Skip to main content
. 2014 Feb 1;37(2):271–281. doi: 10.5665/sleep.3396

Figure 2.

Figure 2

Effects of daytime and nighttime monochromatic light exposure on behavioral measures. Mean (± standard error of the mean) profiles of subjective sleepiness (A-C), auditory reaction time (D-F), and auditory attentional lapses (G-I) before, during, and after exposure to monochromatic light for 6.5 hours (open box). Behavioral test data were analyzed during light exposure including the first cognitive test immediately after the end of light exposure (A, B, D, E, G, H) and only during the light exposure (C, F, I). Karolinska Sleepiness Scale (KSS) ratings and performance parameters (mean reaction time and lapses [defined as responses > 500 ms]) measured during monochromatic light exposure were z-transformed and subjected to two-way repeated measures with random intercepts mixed model analysis of variance with restricted maximum likelihood estimation method followed by Dunnett post hoc multiple comparison test for comparing between groups (A, B, D, E, G, H) and one-way generalized linear model analysis followed by Student-Newman-Keuls post hoc multiple comparison test for comparing between groups (C, F, I). Fixed effects were set as spectral condition and/or exposure duration as appropriate. Exposure to 460 nm monochromatic light during the day did not change subjective sleepiness ratings (A) but improved reaction times (D) and reduced attentional lapses (G) relative to 555-nm exposure. Exposure to 460 nm monochromatic light at night significantly reduced subjective sleepiness ratings (B), reaction time (E), and attentional lapses (H) as compared with 555-nm light. Subjective sleepiness was not different between any of the four groups during light exposure (C). Reaction times were lowest under daytime 460-nm exposure, followed by daytime 555 nm and nighttime 460 nm and highest under nighttime 555 nm (F). Attentional lapses were not different between daytime 460 nm, daytime 555 nm, and nighttime 460 nm and all three were significantly lower than nighttime 555 nm (I). Significant differences between condition means are represented by * and † (P < 0.05). Panels B, E and H were published previously in reference 7.