Effect of intensity of short-wavelength light on electroencephalogram and subjective alertness

Abstract

Short-wavelength light is known to have an effect on human alertness in the night-time. However, there are very few studies that focus on the effect of intensity of light on alertness. This study evaluates the acute alerting ability of short-wavelength light of three different intensities (40 lux, 80 lux and 160 lux). Eight subjects participated in a 60-minute exposure protocol for four evenings, during which electroencephalogram (EEG) as well as subjective sleepiness data were collected. EEG power in the beta range was significantly higher after subjects were exposed to 160 lux light than after they were exposed to 40 lux, 80 lux light or remained in darkness. Also, the alpha theta power was significantly lower under 160 lux light then in darkness. These results show that the effect of intensity on alertness is not linear and further work should be done to investigate the threshold intensity that is required to produce an alerting effect.

Full text

Effect of intensity of short-wavelength
light on electroencephalogram and
subjective alertness
J Lin MSc , S Westland PhD and V Cheung PhD
School of Design, University of Leeds, Leeds, UK
Received 15 May 2019; Revised 22 July 2019; Accepted 1 August 2019
Short-wavelength light is known to have an effect on human alertness in the night-
time. However, there are very few studies that focus on the effect of intensity of
light on alertness. This study evaluates the acute alerting ability of short-
wavelength light of three different intensities (40 lux, 80 lux and 160 lux). Eight
subjects participated in a 60-minute exposure protocol for four evenings, during
which electroencephalogram (EEG) as well as subjective sleepiness data were
collected. EEG power in the beta range was significantly higher after subjects were
exposed to 160 lux light than after they were exposed to 40 lux, 80 lux light or
remained in darkness. Also, the alpha theta power was significantly lower under
160 lux light then in darkness. These results show that the effect of intensity on
alertness is not linear and further work should be done to investigate the threshold
intensity that is required to produce an alerting effect.
1. Introduction
Light has been shown to exert strong non-
visual effects on a range of biological
functions such as the regulation of human
circadian system. Exposure to light in the
evening and night-time, especially of short-
wavelength (but not necessarily limited to
short-wavelengths), has been shown to lead to
an increase in alertness in humans. This effect
is suggested to be related to circadian disrup-
tion. Circadian disruption is associated with
reduced levels of the hormone melatonin and
is primarily (though note that rods and
cones may also participate in this process
1
)
mediated by activation of the intrinsically
photosensitive retinal ganglion cells which
respond to short-wavelength light most
strongly. Exposure to light in the evening
can inhibit the production of melatonin which
otherwise would naturally build-up in the
body during the late-evening hours.
Studies to date have linked the alerting
effects of light to its ability to suppress
melatonin.
2
However, hormonal changes
might not be the only pathway mediating
the non-visual effects of light. More recent
studies have suggested that acute melatonin
suppression is not needed for light to evoke
alertness responses in humans. For example,
it has been shown that both short-wavelength
(blue) and long-wavelength (red) lights
increased alertness at night, as measured by
EEG, but only blue light suppressed mela-
tonin significantly.
3
Light exposure during
the day has no impact on modulating mela-
tonin,
4,5
though it has been shown to affect
objective and subjective alertness in the
afternoon – as measured by EEG and the
Karolinska Sleepiness Scale (KSS).
6
These
studies have clearly demonstrated that light
can increase alertness independent of light-
induced melatonin suppression.
Address for correspondence: J Lin, School of Design,
University of Leeds, Leeds LS2 9JT, UK.
E-mail: cm14jl@leeds.ac.uk
Lighting Res. Technol. 2020; 52: 413–422
ßThe Chartered Institution of Building Services Engineers 2019 10.1177/1477153519872801

Studies on the non-visual effects of light
have suggested that the alerting effect of light
exposure during the daytime is more modest
than its effect during night-time.
5
Bright light
exposure at night has been shown to reliably
increase alertness.
7
These findings suggest
that the impact of light on alertness is a
complex physical, physiological and psycho-
logical activity that can result from different
pathways. Although melatonin level is asso-
ciated with circadian rhythm, EEG power
fluctuations reflect the immediate neuroendo-
crine responses.
8
EEG has been one of the measures often
used to evaluate acute alertness change, since
light can have an impact on EEG measures
without affecting melatonin levels.
6
One study
investigated how 48-minute exposures to
three lighting conditions (red, blue and
dark) affected subjects and found that EEG
alpha and alpha theta power were both
lower after the exposure to red and blue
lights compared to the darkness condition.
9
Similarly, another study compared short-
wavelength light, long-wavelength light and
darkness, and found that alpha power was
significantly lower after 30 minutes under
both the short- and long-wavelength light
than remaining in darkness.
10
Red and blue
lights have been found to increase beta signals
and reduce sleepiness relative to preceding
dim light exposure.
11
By looking into individ-
ual EEG frequencies, it was also suggested
that short-wavelength light in particular
enhances high alpha activity.
12
With these
findings, it is generally agreed that a decrease
in low EEG frequencies (theta alpha) power
and an increase in high EEG frequencies
(beta) power are associated with an increase
in alertness.
Short-wavelength light is becoming a crit-
ical safety concern. For example, there have
been concerns about home lighting in the
evening, where tungsten filament lamps of
low correlated colour temperature (CCT)
have been replaced by solid-state lighting of
higher CCT; the use of emissive electronic
displays may also be responsible for the
increasing sleep problem.
13
Differences in
the properties of lighting were shown to
affect individuals in various ways. Past
studies have tried to define and quantify the
timing, illuminance levels, exposure duration
and wavelength distribution of the light
required to evoke alerting responses.
14
A study has looked at how three different
illuminances ranging from 3 lux to 9100 lux
affect EEG activity over 6.5 hours of expos-
ure, and a dose-response relationship was
found in subjective alertness and EEG
power.
15
Other studies have also demon-
strated a non-linear relationship between
light intensity and circadian shifts.
16,17
It is
generally agreed that brighter light has a
stronger alerting ability than dimmer light.
However, there is no consensus yet about the
threshold of light intensity needed to produce
these effects.
2. Method
The alerting effect of intensity of light in
humans has so far received little attention.
This work is concerned with exploring the
threshold intensity of light needed to evoke
acute alertness responses in humans at night-
time. Based on the notion that melatonin
suppression is not the only mechanism con-
tributing to light-induced alertness, this study
has measured both subjective and objective
alertness using KSS and EEG, respectively.
The objective, specifically, is to investigate the
effect of three intensities (40 lux, 80 lux and
160 lux) of a short-wavelength light (
max
¼
475 nm), compared to remain in darkness
(51 lux), on human alertness during the
evening.
2.1. Test participants
Nine participants (aged 28 3.4 years; five
females) were recruited for the within-subject,
414 J Lin et al.
Lighting Res. Technol. 2020; 52: 413–422

four-session study. All participants went
through a pre-screening procedure where
individual sleep/rise time was collected and
the daily consumption of nicotine, caffeine
and alcohol was reported. Smokers and those
who were rated as extreme late chronotypes
(e.g. those who went to bed after 1am) were
excluded. The study was approved by the
University of Leeds Ethics Committee and all
participants signed informed consent forms
prior to the study. An information sheet was
given and participants were asked to refrain
from caffeine and alcohol intake three hours
prior to the experiment, and to try to main-
tain a regular, constant sleep schedule during
the entire experimental period. As a result,
there is no report of any major health
problems or currently taking any medica-
tions. One participant did not finish all of the
experimental sessions, and their data was
excluded from further analysis. In total, the
results produced from eight participants (aged
28 3.6 years; five females) are reported here.
2.2. Lighting conditions
Light was delivered through 12 luminaires
(LEDs, provided by Thouslite Lighting
System) mounted in the ceiling of a room
with white walls and grey carpets. The light-
ing system provides spectral tunable lighting
based on multi-channel LED technology.
Participants were asked to sit under the light
whilst reading, with a white table in front of
them (Figure 1). The lighting measures were
taken within the flat reading area on the table.
Four light settings were used: a dim (CCT
2000 K, 51 lux) and three short-wavelength
lighting conditions. The short-wavelength
condition had a peak at about 480 nm and
was approximately Gaussian with a half-
width half-height of 35 nm. Three intensities
were 40 lux, 80 lux and l60 lux (1 lux). The
spectra of the three test lighting conditions
were measured with an X-Rite i1Pro spectro-
photometer (Figure 2). The a-opic irradiance
for each lighting was calculated according to
the new CIE S 026 /E:2018
18
(Table 1).
2.3. Experiment protocol
Each participant completed four sessions
over four nights, all starting at the same time
(8 p.m.). Participants were fitted with EEG
electrodes prior to the start of the exposure.
The order of the conditions (Dim, Blue 40 lux,
80 lux and 160 lux) was selected randomly for
each participant to avoid potential sequence
effects. Sessions were separated by at least
72 hours for the same participant to avoid
potential carry-over effects.
During each evening study EEG was con-
tinuously recorded over 60 minutes. EEG
data was collected using B-Alert Live
Software with wireless advanced brain moni-
toring (ABM) EEG device (X10 headset with
standard sensor strips). Recordings consisted
of EEG with nine electrode positions (Fz, Cz,
Figure 1 Lighting room showing the position of the
participant and the luminaires
Effect of short-wavelength light on alertness 415
Lighting Res. Technol. 2020; 52: 413–422

Poz, F3, F4, C3, C4, P3 and P4) and
two reference mastoid electrodes. The elec-
trode impedance test was performed each
time before experiment to ensure the
good conductivity between the scalp and
electrodes, thus the good quality of the
signal. The EEG signal was band-passed to
1 to 40 Hz and decontaminated using ABM’s
validated artefact identification and decon-
tamination algorithms which identify and
remove five artefact types: electromyogram,
electrooculogram, excursions, saturations
and spikes. Power spectral density (PSD)
was computed by performing Fast Fourier
Transform with application of a Kaiser
window. PSD of selected 1-Hz bins was
averaged after application of a 50% over-
lapping window across three one-second
overlays.
Under the Dim condition, participants
were kept in the dim light for 60 minutes.
Under the Blue condition, test lights were
energized for 40 minutes, preceded by a
20-minute dim (51 lux) period.
Subjective sleepiness was evaluated using
the KSS, a self-reporting scale that ranges
from 1 (‘extremely alert’) to 9 (‘very sleepy,
fighting sleep’).
19
This scale has previously
been shown to be sensitive to changes in
sleepiness and the alerting effects of
lights.
12,20
The KSS was rated three times
(every 20 minutes – at the 20th, 40th and 60th
minutes) during each session. Participants
were asked to rate themselves from 1 to 9,
according to their sleepiness. For the duration
of the 60 minutes the participants were free to
read a book. They were also asked to keep
their eyes open and reduce head movement
0
20
40
60
80
100
380 400 420 440 460 480 500 520 540 560 580 600 620 640 660 680
Relative spectral power (%)
Wavelength (nm)
40lx
80lx
160lx
Figure 2 Spectral power distributions of three test lighting conditions
Table 1 -Opic irradiance (mW/m
2
) for three lighting conditions and 1 lux daylight (for reference)
-opic irradiance for S-cone-opic M-cone-opic L-cone-opic Rhodopic Melanopic
40 lux 157.98 131.44 80.48 277.64 324.25
80 lux 340.38 273.25 167.00 580.18 678.81
160 lux 716.54 557.33 340.56 1187.21 1389.97
1 lux D65 0.82 1.46 1.63 1.45 1.33
416 J Lin et al.
Lighting Res. Technol. 2020; 52: 413–422

throughout the experiment. No other activ-
ities (e.g. using electronic devices, eating or
talking) were allowed (Figure 3).
3. Results
3.1. EEG
EEG measures collected from nine elec-
trode sites were averaged to produce overall
EEG PSD, and then grouped into the follow-
ing frequency bins: 5–9 Hz (theta alpha),
8–9 Hz (lower alpha), 11–13 Hz (higher
alpha), and 13–30 Hz (beta). In each fre-
quency range, EEG power averaged over the
40 minutes under test lighting was normalized
to the initial 20 minutes of dim light period.
One-way analysis of variance (ANOVA)
was performed using the normalized power
in each of the frequency ranges studied. Post-
hoc t tests (with Bonferroni corrections)
were used to further compare the signifi-
cance between lighting conditions. Analyses
were performed using IBM SPSS Statistics 25
and the results for beta ranges are listed in
Table 2.
One-way ANOVA revealed a close to
significant main effect of lighting condition
in the normalized theta alpha (F
3,28
¼2.785;
p¼0.059) and a significant main effect of
lighting condition in beta (F
3,28
¼7.571;
p¼0.001). No significant difference was
observed in lower alpha (F
3,28
¼0.477;
p¼0.701) or higher alpha (F
3,28
¼0.385;
p¼0.765) ranges. Post-hoc pairwise compari-
sons found significant differences between
Dim and 160 lux, 40 lux and 160 lux, 80 lux
and 160 lux in beta range (Table 2). Figure 4
shows the results of normalized power for
four lighting conditions in four frequency
ranges studied (where * indicates statistical
significance). Power in theta alpha was lower
after exposure to 40 lux and 160lux blue lights
than after remaining in the Dim condition.
Power in beta range was significantly higher
after exposure to 160 lux blue lights than after
exposure to the other three lighting
Elapsed Time (min)
KSS
KSS
KSS
0102030 405060
EEG
Blue, 40lx
Blue, 80lx
Blue, 160lx
Dim, <1lx
Figure 3 Experimental design
Table 2 Pairwise comparisons for EEG beta
power
Pairs Significance*
Dim–40 lux 1.000
Dim–80 lux 1.000
Dim–160 lux 0.001*
40–80 lux 1.000
40–160 lux 0.005*
80–160 lux 0.025*
*Statistically significant (p0.05).
Effect of short-wavelength light on alertness 417
Lighting Res. Technol. 2020; 52: 413–422

conditions. Compared to Dim, exposure to
160 lux blue light has also reduced lower
alpha power and increased high alpha power,
although these differences did not reach
statistical significance (p40.05). Figure 5
shows the individual EEG beta power values.
3.2. Subjective sleepiness (KSS)
Mean scores over the experimental condi-
tion were normalized to the initial Dim
session. One-way ANOVA was performed
and post-hoc t tests (with Bonferroni correc-
tions) were used to further compare the
significance between lighting conditions. The
results are listed in Table 3.
ANOVA revealed a significant difference
between Dim and 160 lux conditions. Figure 6
shows the results for the normalized KSS
scores under four lighting conditions (where *
indicates statistical significance). Mean score
in 160 lx condition was significantly lower
than score in Dim condition (a lower KSS
score means more alertness). Mean scores
under 40 lux and 80 lux conditions were lower
than score under the Dim, and higher than
score under the 160 lux condition, although
these differences did not reach statistical
significance (p40.05). Figure 7 shows the
individual KSS scores.
4. Discussion
This study investigated how exposures to
short-wavelength lights of three different
intensities (40 lux, 80 lux and 160 lux) affect
objective and subjective alertness during the
night-time. Results showed the effect of
intensity on EEG theta alpha (5–9 Hz) and
beta (13–30 Hz) powers. Exposure to 40 lux
and 160 lux lights reduced theta alpha power
compared to remaining in the Dim condition.
Exposure to 160 lux light significantly
increased beta power compared to the Dim,
40 lux and 80 lux light conditions. It has been
observed and agreed in many studies that the
decrease in theta alpha power and the
*
*
*
0.95
1.00
1.05
1.10
1.15
theta_alpha l_alpha h_alpha beta
Normalized power
Dim 40lx 80lx 160lx
Figure 4 Mean standard error of the mean normalized EEG power for four frequency ranges
Table 3 Pairwise comparisons for KSS
scores
Pairs Significance*
Dim–40 lux 0.240
Dim–80 lux 0.252
Dim–160 lux 0.024*
40–80 lux 0.973
40–160 lux 0.209
80–160 lux 0.198
KSS: Karolinska Sleepiness Scale.
*Statistically significant (p0.05).
418 J Lin et al.
Lighting Res. Technol. 2020; 52: 413–422

increase in beta power indicate greater alert-
ness. It has also been suggested that reduction
in EEG alpha power (8–12 Hz) is related
to alertness increase, and the increase in
high-frequency alpha activity especially is
associated with the circadian regulation of
arousal.
21
We did not find a significant effect
on lower or higher alpha power, which might
be due to the limited sample size. Considering
the practical difficulties in conducting the
study, we included eight participants, which is
relatively small (but nevertheless sufficient to
show some statistically significant results in
theta alpha and beta ranges). The results of
lower and higher alpha actually showed the
consistent trend with other results, although
these are not statistically significant. To
combine the findings in EEG together, this
study showed that exposure to 160 lux blue
light significantly increases alertness com-
pared to Dim, 40 lux and 80 lux conditions;
exposure to 40 lux and 80 lux also increase
alertness compared to Dim light, although
not as significant as 160 lux light.
Furthermore, 40 lux and 80 lux exposure did
not show a large difference between each
other.
In the result alpha power was split into two
parts. The full alpha range was initially
examined and there were no significant
results. However, in some studies, high
alpha power alone was examined (as high
alpha is also suggested to be a specific marker
for alertness). Therefore in this study, alpha
power was then split into higher alpha and
lower alpha for further analysis. As shown in
Figure 4, the two parts seem to have opposite
patterns. In lower alpha, Dim light condition
has the greatest power, whereas in higher
alpha, 160 lux light condition has the greatest
power. One possible explanation might be
that the lower part of alpha is close to the
theta range (where alertness is negatively
related to the power), and the higher part of
alpha is closer to beta (where alertness is
positively related to the power). However,
these results would need further study to
assess this possibility.
The subjective alertness (KSS) reported is
consistent with the above EEG results.
Participants rated themselves as sleepier in
0.85
0.95
1.05
1.15
1.25
Normalized power
Dim 40lx 80lx 160lx
Figure 5 Individual EEG power values in beta range (the
same symbol indicates the values obtained from the
same participant)
*
0.5
0.7
0.9
1.1
1.3
1.5
Dim 40lx 80lx 160lx
Mean KSS score
Figure 6 Mean standard error of the mean normalized
KSS scores (lower scores indicate greater alertness)
0
0.3
0.6
0.9
1.2
1.5
1.8
2.1
Normalized scores
Dim 40lx 160lx
80lx
Figure 7 Individual KSS scores (the same symbol indi-
cates the values obtained from the same participant)
Effect of short-wavelength light on alertness 419
Lighting Res. Technol. 2020; 52: 413–422

the Dim condition, and more alert under the
blue lights. With the Dim condition having
the highest score and 160 lux having the
lowest, this again shows a significant alerting
effect of 160 lux compared to Dim light. Both
40 lux and 80 lux exposure were rated in
between the Dim and 160 lux conditions, at
about the same level. In addition, participants
were asked to report changes on their bed
time and any other unusual feelings (including
a sleep problem) after experiments. As a
result, there is no report on unusual bed
time shifts after exposure; however, one
participant reported suffering from headache
at night after being kept in blue light.
Very few studies have so far looked into the
effect of intensity of short-wavelength light on
alertness. In some studies, blue light of 40 lux
has been suggested to have an alerting effect.
9,10
Some other studies, for example, have com-
pared light of 40 lux at 2500 K, 3000 K and
6500 K,
22
and light of 295lux at 2700 K and
209 lux at 5600 K.
23
These studies, however,
have focussed on spectral composition or
exposure duration, rather than light intensity.
This study extends the research on the
effect of intensity on light-induced alertness.
The findings suggested that, firstly, a certain
level of intensity is needed to induce alertness
(to produce a significant effect a higher level
of intensity might be needed). Secondly, the
relations between intensity and alertness
change is not linear, which certain param-
eters, e.g. the intensity that could maximally
increase alertness might be identified. Lastly,
these threshold intensities could be difficult to
measure. Assuming that the relations between
intensity and alertness are not linear, the closer
to the threshold, the smaller the difference
would be. Study of higher precision is required
in order to detect smaller effect, e.g. a bigger
sample size. The practical difficulties in con-
ducting such experiment limit the sample size
that is normally included, whereas the small
sample size further limits the statistical power
to detect the subtle effect. The other limitation
is that reading a book during the experiment
might have affected participants’ brain activ-
ity, although this is difficult to avoid. In
similar studies, participants will normally be
allowed to conduct light tasks such as cross-
word puzzles
11
over the relatively long exposure
hours. However, they were asked to read
continuously and try to maintain their mood
and attention throughout the session, so to
control the potential effect as much as possible.
5. Conclusions
This study provides some evidence that short-
wavelength light exposure in the evening can
increase human alertness and that this can occur
relatively quickly (even though some other
studies have suggestion that melatonin inhib-
ition, for example, may have a longer time
course). Both objective and subjective results
also suggest that for the lighting conditions
tested in the present study, light of higher
intensity has a stronger alerting effect than light
of lower intensity. These findings, in themselves,
do not enable a threshold effect to be identified.
However, the methodology described in this
study may provide a basis for future on-going
work to address this question explicitly.
Declaration of conflicting interests
The authors declared no potential conflicts of
interest with respect to the research, author-
ship, and/or publication of this article.
Funding
The authors received no financial support for
the research, authorship, and/or publication
of this article.
ORCID iD
J Lin https://orcid.org/0000-0002-1633-
4553
420 J Lin et al.
Lighting Res. Technol. 2020; 52: 413–422

References
1 Hattar S, Lucas RJ, Mrosovsky N, Thompson
S, Douglas RH, Hankins MW, Lem J, Biel M,
Hofmann F, Foster RG, Yau KW.
Melanopsin and rod–cone photoreceptive sys-
tems account for all major accessory visual
functions in mice. Nature 2003; 424: 75.
2 Figueiro MG, Bullough JD, Bierman A, Fay
CR, Rea MS. On light as an alerting stimulus
at night. Acta Neurobiologiae Experimentalis
2007; 67: 171.
3 Figueiro MG, Bierman A, Plitnick B, Rea MS.
Preliminary evidence that both blue and red
light can induce alertness at night. BMC
Neuroscience 2009; 10: 105.
4 Figueiro MG, Rea MS. The effects of red and
blue lights on circadian variations in cortisol,
alpha amylase, and melatonin. International
Journal of Endocrinology. DOI: 10.1155/2010/
829351.
5 Figueiro MG, Rea MS. Sleep opportunities
and periodic light exposures: impact on bio-
markers, performance and sleepiness. Lighting
Research and Technology 2011; 43: 349–369.
6 Sahin L, Figueiro MG. Alerting effects of
short-wavelength (blue) and long-wavelength
(red) lights in the afternoon. Physiology and
Behavior 2013; 116: 1–7.
7 Badia P, Myers B, Boecker M, Culpepper J,
Harsh JR. Bright light effects on body tem-
perature, alertness, EEG and behavior.
Physiology and Behavior 1991; 50: 583–588.
8 Jung TP, Makeig S, Stensmo M, Sejnowski TJ.
Estimating alertness from the EEG power
spectrum. IEEE Transactions on Biomedical
Engineering 1997; 44: 60–69.
9 Figueiro MG, Sahin L, Wood B, Plitnick B.
Light at night and measures of alertness and
performance: implications for shift workers.
Biological Research for Nursing 2016; 18:
90–100.
10 Okamoto Y, Rea MS, Figueiro MG. Temporal
dynamics of EEG activity during short-and
long-wavelength light exposures in the early
morning. BMC Research Notes 2014; 7: 113.
11 Plitnick B, Figueiro MG, Wood B, Rea MS.
The effects of red and blue light on alertness
and mood at night. Lighting Research and
Technology 2010; 42: 449–458.
12 Lockley SW, Evans EE, Scheer FA, Brainard
GC, Czeisler CA, Aeschbach D. Short-wave-
length sensitivity for the direct effects of light
on alertness, vigilance, and the waking elec-
troencephalogram in humans. Sleep 2006; 29:
161–168.
13 Chang AM, Aeschbach D, Duffy JF, Czeisler
CA. Evening use of light-emitting eReaders
negatively affects sleep, circadian timing, and
next-morning alertness. Proceedings of the
National Academy of Sciences 2015; 112:
1232–1237.
14 Cajochen C. Alerting effects of light. Sleep
Medicine Reviews 2007; 11: 453–464.
15 Cajochen C, Zeitzer JM, Czeisler CA, Dijk DJ.
Dose-response relationship for light intensity
and ocular and electroencephalographic cor-
relates of human alertness. Behavioural Brain
Research 2000; 115: 75–83.
16 Boivin DB, Duffy JF, Kronauer RE, Czeisler
CA. Dose-response relationships for resetting
of human circadian clock by light. Nature
1996; 379: 540.
17 Zeitzer JM, Dijk DJ, Kronauer RE, Brown
EN, Czeisler CA. Sensitivity of the human
circadian pacemaker to nocturnal light:
melatonin phase resetting and suppression.
The Journal of Physiology 2000; 526:
695–702.
18 Commission Internationale de l’E
´clairage. CIE
System for Metrology of Optical Radiation for
ipRGC-Influenced Responses to Light. CIE S
026/E:2018. Vienna: CIE, 2018.
19 Shahid A, Wilkinson K, Marcu S, Shapiro
CM. Karolinska sleepiness scale (KSS). In:
STOP, THAT and One Hundred Other Sleep
Scales. New York: Springer, 2011.
20 Cajochen C, Brunner DP, Kra
¨uchi K, Graw P,
Wirz-Justice A. EEG and subjective sleepiness
during extended wakefulness in seasonal
affective disorder: circadian and homeostatic
influences. Biological Psychiatry 2000; 47:
610–617.
21 Aeschbach D, Matthews JR, Postolache TT,
Jackson MA, Giesen HA, Wehr TA. Two
circadian rhythms in the human electro-
encephalogram during wakefulness. American
Journal of Physiology-Regulatory, Integrative
and Comparative Physiology 1999; 277:
R1771–1779.
Effect of short-wavelength light on alertness 421
Lighting Res. Technol. 2020; 52: 413–422

22 Chellappa SL, Steiner R, Blattner P, Oelhafen
P, Go
¨tz T, Cajochen C. Non-visual effects of
light on melatonin, alertness and cognitive
performance: can blue-enriched light keep us
alert? PloS One 2011; 6: e16429.
23 Nagare R, Plitnick B, Figueiro MG. Effect of
exposure duration and light spectra on night-
time melatonin suppression in adolescents and
adults. Lighting Research and Technology 2019;
51: 530–543.
422 J Lin et al.
Lighting Res. Technol. 2020; 52: 413–422