress unwanted saccades and control their fixation behaviour voluntarily. This is consistent with a fronto-striatal pathophysiology

Think: It is not that they can’t do it, it's more that its harder to voluntarily control

William’s Syndrome

Genetic disorder characterised by:

• Disassociation of higher cognitive functions

• Distinct social phenotype

• Drive toward social behaviour, often labeled as ‘hyper-sociability’ or ‘pro-social compulsion’

They demonstrate

• Desire to interact regardless of familiarity

• Hold prolonged face gaze during interactions

• Atypical interpretations of facial cues for emotion and identity

Prevalence is 1 in 20 000 people

Think: Anti-autism

Describe Ruby & Hancock’s (2009) study (was a reading)

Compared how faces capture the attention of people with autism vs William’s syndrome

• Used an embedded face task (basically hide a face in a scene)

  • Scenes were presented for 5 seconds and they were instructed to look at each picture for as long as they were on the screen

• Also used a scrambled picture task (same thing but a picture was broken into tiles and components of the picture were scrambled, with one containing a face)

Results:

Embedded Face Task:

• There was no difference between William’s and TD in how long it took them to detect the face

  • However, Willliam’s kids fixated on the face for significantly longer and TD

• ADD kids took significantly longer to detect the face

  • When they did detect it, they spent less time fixating on it (didn’t care)

Scrambled Picture Task:

• TD and William’s took the same amount of time to locate the face, but William’s fixated significantly longer

• ADD did not fixate on faces for long at all

  • When a fixation was made, it took significantly longer to get there than other groups

Summary: Individuals with ADD showed dramatically less face gaze with shorter face fixations (don’t want to look at them, and when they do, make it short)

Individuals with William’s syndrome exhibit prolonged face fixations and face gaze.

What tasks can be used to assess facial attention?

• Embedded face task

• Scrambled picture task

Both ‘hide’ faces in scenes that don’t make much sense and track eye fixations

Evaluate this statement:

People with autism cannot process facial information

Evidence suggests that it may not be a reflection of ability, but rather that they are just not looking.

Ruby & Hancock (2009) found that in a face scramble and embedded face task, individuals with ADD took significantly longer to locate the face in the image, and when they did fixate, the fixations were significantly shorter than people with William’s syndrome and TD individuals.

There may still be deficits in facial processing, but this observation likely has something to do with those lack of fixations to actually detect the information in the first place.

What is the general pattern of fixations and eye movements for autistic people in regard to faces?

They show significantly less face gaze (less likely to fixate on faces) and when they do, the fixation duration is much shorter (Ruby & Hancock, 2009)

What is the general pattern of fixations and eye movements for William’s syndrome in regard to faces?

They have prolonged durations of face fixation and gaze

Describe Muszcat et al's (2015) study

They compared face scanning between people with ADHD, ADD or TD when looking at dogs or humans

Found that:

• ADD and ADHD kids spent less time viewing the eye region than other regions of the face

  • Supports that difficulties in processing social cues may be shared by both conditions

• People with ADHD had higher intellectual and adaptive functioning than people with ADD, but it didn’t impact the amount of time or region of interest

Takeaway: People with ADD and ADHD are less likely to look people in the eye

Limitation: Lab based - is this generalisable?

What did Pelc et al. observe?

In a face recognition task, children with ADHD had more difficulties with anger and sadness faces than other emotions

• Problems in emotion recognition were also identified in boys at risk for ADHD

Describe Fletcher-Watson et al.’s Study

Naturalistic study where they tracked eye movements of people with ASD to investigate attention to social information

• Presented ASD and TD adults/kids with a complex social scene (left), alongside a non-social scene (right)

• Px were to free view the scenes

Results:

• The people in the ADD group appeared to have a normal attentional preference for social information

• People in the TD group were more inclined to look at the object being looked at by the person in the scene than the ASD group.

Takeaway: Social attention deficits in ADD may not be because they don’t attend to social information per say, but they don’t attend in the same way TD people (i.e. staring at the person instead of the object the person is looking at) which may produce serious difficulties in real-life scenarios

This study was good because it was in a naturalistic setting, not lab

Body Dysmorphic Disorder

A preoccupation with an imagined defect in one’s appearance

• Distress causes functional impairment

• Characterised by some of the following behaviours:

  • Excessive mirror gazing (or avoidance)

  • Comparison of self to others

  • Excessive camouflage

  • Skin picking

  • Reassurance seeking

  • Avoidance of social situations

  • Intimacy avoidance

  • Increased use of alcohol or illegal substances

Describe how looking in a mirror can act as a trigger for BDD symptoms according to Veale’s model

Cyclic in nature and the process perpetuates its own existence

• They look in the mirror and selective attention to negative stimulus (ie. stomach) triggers a negative evaluation of self, but also promotes the processing of self as an aesthetic object

  • These all bidirectionally perpetuate each other

• The negative appraisal leads to

  • Safety behaviours like camouflaging or avoidance

  • Negative moods (disgust and depression)

  • Rumination on ugliness and comparisons to ideal

Describe Greenberg et al.’s Study

They studied how HC groups vs people with BDD viewed pictures of themselves and others, and compared them

• Individuals with BDD fixated for longer on their own features they perceived as unattractive

  • This could be due to an existing attentional bias that maintains the symptoms, but it could simultaneously contribute/reinforce the bias itself

• BDD also allocated more visual attention to the attractive features of others, which perpetuates upwards comparisons

• HCs had a more holistic view of themselves, indicating a more positive attentional process

Takeaway: there appears to be a cyclic process involving an attentional bias with people with BDD where they fixate on ‘unattractive’ parts of themselves, but ‘attractive’ parts of others, which perpetuates upwards comparisons

Describe the cognitive and attentional patterns associated with anorexia nervosa and bulimia nervosa

• Repeated checking of shape and weight as well as comparison of self to other’s body weight

• Cognitive bias is perpetuated by an attentional bias towards perceived flaws with self compared to aspirations to others

• Over-evaluation of body shape and weight

Takeaway: Suggests that individuals with these disorders may allocate different visual attention to different features which is perpetuated by the cognitive bias but simultaneously maintains it

Describe Blechert et al.’s (2009) Study

Investigated whether individuals with bulimia nervosa preferentially fixate on images of others with low BMI

Results: Yes, it appears so

• BN Px had longer fixations on individuals with low BMIs compared to HCs

• They spent significantly less time fixating on individuals with high BMI

• There was no difference between the amount of time spent fixating on their OWN image between the BN and HC groups

Takeaway: Disadvantageous social comparison strategies might be related to body dissatisfaction and maintenance of BN symptoms

Further support for the notion that upward comparisons play a huge role for EDs

Describe Horndasch et al’s (2021) Study

Compared visual attention of adolescent ED patients and HCs when looking at images of women wearing bathing suits or underwear

• 3 picture conditions; underweight, average and overweight

Results:

• All Px directed their visual attention to the abdomen, buttocks, hips and upper legs (no differences between ED and HC)

  • Think: No one is safe! This is why social media can be harmful

• HOWEVER, ED Px fixated on unclothed body parts for longer than HCs

Takeaway: People with EDs appear to have an attentional bias towards parts of bodies often deemed unattractive gives rise to potential attentional training interventions to reduce bias and prevent harm from idealised media images.

Describe Castellini et al.’s (2012) Study

Compared fMRI images been people with anorexia and HCs when viewing altered images of themselves with different body weights

• There was an elevated neural response in attentive, executive and self-evaluation networks in the brain for people with AN

• ESPECIALLY when they viewed themself in the overweight condition

• Everyone seemed to follow the same pattern of escalation with increase in body weight, but AN Px were stronger

Takeaway: A pathological neural response exists for people with AN which contributes to distorted body image through visual processing

Describe Giel et al.’s (2001) study.

Assessed attentional processing of food pictures in patients with anorexia and HCs who had fasted for 8 hours or hadn’t.

• Tracked eye movements for food and non-food control images

• AN Px showed more attentional disengagement to food pictures compared with control (think: cognitive avoidance strategy?)

Takeaway: AN people appear to disengage attention from food related stimuli

Describe attentional biases for individuals with alcohol related disorders

• Increased dwell times for alcohol related stimuli

• Reduced inhibitory control on saccadic movements

• Increased pupillary reactivity to visual stimuli regardless of emotional content

• Limited visual attention to prevention messages (!)

Takeaway: There appears to be this pattern of ‘I shouldn’t do that’ so it makes them want to look at it….?

Think: Trying to cognitively engage with the substance rather than ingesting it because it is more acceptable and discrete