About the Project
Neuroadaptive interfaces utilise measures from the brain and body to adapt user interfaces in an intelligent, context-aware fashion. This emerging technology has the potential to enhance operator performance by optimising information provision and mitigating undesirable psychological states. However, neuroadaptive interfaces are reliant on robust, real-time monitoring of the operator via neurophysiology and psychophysiology.
It is assumed that laboratory-grade apparatus will deliver greater sensitivity to key variables, such as mental workload in comparison to wearable sensors. However, the latter are more comfortable and have fewer implementation requirements. The primary goal of the research programme is to quantify the trade-off between sensitivity and implementation requirements during neurophysiological and psychophysiological monitoring of operator functional state.
A series of experiments are planned to manipulate cognitive workload and sustained attention. The thesis will assess the sensitivity of measures in both cases while manipulating: (1) signal source (EEG, fNIRS, cardiovascular), (2) laboratory vs. consumer wearables, and (3) machine learning methodology. The goal of the research is to specify a minimal viable system that maximises sensitivity while minimising implementation requirements for real-time, classification of operator functional state.
Cognitive workload and sustained attention will be manipulated within an applied task environment and a number of experiments will be performed. During phase one of the programme, the sensitivity of workload/attention classification will be investigated by manipulating: (1) signal source (EEG, fNIRS, cardiovascular), (2) implementation requirements of EEG (many vs. sparse channels, wet vs. dry electrodes) and fNIRS (many vs. sparse channels), (3) machine learning method (high bias/low variance vs. low bias/high variance), and (4) subject-dependent vs. subject-independent classification.
The second stage of the research programme will test the validity of a classification algorithm to quantify the functional state of the operator in real-time and to create real-time visualisation of operator state.
This PhD studentship is funded by the Defence Science and Technology Laboratory (DSTL).
The project is based in the School of Psychology at Liverpool John Moores University and will be supervised by Professor Stephen Fairclough.
If you wish to apply for this studentship
- You must be a UK or EU national
- You must be available to enrol as a PhD student at Liverpool John Moores University on 1st June 2021
The successful candidate is expected to have:
(1) A good undergraduate degree in Psychology, Computer Science or a related discipline (Ergonomics/Human Factors)
(2) Experience of running psychological experiments and managing datasets
(3) Skills in statistical analyses
(4) An ability to work autonomously to deadlines and be self-organised
(5) Good written communication
(6) Good verbal communication skills, including presentation
It would be desirable if the candidate had:
- A Masters degree in a relevant discipline
- Knowledge and/or practical experience of using EEG
- Knowledge and/or practical experience of using fNIRS
- Knowledge and/or practical experience of using cardiovascular psychophysiology
- Experience of machine learning analyses, e.g., using R or MatLab
- Experience of creating experiments with multimodal data streams, e.g., using Lab Streaming Layer (LSL) or equivalent
If you wish to apply for this studentship, please prepare a 1-page statement (12-font) explaining why you are interested in this topic along with your cv and send to:
Professor Stephen Fairclough (firstname.lastname@example.org) with the subject marked as “PhD application”
The deadline for applications is Friday 7th May 2021.
Any offer to a prospective PhD student is contingent on candidate being cleared via administrative checks made by DSTL and the Ministry of Defence.
Why not add a message here
Based on your current searches we recommend the following search filters.
Based on your current search criteria we thought you might be interested in these.