Table 4.

Spearman correlation values of the performance metrics against segment time between user and expert data

Metrics	Rho-value (user)	Rho-value (expert)
1. Cluster number of eye gaze and virtual orientation	0.37	0.35
2. Cluster centres of eye gaze and virtual orientation	0.41	0.28
3. Cluster number of eye gaze	0.27	0.19
4. Cluster centres of eye gaze	0.31	0.14
5. No. of blinks/s	0.27	0.17
6. Screen time	0.28	0.17
7. No. of collisions/s	0.56	0.49
8. Path with speed performance	0.82	0.91
9. Path performance	0.57	0.48
10. Cluster number of head pose and virtual orientation	0.44	0.49
11. Cluster centres of head pose and virtual orientation	0.50	0.38
12. Cluster number of head pose	0.23	0.14
13. Cluster centres of head pose	0.40	0.11
14. Cluster centres of user inputs	0.59	0.44
15. Cluster number of user inputs	0.49	0.54
16. Average braking	0.21	0.25
17. Average throttle	0.50	0.60
18. Average steering	0.48	0.33
19. Eye Fixations	0.20	0.17

All p values are $<0.001$. The closer the value is to 1 the closer the particular variable is more correlated to the segment time. As anticipated the physiological signals have on average lower significance to the time than the non-physiological. (Values $\ge 0.40$ are in bold)