Eye movement based metric serving as an on-site test metric for time awake
Eye movements can be used to study tiredness, lack of sleep as well as disturbances of attentiveness.

Prolonged time awake increases the need to sleep and causes sleepiness. It has been estimated that sleepiness causes 10–20 % of traffic accidents hence being a major identifiable and preventable cause of accidents. Even though the severity of sleepiness-related accidents and hazards have been recognized, there is no reliable on-site tester for estimating total time awake of a person. An objective and practical metrics for measuring sleepiness outside the laboratory is needed.

In her doctoral thesis Licentiate of Philosophy Kati Pettersson is presenting a novel approach to this and examines whether an eye movement based metric could serve as an on-site test metric for time awake.

Algorithm development & development of an eye movement based metric

The thesis consists of two parts. Algorithm development for electro-oculographic (EOG) feature extraction to enable effective and practical analyses of measurements conducted outside the laboratory, and development of an eye movement based metric to estimate prolonged time awake.

"The rationale for the studying the use of eye movements to estimate overall time awake is as follows: Different cognitive functions, especially attentional ones are vulnerable to sleepiness. The attentional and oculomotor processes share neuroanatomical networks in the brain and saccadic eye movements have been used to study attentional functions. Moreover, saccadic eye movements are sensitive to sleepiness," she says.

Cognitive performance comparable to alcohol intoxication

Increasing sleep drive induces rapid and uncontrolled sleep initiation leading to unstable cognitive performance which is comparable to alcohol intoxication.

Eleven healthy adults were tested during 60 hours of time awake

Saccadic eye movements were measured from eleven healthy adults every sixth hour with EOG in a 8-minute saccade task during 60 hours of prolonged time awake. The study finds that the saccade task performance, estimated as the number of saccades, decreased as a function of time awake on an individual level.

The saccadic performance differed between the participants but was stable within participants; this was tested with 5 participants.

The circadian rhythm affected the saccade task performance. Thus, the three process model of alertness (TPMA) was fitted to, and the circadian component was removed from the measured data. After removing the C-component, the linear model revealed a significant trend for six out of eleven participants.

Individual calibration is still needed

The results imply that saccades measured with EOG could be used as a time awake metric outside the laboratory. The metric needs individual calibration before the time awake of a person can be estimated.

"More research is needed to study individual differences, optimize the measurement duration, and stimulus parameters," Kati Pettersson says.

The doctoral thesis and the defence:

Kati Pettersson will defend the thesis entitled "Saccadic eye movements estimate prolonged time awake" in the Faculty of Science, University of Helsinki on Wednesday, September the 12th at noon. Professor Hans Van Dongen will serve as the opponent. The defence will take place at the Main building, Unioninkatu 34, Auditorium XII.

Lic. Phil. Pettersson has conducted the thesis work at the Finnish Institute of Occupational Health, under the supervision of Docent Kiti Müller (Nokia Bell Labs, Espoo & Department of Neurology, University of Helsinki) and Professor Edward Hæggström (Department of Physics, University of Helsinki).

The electronic thesis can be found here: https://helda.helsinki.fi/handle/10138/239934

Researcher Kati Pettersson, +358 40 533 7881, kati.pettersson@ttl.fi