Bachelor and Master Theses

Purpose

The project is expected to, through the feedback system that is developed, generate new knowledge in the field of neuroscience and learning with the purpose to develop efficient techniques for training and rehabilitation of people with cognitive impairments.

Background

The human brain typically integrates a rich variety of feedback in order to adapt and control behavior. Providing a person with feedback based on his/her performance on a specific task has been shown to create a powerful mean to adapt and learn that specific task (Carmena et al., 2003). However, the delay of this feedback has been shown critical for learning (Belinskaia et al., 2020). In RECOG, we are particularly interested in developing a real-time feedback system to improve time-critical aspects in order to enable reliable and accurate feedback to be seamlessly integrated into a cognitive memory task.

Important advances in the domain of artificial intelligence (AI) and the use of these techniques in neuroscience have unleashed many new promising opportunities to identify patterns in time series data related to specific functions of interest. In particular, pupil dilation is an interesting feature to enable the targeting of functions related to cognition. The pupil size changes continuously in response to variations in ambient light levels (Toates, 1972), but this behavior is not solely due changes in light intensity but early observations show that the pupil size varies systematically in relation to cognitive demands, attention and effort (Kahneman 1973, Hess and Polt 1964).

The project aims at exploiting pupil size changes in real time to decode a person’s cognitive workload simultaneously as the person is performing a memory task. The level of cognitive workload at a certain time instance will then be used to adapt the difficulty of the memory task, thus enabling effective cognitive training  through the use of real-time biofeedback.

Merging systems engineering, fundamental neuroscientific principles of learning and advances in AI puts this project at the frontier of novel real-time feedback technology applicable to many different industrial areas. The project will be carried out in close connection to an ongoing research project at MDU involving three partner companies; Smart Eye, Prevas and National Instruments.

References:

Belinskaia, A., Smetanin, N., Lebedev, M., and Ossadtchi, A. (2020). Short-delay neurofeedback facilitates training of the parietal alpha rhythm. J. Neural Eng. 17, 066012.

Carmena, J.M., Lebedev, M.A., Crist, R.E., O’Doherty, J.E., Santucci, D.M., Dimitrov, D.F., Patil, P.G., Henriquez, C.S., and Nicolelis, M.A.L. (2003). Learning to Control a Brain-Machine Interface for Reaching and Grasping by Primates. PLoS Biol 1.

Hess, E.H., and Polt, J.M. (1964). Pupil Size in Relation to Mental Activity during Simple Problem-Solving. Science 143, 1190–1192.

Kahneman, D. (1973). Attention and effort. (Englewood Cliffs, N.J.: Prentice-Hall).

Toates, F.M. (1972). Accommodation function of the human eye. Physiological Reviews 52, 828–863.

The student should have knowledge in the following areas:

- LabVIEW

- MATLAB

- FPGA programming

- AI

Additional supervisors will be appointed before the start of the thesis.

If interested or have questions, contact Elaine at elaine.astrand@mdu.se