Using near infrared spectroscopy (NIRS) to assess workload and inhibitory control in real and simulated driving environments

Foy, Hannah J. (2017) Using near infrared spectroscopy (NIRS) to assess workload and inhibitory control in real and simulated driving environments. PhD thesis, University of Nottingham.

[img] PDF (Thesis - as examined) - Repository staff only - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (4MB)

Abstract

Both mental workload and inhibitory control have previously been found to relate to road traffic collisions and these factors are also believed to be highly dependent on the prefrontal cortex. Driving simulators create a safe environment in which to manipulate these factors and examine human behaviours. However, the validity of simulators is often called in to question and previous simulator validation research has a number of common methodological issues. As such, there were two main aims of this thesis. The first was to examine the efficacy of functional near infrared spectroscopy in measuring changes in prefrontal cortex activation as a result of mental workload and inhibitory control manipulations in simulated driving scenarios. The second aim was to examine the similarities and differences in driving behaviour, eye movements and prefrontal cortex activation in real and simulated driving environments.

In Experiment One mental workload was examined using a simulated driving task which used different road types to manipulate workload in a medium fidelity driving simulator. Results showed changes in subjective workload ratings with changes in road types, with dual carriageway roads having the lowest mental workload followed by A-roads, city centre routes and suburb roads, which had the highest workload demands. Increases in mental workload were accompanied by reductions in speed and mean fixation duration and increases in accelerations, horizontal spread of search, skin conductance responses and the concentration of oxygenated haemoglobin in the prefrontal cortex.

In Experiments Two and Three, inhibitory control was examined in both a high and medium fidelity simulator. Inhibition was manipulated by using different hazard types and by changing the role of the participant to active or passive. Results showed changes in prefrontal cortex activation with changes in hazard type in the high fidelity simulator. Although the same pattern of results was found in the medium fidelity simulator the magnitude of change was lower and results were not statistically significant. Subjective ratings of risk and mental workload also showed an increased magnitude of change from everyday to hazardous driving in the high fidelity simulator suggesting that it may be a more useful tool for comparing hazardous scenarios than the medium fidelity simulator. Results also revealed greater activation for active than passive participants for oxygenated haemoglobin.

In Experiment Four, driving behaviour and eye movements were compared between real world and simulated driving at both high and medium fidelities using the same road definitions as in Experiment One. Results showed that participants had reduced speeds and spread of search and increased accelerations and mean fixation durations in the car. However, changes in driving behaviours and eye movements in all environments were consistent with the patterns seen in Experiment One. More specifically, there were increases in acceleration and spread of search and reductions in speed and fixation duration with increased mental workload. These results demonstrate relative validity for both simulators. Behavioural equivalence was found between the high and medium fidelity simulator for all eye movement and speed measures and two (right and absolute) of the five acceleration signatures.

Finally, Experiment Five compared the similarities and differences in driver behaviour and prefrontal cortex activation as measured using functional near infrared spectroscopy in on-road and high fidelity simulated driving. Results showed reduced speed and increased acceleration in the car compared to the simulator but as with Experiments One and Four there were increases in acceleration and reductions in speed with increased workload in both environments, showing support for relative validity for driver behaviour measures. Prefrontal cortex activation did not show meaningful patterns of activation in either environments or across changes in road types.

It was concluded that functional near infrared spectroscopy is a valuable neuroimaging technique that can be used to detect mental workload changes and prefrontal cortex responses to hazardous situations in controlled, simulated environments. However, it is currently not an effective technique for determining these changes in mental workload in real world driving where experimental control is reduced. With respect to simulator validation, driver behaviour and eye movements differ numerically between on-road and simulated driving but show the same direction of change with changes in road demands. As such, there is relative validity for both medium and high fidelity simulated driving and behavioural equivalence between the two levels of simulator fidelity.

Item Type: Thesis (University of Nottingham only) (PhD)
Supervisors: Chapman, Peter
Newport, Roger
Keywords: Driving Simulation; Prefrontal Cortex; Mental Workload; Inhibitory Control;
Subjects: B Philosophy. Psychology. Religion > BF Psychology
T Technology > TL Motor vehicles. Aeronautics. Astronautics
Faculties/Schools: UK Campuses > Faculty of Science > School of Psychology
Item ID: 47080
Depositing User: FOY, HANNAH
Date Deposited: 05 Jan 2018 11:23
Last Modified: 05 Jan 2018 17:31
URI: https://eprints.nottingham.ac.uk/id/eprint/47080

Actions (Archive Staff Only)

Edit View Edit View