Table 5.

Durbin-Watson results for whiteness of residuals. Values in bold exceeded DW critical values, but were within tolerances suggested by the original authors of the MVGC method]. All values exceeding critical DW values (bold) were however within the range 1.91 < DW < 2.1, with the exception of a single participant’s (*) pupil model residuals (DW = 1.78). In line with the tolerances used in the original study, 80% of residuals showed no significant autocorrelation, indicating the autoregressive models were overall sufficient to account for the autocorrelation present in the time series.

	Resp	Fz	Pupil
P1	2.0001 (0.99)	2.0040 (0.60)	2.0040 (0.18)
P2	1.9979 (0.77)	2.0017 (0.82)	1.9797 (<0.01)
P3	2.0037 (0.65)	2.0064 (0.43)	1.9965 (0.66)
P4	1.9830 (0.05)	1.9244 (<0.01)	2.0004 (0.97)
P5	1.9778 (<0.01)	1.9999 (0.98)	2.0015 (0.82)
P6	1.9881 (0.10)	1.9997 (0.96)	1.9306 (<0.01)
P7	2.0197 (0.0213)	2.0028 (0.75)	2.0017 (0.85)
P8	1.9961 (0.57)	2.0007 (0.92)	1.9120 (< 0.01)
P9	2.0017 (0.80)	2.0005 (0.95)	1.7753 * (<0.01)
P10	2.0128 (0.09)	1.9996 (0.96)	1.9997 (0.96)