MICHAL GABAY
Department of Neurobiology, The George S. Wise Faculty of Life Sciences and Sagol School of Neuroscience
Project description, including a specific plan for next year
We will use advanced VR goggles adapted with a binocular high-speed eye-tracking (ET) device (Tobii-installed VR HTC-Vive) that allows calculating pupil diameter, eye gaze and track head movements while participants are engaged in a virtual reality (VR) task. During the VR task, eye tracking data as well as behavioral data are recorded. These will be fed to train statistical and machine learning (ML) models to detect the individual experience measures and learning strategy. Our research proposal consists of two parts with two experimental paradigms.
In the comming year, we plan to finalize the data collection and analysis of the first experiment. In this unique dynamic VR task we developed, participants explore a house using body posture, head movements, and eye gaze. They search and look at objects - pets and control shapes that appear in pre-defined locations in a random order. Following that, the participants rank their preference, valence and arousal of the stimuli they saw. ET and behavioral features extracted, will be used to train ML models to predict individual rankings. Thereafter, we plan to continue the development and graphics design of second task, which will allow studying differences in spatial learning strategy. We plan to collect participants data and develop statistical and ML models to predict individual strategy from the physiological and behavioral signals.
About me
Michal Gabay, has a BSc and an MSc in biomedical-engineering from the Technion and Tel-Aviv University respectively, graduated with honors. During these studies she specialized in physiological mathematical models, and signal and image processing. Thereafter, Michal has joined a biomedical startup company as a machine learning (ML) and computer vision algorithms developer with the objective of early breast cancer detection. Through her time there, Michal was promoted to a position of algorithms team leader where she gained additional experience in project and team members' management. After 6 years, and motivated by a great interest in individual learning differences and their underlying neural mechanism, Michal has joined Shconberg's lab in Sagol School for Neuroscience as a data scientist. Following a year of specializing in multivariate pattern analysis of functional magnetic resonance imaging, Unity coding and developed tasks and environments in virtual reality (VR), last year she started her PhD there. To summarize, Michal has over 9 years’ experience in ML development, and physiological signal and image processing both in the industry and academy, which she aims to use in order to find the relation between individuals' learning differences and their physiological and neurological signals (as eye-tracking, and MRI), using a gamified VR environment, to induce a close to real life experience.