Low–Cost Eye–Trackers: Useful for Information Systems Research?

Evaluating Operator Training Performance Using Recurrence Quantification Analysis of Autocorrelation Transformed Eye Gaze Data

OBJECTIVE

This research aimed to investigate the relationship between gaze behaviour dynamics and operator performance.

BACKGROUND

Individuals differ in their approach when learning a new task often resulting in performance disparity. During training some individuals learn the structure and dynamics of the task and develop a systematic approach, whereas others may achieve the same result albeit with increased perceived workload, or indeed some may fail to achieve superior performance levels. Previous research has shown that comparing gaze of experts with novices can provide unique insights into cognitive functioning of superior performers.

METHODS

Twenty-five individuals participated in a computer-based simulation task. The concept of coefficient of variation (CoV) of task scores was used to compute the participants' consistency of performance. Based on CoV, the cohort was split into two performance categories. The temporal patterns in participants gaze data were transformed using autocorrelation, and recurrence quantification analysis (RQA) was employed to analyse and quantify the patterns.

RESULTS

A Mann-Whitney U analysis demonstrated significantly (p < .01) higher determinism, entropy and laminarity in the superior group compared to the moderate group. Pearson's correlation revealed a significant (p < .01) negative correlation between the consistency of task performance (CoV) and the RQA measures.

CONCLUSION

The results demonstrated that eye gaze dynamics can be used as an objective measure of performance. Participants classified as superior performers consistently demonstrated a systematic gaze activity which were in line with the task structure.

APPLICATION

The methods presented here are applicable to observe and evaluate operators' strategic distribution of gaze. Specifically, for tactical monitoring and decision making in task environments where spatial locations of elements-of-interest vary continuously.

Prediction of depression symptoms in individual subjects with face and eye movement tracking

BACKGROUND

Depression is a challenge to diagnose reliably and the current gold standard for trials of DSM-5 has been in agreement between two or more medical specialists. Research studies aiming to objectively predict depression have typically used brain scanning. Less expensive methods from cognitive neuroscience may allow quicker and more reliable diagnoses, and contribute to reducing the costs of managing the condition. In the current study we aimed to develop a novel inexpensive system for detecting elevated symptoms of depression based on tracking face and eye movements during the performance of cognitive tasks.

METHODS

In total, 75 participants performed two novel cognitive tasks with verbal affective distraction elements while their face and eye movements were recorded using inexpensive cameras. Data from 48 participants (mean age 25.5 years, standard deviation of 6.1 years, 25 with elevated symptoms of depression) passed quality control and were included in a case-control classification analysis with machine learning.

RESULTS

Classification accuracy using cross-validation (within-study replication) reached 79% (sensitivity 76%, specificity 82%), when face and eye movement measures were combined. Symptomatic participants were characterised by less intense mouth and eyelid movements during different stages of the two tasks, and by differences in frequencies and durations of fixations on affectively salient distraction words.

CONCLUSIONS

Elevated symptoms of depression can be detected with face and eye movement tracking during the cognitive performance, with a close to clinically-relevant accuracy (~80%). Future studies should validate these results in larger samples and in clinical populations.

Evaluating the data quality of the Gazepoint GP3 low-cost eye tracker when used independently by study participants

The portability of low-cost eye trackers makes them attractive for research outside of the laboratory. Such research may require independent eye-tracker use. The present work compared the data quality of the Gazepoint GP3 when used independently by research participants with expert eye-tracking users. Twenty participants completed a training and a testing session 1 week apart. At training visits, participants were taught how to set up and use eye-tracking hardware and software and how to complete two tasks: a calibration task to measure accuracy and precision, as well as a visual search task to assess target fixations. At the testing session, participants set up the Gazepoint eye tracker and completed the two tasks without assistance. Participant accuracy and precision and visual search performance were compared to values obtained from two expert eye-tracking users. Additionally, the eye-tracker sampling rate, which is sensitive to factors such as head motion, was assessed in both participants and the expert users. Participant accuracy and precision closely approximated expert user values. Participant target fixations were detected with a 92.5% sensitivity and 76.8% specificity, closely mirroring expert user sensitivity and specificity. The sampling rate distribution was also similar between the participants and expert user (the means of those distributions were 16.99 ± 3.0 ms and 16.43 ± 2.3 ms, respectively). When used independently, data quality obtained from a low-cost, portable eye-tracking setup closely approximated values obtained from an expert user and was adequate enough to be a feasible option for some studies that require independent use by study participants.

Sample entropy analysis of pupillary signals in glaucoma patients and control via light-induced pupillometry

This paper evaluated the pupillary light reflex of glaucomatous eyes in the presence of constant lighting via light-induced pupillometry using sample entropy. The study used 20 patients and 15 controls, applied three different light intensities to their eyes, and recorded the behavior of the pupil. This study has validated that there is a difference in the entropy of pupillary data in glaucoma and healthy eyes. We concluded that entropy analysis is an excellent method to differentiate glaucoma eyes with the control through light-induced pupillometry. Hence, pupillometry has potential clinical applications in glaucoma investigation.

Enhancing the usability of low-cost eye trackers for rehabilitation applications

Eye tracking is one of the most widely used technique for assessment, screening and human-machine interaction related applications. There are certain issues which limit the usage of eye trackers in practical scenarios, viz., i) need to perform multiple calibrations and ii) presence of inherent noise in the recorded data. To address these issues, we have proposed a protocol for one-time calibration against the "regular" or the "multiple" calibration phases. It is seen that though it is always desirable to perform multiple calibration, the one-time calibration also produces comparable results and might be better for individuals who are not able to perform multiple calibrations. In that case, "One-time calibration" can also be done by a participant and the calibration results are used for the rest of the participants, provided the chin rest and the eye tracker positions are unaltered. The second major issue is the presence of the inherent noise in the raw gaze data, leading to systematic and variable errors. We have proposed a signal processing chain to remove these two types of errors. Two different psychological stimuli-based tasks, namely, recall-recognition test and number gazing task are used as a case study for the same. It is seen that the proposed approach gives satisfactory results even with one-time calibration. The study is also extended to test the effect of long duration task on the performance of the proposed algorithm and the results confirm that the proposed methods work well in such scenarios too.

Investigating the Use of Two Low Cost Eye Tracking Systems for Detecting Pupillary Response to Changes in Mental Workload

Eye tracking technologies are being utilized at increasing rates within industry and research due to the very recent availability of low cost systems. This paper presents results from a study assessing two eye tracking systems, Gazepoint GP3 and Eye Tribe, both of which are available for under $500 and provide streaming gaze and pupil size data. The emphasis of this research was in evaluating the ability of these eye trackers to identify changes in pupil size which occur as a function of variations in lighting conditions as well as those associated with workload. Ten volunteers participated in an experiment in which a digit span task was employed to manipulate workload as user’s fixated on a monitor which varied in background luminance (black, gray and white). Results revealed that both systems were able to significantly differentiate pupil size differences in high and low workload trials and changes due to the monitor’s luminance. These findings are exceedingly promising for human factors researchers, as they open up the opportunity to augment studies with non-obtrusive, streaming measures of mental workload with technologies available for as little as $100.