Saccades Methodology

Question 1

Why did you choose to do only two eye-tracking sessions pre and postexercise?

Answer 1

The decision to conduct only two eye-tracking sessions, one before and one after the exercise intervention, was based on the need to compare baseline oculomotor performance with post-exercise performance. This design is commonly used to identify any immediate effects of exercise on oculomotor function, minimizing participant fatigue and maintaining the integrity of the data by reducing potential confounding factors such as learning or adaptation effects that might occur with multiple testing sessions.

Question 2

Why did you not choose a longer oculomotor task? Why just 10 minutes?

Answer 2

The 10-minute duration for each block of the oculomotor task was chosen to balance the need for a sufficient number of trials to achieve reliable data with the need to prevent participant fatigue and loss of concentration, which can degrade data quality. Longer tasks can lead to increased variability in participant performance due to fatigue, and the 10-minute duration is a compromise to maintain high-quality, consistent data.

Question 3

Are 120 trials of pro and antisaccades sufficient to detect changes in oculomotor performance?

Answer 3

Yes, 120 trials of pro- and antisaccades are generally sufficient to detect changes in oculomotor performance. Previous studies have shown that this number of trials provides a reliable measure of saccadic reaction times, accuracy, and errors, allowing for robust statistical analyses. This number of trials strikes a balance between obtaining enough data for statistical power and avoiding participant fatigue.

Question 4

Describe the calibration technique used before the start of the oculomotor task. What was the reasoning behind the degree of pupil error used?

Answer 4

The calibration technique involved a nine-point calibration of the viewing space, followed by immediate verification to ensure that no point in the calibration space contained more than 1° of error. This high level of accuracy is critical in oculomotor studies to ensure that the measured gaze position accurately reflects the participant’s true gaze. The 1° error threshold is a standard used in eye-tracking studies to ensure data validity and reliability.

Question 5

Describe what the cut-off parameters were for reaction times, saccade duration, gain, and directional errors.

Answer 5

• Reaction times: Trials with reaction times (RTs) less than 50 ms (anticipatory responses) and RTs greater than 2.5 standard deviations from a participant- and task-specific mean were excluded.
• Saccade duration: Saccade onset was defined as when velocity exceeded 30°/s and acceleration exceeded 8000°/s²; saccade offset was when velocity fell below 30°/s for at least 40 ms. (The saccade was considered to have ended when the eye’s velocity dropped below 30°/s and remained there for at least 40 milliseconds)
• Gain: Analyzed as the ratio of saccade amplitude to target amplitude which was 13.5° for proximal targets and 16.5° for distal targets.
  Directional errors: Trials with directional errors, such as performing a prosaccade instead of an instructed antisaccade (or vice versa), were excluded from RT analysis because they involve different planning mechanisms.

Question 6

How did you pre-process the oculomotor data? What software did you use and why?

Answer 6

The oculomotor data were pre-processed using MATLAB (R2018b) with the Psychophysics Toolbox extensions and the EyeLink Toolbox. Data were filtered offline using a dual-pass Butterworth filter with a low-pass cut-off frequency of 15 Hz. MATLAB and its toolboxes are widely used in psychophysics and oculomotor research due to their flexibility, powerful data analysis capabilities, and compatibility with various eye-tracking hardware.

Question 7

How did you decide the distance between the monitor and chin-rest for the oculomotor task?

Answer 7

The distance between the monitor and the chin-rest (550 mm) was chosen based on standard practices in eye-tracking research to ensure that visual stimuli are presented within a comfortable and optimal viewing range. This distance helps minimize head movements and maintains a consistent visual angle for all participants, ensuring the accuracy of eye-tracking measurements.

Question 8

What performance metrics were you privy to in real time during the oculomotor task?

Answer 8

During the oculomotor task, the experimenter had access to real-time point-of-gaze information, trial-by-trial saccade kinematics, and data related to the accuracy of the eye-tracking system. This real-time feedback allowed the experimenter to monitor the performance and accuracy of the eye-tracking system continuously.

Question 9

How were you able to ensure that experimenter bias did not interfere with oculomotor performance, if you were privy to their performance metrics in real-time?

Answer 9

To minimize experimenter bias, the real-time feedback visible to the experimenter was used solely to monitor technical performance and not to influence participant behavior. Additionally, the experimenter maintained a standardized protocol and remained neutral during task execution to ensure that all participants received the same instructions and conditions, thereby reducing the risk of bias.

Question 10

Describe “proximal” versus “distal” targets in your oculomotor task. Why were they used, how often did they appear, and was their order random?

Answer 10

• Proximal targets were located 13.5° from the central fixation point, and distal targets were 16.5° away. These different target eccentricities were used to prevent participants from adopting stereotyped responses and to introduce variability in the task. The targets appeared randomly within a block of 60 trials, with the location (left or right) and type of saccade (pro- or antisaccade) being pseudorandomized to ensure a balanced and unpredictable presentation.

Question 11

Describe the necessity for a “gap paradigm” in your oculomotor task.

Answer 11

The “gap paradigm” involves extinguishing the fixation cross 200 ms before the target onset, creating a temporal gap. This paradigm was used to increase the difficulty of the task and to reduce the predictability of the target onset, thereby increasing the need for rapid and accurate saccadic responses. The gap paradigm is known to elicit shorter reaction times and more express saccades, making it a useful tool for studying oculomotor control.

Question 12

Why were the targets only presented briefly (i.e., for 50 ms) instead of for the duration of the eye-movement?

Answer 12

Extraretinal feedback refers to the internal monitoring of eye position and movement in the absence of visual input. When a target is presented briefly (50 ms), it disappears before the saccade is completed, preventing participants from using visual feedback to correct their saccade endpoints. By equating pro- and antisaccades for the absence of extraretinal feedback, the study ensures that any observed differences in performance between the two tasks can be attributed to the cognitive processes involved, such as response inhibition and vector inversion, rather than differences in the availability of visual feedback for endpoint correction.

Question 13

What were the instructions given to the participant regarding how to perform the oculomotor task?

Answer 13

Participants were instructed to perform the saccades “quickly and accurately” in response to the targets. For prosaccades, they were to look directly at the target location, while for antisaccades, they were to look at the mirror-symmetrical location opposite to the target. These instructions were given to ensure clarity and consistency in task performance across all participants.

Question 14

Why did you choose a separate “block design” instead of a “task-switching” design for the presentation of the saccade types?

Answer 14

A separate block design for pro- and antisaccade trials was chosen to reduce cognitive load and interference effects that might arise from frequent task switching. This design allows participants to focus on one type of saccadic task at a time, leading to more consistent performance and reducing the likelihood of errors that could arise from switching between tasks.

Question 15

How can you be sure that the reduction in reaction times observed is not due to practice effects?

Answer 15

The antisaccade reaction time (RT) reduction has been shown to be independent of a practice-related performance benefit for several reasons:
○ Frequentist and Bayesian Analyses:
Studies have demonstrated that antisaccade RTs remain equivalent when compared to those interspersed with a non-exercise control interval, as reported in multiple analyses. These findings indicate that the observed reductions in RTs post-exercise are not simply due to repeated exposure or practice effects, but rather are attributable to the effects of exercise itself (Dirk et al., 2020; Dyckman & McDowell, 2005; Klein & Berg, 2001; Samani & Heath, 2018; Shukla & Heath, 2022; Tari et al., 2020, 2023).
○ Antisaccade Durations and Gain Variability:
The fact that antisaccade durations and gain variability did not vary significantly across pre- to post-exercise assessments supports the notion that the post-exercise RT reduction was not due to a change in strategy aimed at reducing RT at the expense of accuracy (Fitts, 1954). If practice effects were responsible, one would expect to see changes in these metrics as participants might trade accuracy for speed (i.e., a speed-accuracy trade-off).

Question 16

Why did you choose to only dim the lights in the experiment room instead of turning them off completely during the oculomotor task?

Answer 16

The lights were dimmed to an ambient level of 39 cd/m² to reduce glare and reflections on the screen while maintaining enough light for the experimenter to monitor the equipment and participants. Completely dark environments can be uncomfortable for participants and can introduce variability in pupil size and eye-tracking accuracy. The chosen light level strikes a balance between these factors.

Question 17

Why did you wait for heart rate and blood pressure to return to baseline post-exercise before beginning the second round of oculomotor testing and was 5 minutes sufficient time to wait?

Answer 17

Waiting for heart rate (HR) and blood pressure (BP) to return to baseline ensures that any changes in oculomotor performance can be attributed to the exercise intervention itself rather than residual physiological effects of exercise. Five minutes was considered sufficient based on pilot data and previous literature suggesting that HR and BP typically return to baseline within this period after moderate exercise. Ensuring physiological measures are at baseline reduces confounding variables in the post-exercise assessment.

Question 18

Was there significant symptom exacerbation post-oculomotor task?

Answer 18

The SCAT-5 symptom frequency and severity did not reliably differ between pre- and post-exercise oculomotor assessments, indicating that there was no significant symptom exacerbation post-oculomotor task. This suggests that the oculomotor assessments were well-tolerated by the participants.

Question 19

How do you know that the post-exercise reduction in antisaccade reaction times is not just an enhancement of information processing and instead it’s a postexercise EF benefit in inhibitory control?

Answer 19

• The post-exercise reduction in antisaccade reaction times indicates a postexercise EF benefit in inhibitory control rather than just an enhancement of information processing due to several reasons:
  ○ Antisaccades specifically measure inhibitory control.
  ○ This task requires suppressing a reflexive prosaccade in favor of a volitional antisaccade, a process heavily reliant on EF (Munoz & Everling, 2004).
  ○ There is no significant pre- to post-exercise change in prosaccade reaction times.
  ○ Prosaccades are prepotent and less reliant on EF, suggesting that the reduction in antisaccade RTs is not due to general enhancement in information processing.
  ○ The improvement in antisaccade performance post-exercise aligns with previous research demonstrating that exercise specifically benefits EF-related tasks rather than general cognitive processing (Heath et al., 2018; Samani & Heath, 2018; Shukla & Heath, 2022).

Question 20

You suggested that the frequency of antisaccade directional errors might provide a resolution to detect persistent SRC-related EF deficits. Could you elaborate on the potential clinical implications of this finding and how it might inform the assessment and management of SRC?

Answer 20

• The persistence of increased antisaccade directional errors in the SRC group, even after the exercise intervention, suggests that this measure may be a sensitive indicator of lasting EF deficits following concussion.
  Clinically, this finding could inform the development of oculomotor-based assessments for concussion diagnosis and recovery monitoring. By tracking changes in antisaccade directional error rates over time, clinicians may be able to detect subtle EF impairments that persist beyond the resolution of other concussion symptoms.

Question 21

The results showed that the exercise intervention led to a reduction in antisaccade reaction times for both the SRC and HC groups, but did not affect the frequency of directional errors in the SRC group. How do you interpret these findings in terms of the specific cognitive processes involved in antisaccade performance, such as response inhibition and vector inversion?

Answer 21

• The reduction in antisaccade reaction times (RTs) following the exercise intervention in both the SRC and HC groups suggests that acute exercise improves the efficiency of response inhibition and vector inversion processes, which are critical for successful antisaccade performance. Response inhibition involves suppressing the automatic tendency to make a prosaccade towards the target, while vector inversion requires inverting the spatial coordinates of the target to generate a saccade in the opposite direction. The improved RTs indicate that exercise enhances the speed at which these cognitive processes can be executed.
• However, the lack of change in directional error rates for the SRC group suggests that while exercise may enhance the speed of response inhibition and vector inversion processes, it does not necessarily improve their accuracy in individuals with concussion. This finding highlights the dissociation between the efficiency (RT) and effectiveness (accuracy) of cognitive control processes in antisaccade performance.
• The increased frequency of antisaccade directional errors in the SRC group may be related to a failure to evoke the high-level executive function (EF) task-set supporting a non-standard response (Everling & Johnston, 2013). A single bout of exercise might not provide a sufficient stimulus to improve the computational demands associated with adopting the task-set, or task rules, necessary to evoke a directionally correct antisaccade. These findings suggest that the frequency of antisaccade directional errors provides the resolution to detect persistent SRC-related EF deficits.

Antisaccades are not purely inhibitory control tasks (Munoz & Everling, 2004); they also require working memory and cognitive flexibility to maintain the goal of performing an antisaccade instead of a prosaccade. Deliberate practice is necessary to reduce antisaccade directional errors, as they represent a nonstandard task-set (Dyckman & McDowell, 2005). Thus, the increased directional errors in the SRC group, despite return-to-sport clearance, suggest an oculomotor EF deficit (Everling & Johnston, 2013).

Question 22

You suggested that the frequency of antisaccade directional errors might provide a resolution to detect persistent SRC-related EF deficits. Could you elaborate on the potential clinical implications of this finding and how it might inform the assessment and management of SRC?

Answer 22

The persistence of increased antisaccade directional errors in the SRC group, even after the exercise intervention, suggests that this measure may be a sensitive indicator of lasting EF deficits following concussion. Clinically, this finding could inform the development of oculomotor-based assessments for concussion diagnosis and recovery monitoring. By tracking changes in antisaccade directional error rates over time, clinicians may be able to detect subtle EF impairments that persist beyond the resolution of other concussion symptoms.

Question 23

Given the promising findings of your study, what future research directions would you propose to further investigate the effects of acute exercise on cognitive function in individuals with SRC, and how might these findings inform the development of exercise-based interventions for SRC recovery?

Answer 23

• Future research could investigate the dose-response relationship between exercise intensity, duration, and cognitive function in individuals with SRC, to determine the optimal exercise parameters for enhancing EF recovery.
• Additionally, longitudinal studies could examine the long-term effects of regular exercise on cognitive function and symptom resolution in SRC patients.
  These findings could inform the development of personalized exercise-based interventions for concussion rehabilitation, targeting specific EF domains and tailoring exercise prescriptions to individual recovery trajectories.

Question 24

Design a follow-up study, not considering funding or resources, to address your limitations and improve upon your methodology

Answer 24

Methods:
* Recruit a larger sample (i.e.,120 SRC participants (60 males, 60 females) and 120 healthy controls
* Match controls on age, sex, BMI, and cardiovascular fitness (i.e., VO2) if possible
* Determine SRC participants’ heart rate threshold (HRt) using the Buffalo Concussion Bike Test.
* Longitudinal assessment of EF pre- and post-exercise at baseline, 2, 4, and 8 weeks post-concussion.
* Monitor CBF changes during exercise using fNIRS and TCD.
* Track SRC symptoms using the complete SCAT5 tool at each time point and 24 hours post-exercise.
* Develop personalized aerobic exercise programs based on HRt.
* Collect data on menstrual cycle.
* Compare outcomes between male and female SRC participants and controls, considering covariates such as years of sport played and lifetime concussion history.

Expected Results:
* Improved EF, symptom resolution, and faster return-to-play in SRC participants following personalized exercise programs.
* Increased CBF during exercise, correlating with EF improvements in both sexes.
Potential sex differences in exercise-induced CBF changes.

Question 25

Describe the EyeLink 1000 Plus; it’s validity and reliability, strenghts and limitations.

Answer 25

• The EyeLink 1000 Plus by SR Research Ltd. (Ottawa, Canada) is a high-performance, video-based eye-tracking system used in research to accurately measure eye movements, gaze position, and pupil size. It offers a sampling rate of up to 2000 Hz and can be used in various configurations. Its strengths include exceptional accuracy, fast sampling rate, versatile configurations, real-time data access, and robust software.

However, it has limitations such as high cost, the need for head stabilization, required technical expertise, potential challenges with certain participant characteristics, such as glasses, contact lenses, or eye makeup, which can affect the quality of eye-tracking data, and complex data processing.

Question 26

You mentioned that The gaze position of the participants’ left eye was sampled at 1000 Hz. Describe why not the right eye, and why not sample at 2k or less than 1k?

Answer 26

• Choosing to sample the left eye over the right eye is often an arbitrary decision, as there is typically no significant difference in gaze position between the two eyes in healthy individuals. Both eyes generally look at the same point in space during normal viewing conditions. This assumption is based on the fact that our eyes are yoked together and move in a coordinated manner to maintain binocular vision.
• A sampling rate of 1000 Hz means that the eye-tracker captures gaze position data every millisecond, providing a high temporal resolution. This enabled us to capture rapid eye movements, such as saccades, which can occur within a few milliseconds. Higher sampling rates generally provide more detailed and accurate data, as they capture more data points within a given time frame. This increased data density can be beneficial for analyzing fine-grained eye movement patterns and detecting subtle changes in gaze position. We chose a sampling rate of 1000 Hz because it provides an optimal balance between temporal resolution and data manageability for our specific research needs.

Question 27

Initially, pre-exercise, the SRC group showed an oculomotor EF impairement as observed by significantly greater antisaccade RTs compared to the HC group. However, post-exercise, you did not observe a significant difference between the two groups with regards to the antisaccade RT reductions. This suggests that the SRC group had a significantly larger magnitude of decrease in antisaccade RTs pre-to-postexercise (51.33, SD = 34) versus (19.84, SD = 18) respectively, in order to make the baseline differences in antisaccade RTs insignificant. What is your interpretation of this outcome, and how do you explain this finding?

Answer 27

• Initially, the SRC group showed impaired oculomotor function, indicated by longer antisaccade reaction times (RTs) compared to the HC group. Post-exercise, the difference between the groups in antisaccade RT reductions was insignificant, suggesting a larger pre-to-postexercise improvement in the SRC group (51.33, SD = 34) versus the HC group (19.84, SD = 18).
• Larger Margin for Improvement: The SRC group had more room for improvement due to their initial impairment, while the HC group, with intact oculomotor function, had less potential for further enhancement, possibly due to a ceiling effect.
• Altered Neurophysiology: Individuals with a history of concussions may have changes in cerebral blood flow, neurotransmitter balance, and neural network efficiency, making them more responsive to the effects of acute exercise on brain function. Exercise-induced compensatory mechanisms—such as increased cerebral blood flow, neurotransmitter release, and improved synaptic efficiency—could have had a more significant impact on the SRC group.
• Impaired Neural Networks: The neural networks responsible for oculomotor control and executive function may be more impaired in the SRC group due to the lasting effects of concussions. Acute exercise might temporarily normalize these impaired networks through compensatory mechanisms. In contrast, the HC group’s neural networks, already functioning optimally, were less affected by the exercise intervention.
• Targeted Effect of Exercise: The exercise-induced changes in cerebral blood flow, neurotransmitter levels, and neural activity are particularly beneficial in counteracting the specific impairments caused by concussions. This targeted effect could lead to a more substantial improvement in the SRC group compared to the HC group.
• The greater improvement in oculomotor function observed in the SRC group post-exercise can be attributed to their larger room for improvement, altered neurophysiology, differential effects on impaired neural networks, and the specific interaction between exercise and concussion-related deficits. These factors allowed the compensatory mechanisms triggered by acute exercise to have a more pronounced impact on the SRC group’s oculomotor function compared to the HC group.

Question 28

What do you mean by this sentence: For the oculomotor task, gaze position data were filtered offline using a dual-pass Butterworth filter with a low-pass cut-off frequency of 15 Hz. A five-point central-finite difference algorithm was used to compute instantaneous velocities and accelerations. Saccade onset was determined when velocity and acceleration exceeded 30°/s and 8000°/s2, respectively. Saccade offset was determined when velocity fell below 30°/s for 40 ms.

Answer 28

• A mathematical algorithm called a five-point central-finite difference was used to calculate how fast the eyes were moving (velocity) and how quickly they were changing speed (acceleration) during the task.
• This helps to understand the dynamics of eye movements and how smoothly or rapidly they occur.
• The gaze position data collected during the oculomotor task was cleaned up using a special filter called a Butterworth filter.
• This filter helps remove any unwanted noise or small variations in the data, making it smoother and more accurate for analysis.
  Hz (Hertz) refers to the unit of frequency, which measures the number of cycles or oscillations per second.
• We chose 15 HZ because saccades occur at relatively low frequencies, typically below 15 Hz.
• Filtering out the higher frequencies above 15 Hz helps to isolate the true eye movement signals and remove any unwanted fluctuations or artifacts.
• We used specific speed and acceleration thresholds to identify when saccades started and ended:
• Saccade start: When the eye speed exceeded 30 degrees per second and the acceleration exceeded 8,000 degrees per second squared.
• Saccade end: When the eye speed fell below 30 degrees per second for at least 40 milliseconds.
• This allowed us to precisely pinpoint the timing of each saccade.
• These thresholds are commonly used in the oculomotor research literature to reliably detect the beginning and end of saccadic eye movements.
• Saccades are characterized by very rapid changes in eye position, so these velocity and acceleration values help to distinguish them from slower, more gradual eye movements.

Question 29

Why did you choose reaction time (RT), directional errors, saccade duration, and saccade gain variability as the occulomotor variables to compare SRC and healthy controls?

Answer 29

• Reaction time is a fundamental measure of processing efficiency, with antisaccade RTs in particular reflecting the time-consuming demands of inhibitory control and spatial remapping required for this task (Munoz & Everling, 2004). Numerous studies have demonstrated that individuals with SRC exhibit prolonged antisaccade RTs compared to healthy controls, indicating impairments in this key aspect of executive function (Johnson et al., 2015a, 2015b; Webb et al., 2018; Ayala & Heath, 2020).
• Directional errors on antisaccade trials provide a measure of the effectiveness of inhibitory control. Generating a saccade in the opposite direction of a visual stimulus requires the active suppression of the prepotent prosaccade response and the implementation of the appropriate task set. Increased antisaccade directional errors in SRC populations reflect a failure to consistently engage this high-level executive control (Everling & Johnston, 2013; Ayala & Heath, 2020; Webb et al., 2018).
• Saccade duration and gain variability offer insights into the visuomotor processing underlying oculomotor control. Antisaccades, which decouple the typical stimulus-response mapping, have been shown to exhibit longer durations and more variable endpoints compared to prosaccades (Edelman et al., 2006). These metrics can provide additional information about the integrity of sensorimotor transformation processes that may be disrupted following SRC.

By examining this comprehensive set of oculomotor variables, the current study was able to provide a multifaceted assessment of both the efficiency and effectiveness of executive function in individuals with SRC.

Question 30

What does the similar saccade gain variability between the SRC and HC groups suggest about the neural processes underlying the cognitive deficits following concussion?

Answer 30

• Saccade gain variability is thought to reflect the integrity of the sensorimotor transformation processes that convert visual target information into the appropriate oculomotor commands. This process involves the integration of visual, spatial, and motor signals within a distributed neural network, including the superior colliculus, frontal eye fields, and posterior parietal cortex.
• The fact that saccade gain variability was equivalent between the SRC and HC groups suggests that the fundamental visuomotor control mechanisms underlying saccadic eye movements remain largely intact following concussion. This implies that the cognitive deficits observed in the SRC group, as evidenced by their longer antisaccade reaction times and increased directional errors, are not likely due to disruptions in the sensorimotor transformation processes per se.
• Rather, the concussion-related impairments appear to be more specifically targeted at the higher-level executive control processes required to suppress the prepotent prosaccade response and implement the appropriate antisaccade task set. These executive functions are mediated by a distinct neural network, including the dorsolateral prefrontal cortex, anterior cingulate cortex, and basal ganglia, which seem to be more vulnerable to the neurometabolic cascade and functional disruptions induced by concussive injury.
• This dissociation between preserved sensorimotor processing and impaired executive control aligns with the notion that concussion primarily impacts the integrity of frontal-subcortical circuits and associated cognitive control mechanisms, while leaving more posterior sensory and motor systems relatively spared (Giza & Hovda, 2014; Rabinowitz & Levin, 2014).

By examining a comprehensive set of oculomotor metrics, the current study was able to isolate the specific neural processes underlying the executive function deficits observed in the SRC group. This provides important mechanistic insights that can help guide the development of targeted rehabilitation strategies to address the cognitive sequelae of concussive injuries.

Question 31

What would have been different had you assessed executive function via the pro-and-antisaccade task during the exercise instead of before and after? How would this have affected your results in terms of the differences between the SRC and healthy control groups?

Answer 31

Administering the pro- and antisaccade task concurrently with the exercise intervention, rather than before and after, could have provided additional insights into the dynamic interplay between exercise, cerebral blood flow, and executive function in the context of concussive injury.

Performing the oculomotor assessment during the exercise bout could have revealed potential differences in the temporal patterns of executive function changes between the groups. This may have uncovered impairments in the SRC group’s ability to appropriately modulate neural activity and information processing in response to the increased metabolic demands.

However, we opted not to take this approach, as we were concerned about unduly stressing our concussed participants by combining the cognitive and physical challenges. While this limited our ability to examine the real-time dynamics, the pre- and post-exercise assessment still provided valuable insights into the executive function deficits observed in the SRC group and the potential EF benefits of an acute bout of exercise