Figure 1.
Study population mean-waveforms.
Mean-waveforms for light exposure (Light) and environmental temperature (Env Temp) (A), wrist temperature (WT) and sleep (B), and activity and position (C). The shaded area shows the mean sleep period. All variables are expressed as mean ± SEM (n=103). The mean values (±95% Confidence Interval) for Mesor, Amplitude and Acrophase, as well as the %V as calculated by the cosinor analysis for the above-mentioned variables, are shown in D. * indicates p<0.001 according to the cosinor analysis.
Figure 2.
WT pattern purified for activity.
Demasked WT pattern, expressed as mean ± SEM after application of the purification by intercepts or categories method (correcting for the effect of activity). The shaded area shows the mean sleep period. The table below the graph shows the corresponding Mesor, Amplitude and Acrophase as well as the %V for WT, demasked by means of the purification by categories or intercepts method, (data are expressed as Mean±95% Confidence Interval). * indicates p<0.001 according to the cosinor analysis.
Figure 3.
WT pattern purified for each studied masking variable.
Demasked WT pattern, expressed as mean ± SEM, following the application of the purification by categories method according to environmental temperature level (A), light exposure (B), activity (C), position (D) and sleep status (E). The shaded area shows the mean sleep period. Note that in the case of the lowest level of activity, the same data set as for Figure 1 has been used, although in this case, with intervals of ten minutes instead of one hour. See the material and methods section for more details. The values for Mesor, Amplitude and Acrophase as well as the %V of each demasked wrist temperature pattern are expressed as Mean±95% Confidence Interval, and are included in F. * indicates p<0.001 according to the cosinor analysis.
Figure 4.
WT pattern after constant routine approach.
Demasked WT waveforms obtained using the constant routine approach (see the material and methods section for details), employing either purification by categories (A) or by intercepts (B); in the latter case, three different environmental temperatures (15°, 20° and 25°C) are considered. The shaded area shows the mean sleep period. Data are expressed as Mean ± SEM. The values for Mesor, Amplitude and Acrophase as well as the %V of wrist temperature demasked by the simulated constant routine are expressed as Mean±95% Confidence Interval and are included at the bottom of the graph. * indicates p<0.001 according to the cosinor analysis.
Figure 5.
Circadian modulation of masking variables.
Circadian modulation of the polynomial coefficients for the different masking variables: (A) light and environmental temperature (Env Temp), (B) activity and position and (C) sleep. The independent term for WT is represented in D. The shaded area shows the mean sleep period. Note that the independent term corresponds to an environmental temperature of 0°C. All values are expressed as Mean ± SEM. See the materials and methods section for details.