Table 1.

Mean values shown; standard deviation values in [brackets].

	Photopic light (lux)	Circadian light (CLA)	Activity Index (Δg-forcerms)	Interdaily Stability (IS)	Intradaily Variability (IV)	Phasors
						Sliding-window		All-at-once
						Magnitude	Angle (hours)	Magnitude	Angle (hours)
Day-shift (N=38)	302 [188]	369 [227]	0.0094 [0.0014]	0.692 [0.144]	0.447 [0.170]	0.50 [0.11]	0.65 [0.74]	0.46 [0.12]	0.68 [0.71]
Rotating-shift (N=61)	188 [152]	209 [166]	0.0097 [0.0018]	0.252 [0.129]	0.458 [0.135]	0.33 [0.10]	0.64 [0.73]	0.12 [0.10]	2.3 [3.2]
t-test p-value (two-tail)	<0.01	<0.001	0.26	<0.001	0.72	<0.001	0.94	<0.001	<0.001

Photopic light exposures: The Daysimeter system utilizes a fully calibrated (spectral, spatial, intensity) photopic light sensor measuring in lux (lm/m²). Group means [standard deviation] are computed from individual subject means of each subject's entire recording session.

Circadian light exposures: Using the model from Rea et al.,¹⁶ circadian light exposure values are determined from the photopic and blue sensor data measured in CL_A. Group means [standard deviation] are computed from individual subject means of each subject's entire recording session.

Activity index: Values from two orthogonally oriented accelerometers (measuring up/down and forward/back head motions) are used to compute an activity index that is logged at regular time intervals along with the light readings. Each logged measure of activity is the rms combination of the standard deviation of acceleration taken once per second over a 30-second interval of both accelerometers. Group means [standard deviation] are computed from individual subject means of each subject's entire recording session.

Interdaily stability: The IS statistic developed by Van Someren et al.²⁴ measures the consistency of activity among days and ranges from 0 to 1. A value of 1 results when every day's activity is identically to the other days, while conversely a value of 0 results from no similarity among days.

Intradaily variability: The IV statistic developed by Van Someren et al.²⁴ measures the fragmentation of rest and activity based on a scale from 0 (no variability from hour to hour) to upwards of 2. A larger value indicates more fragmentation of rest and activity, or conversely, less consolidation of rest/activity patterns.

Phasor magnitude: A correlation between light exposure (CS) and activity (calculated using either the sliding-window or the all-at-once technique) for the observation period, in this case 5–7 days. A higher magnitude indicates the subject has a consistent, 24-hour schedule with respect to activity and light. Lower magnitudes indicate low correlation between daily cycles of light and activity irrespective of phase differences.

Phasor angle: A phase relationship between light exposure (CS) and activity (calculated using either the sliding-window or the all-at-once technique) for the observation period, in this case 5–7 days. A positive angle (first quadrant) indicates a delay in activity with respect to light and a negative angle (fourth quadrant) indicates an advance in activity with respect to light for diurnal species.