Which EEG Electrodes Should Be Considered for Alertness

Assessment?

Agnieszka Wolska

, Dariusz Sawicki

, Marcin Kołodziej

, Mariusz Wisełka

and

Kamila Nowak

Central Institute for Labour Protection - National Research Institute (CIOP-PIB), Warsaw, Poland

Warsaw University of Technology, Institute of Theory of Electrical Engineering,

Measurements and Information Systems, Warsaw, Poland

Keywords: EEG, Alertness, Exposure to Light, Alpha and Beta Ranges, Electrodes Selection for Analysis.

Abstract: The analysis of EEG signal is one of the objective methods used in alertness assessment. Many publications

confirm the correct assessment of alertness level based on the analysis of selected brain waves. EEG

registration is a difficult task; one of the important problems is the necessity to choose which EEG electrode

to download the signal for analysis. The authors use different electrodes, often without justifying the choice.

Equally often, the only justification is to say that the analyzed signal was the strongest among those available,

or the least contaminated with artifacts. The aim of the article is to try to answer the question: signals which

electrodes (channels) should be included in the alertness assessment. 33 participants took part in the

experiment. Blue and red light was used to stimulate alertness. The impact of such light is documented in

many publications. Alertness changes due to specific color of light were evaluated – the changes of alpha and

beta bands were analyzed. Statistical analysis has shown that for alertness assessment the following electrodes

should be considered: C3 and FC1 for alpha band and F3 and FP1 – for beta band signals.

1 INTRODUCTION

Apart from the visual response to light (light enables

us to see), the non-visual response to light (melatonin

suppression, core body temperature regulation,

alertness and cognition, circadian clock changes) has

been examined in detail since early 2000, when a new

photoreceptor - intrinsically photosensitive retinal

ganglion cell (ipRGc) containing the melanopsin had

been discovered. The non-visual response depends on

the light wavelength and intensity (irradiance level at

the eye), time and duration of exposure. It was

scientifically proven that light of particular

wavelengths is able to affect human health,

physiological and psychological behavior and

wellbeing (Bellia et al., 2011, Wolska et al, 2018,

Łaszewska et al., 2017, Sahin et al., 2014, Cajochen

et al., 2005, 2007, 2010). Many studies confirmed that

exposure to blue or red light increases the level of

https://orcid.org/0000-0003-3912-605X

https://orcid.org/0000-0003-3990-0121

https://orcid.org/0000-0003-2856-7298

https://orcid.org/0000-0002-7145-6457

alertness (Figueiro et al., 2009, 2016, Figueiro & Rea,

2010, Plitnick et al., 2010, Sahin & Figueiro, 2013,

Okamoto et al., 2014, Łaszewska et al., 2017,

Scheuermaier et al, 2018, Iskra-Golec et al , 2017,

Phipps-Nelson at al, 2009).

Maintaining a high level of alertness is a very

important factor on many workstations, especially

where the human error could result in occupational

accident or threat to life or health of many people. The

new knowledge of the dual role of light contributed

in numerous studies concerning influence light on

alertness level. Among the methods of objective

alertness assessment, EEG signal analysis seems to be

the most frequent used and relatively easy to use.

From the EEG signal it is possible to differentiate

bands: alpha (8-12 Hz), beta (13-30 Hz), delta (0.5-4

Hz), and theta (4.5-8 Hz) using fast Fourier transform.

The EEG signal is closely related to the activity of the

person. As the activity increases, the EEG shifts to

higher dominating frequency and lower amplitude.

Wolska, A., Sawicki, D., Kołodziej, M., Wisełka, M. and Nowak, K.

Which EEG Electrodes Should Be Considered for Alertness Assessment?.

DOI: 10.5220/0008168600400049

In Proceedings of the 3rd International Conference on Computer-Human Interaction Research and Applications (CHIRA 2019), pages 40-49

ISBN: 978-989-758-376-6

(Malmivuo and Plonsey, 1995). In the awaking state

(and with the eyes open) theta and alpha waves

practically does not exist or their amplitudes are

minimal (Sahin and Figueiro, 2013, Baek and Min,

2015, Klimesch, 2012, Okamoto, 2014, Sawicki et

al., 2016). The increase of their amplitude in the

awaking state is interpreted as the decrease of

alertness and vice versa. Additionally sometimes

beta, and delta bands are also taken into account in

studies on alertness.

However, a unified or standardized method of

analyzing the EEG signal for this purpose does not

exist. The differences concern both the EEG

registration (EEG equipment, number and placement

of electrodes, type of electrodes etc.) and analysis of

EEG signal for the alertness assessment (procedure of

raw signal filtering – pre-processing, artifacts

removing method, digital processing of cleaned

signal, mathematical interpretation of recorded time

series, spectral analysis, and selection of the

electrodes for the analysis etc.). As it was mentioned

before, for the purpose of assessing alertness after

exposure to particular light usually two or three EEG

bands are considered. Alpha, theta and beta (Lavoie

et al., 2003, Figueiro et al., 2009, Plitnick et al., 2010,

Sahin at al., 2013, Yokoi et al., 2003), alpha, theta and

delta (Phipps-Nelson et al., 2009, Iskra-Golec et al.,

2017), alpha and theta (Sawicki et al., 2016), alpha

and beta (Scheuermaier et al., 2018). However some

studies on vigilant attention and sleeping considered

all bands and even for particular frequencies of those

bands (Chellappa et al, 2017).

And even the same equipment for EEG

registration and placement of electrodes according to

the International 10-20 system is used and the same

frequency ranges are analysed – the brain sites for

analysis (what corresponds with particular electrodes

placement on the scalp) are often different. Based on

many studies on EEG registration for alertness level

assessment some examples of electrodes sets could be

specified:

 midline central: Fz, Cz, Pz and Oz (Łaszewska et

al., 2017, Yokoi at al., 2003, Figueiro, 2009);

 motor cortex: C3 and C4 (Lavoie et al., 2003);

 frontal and occipital from midline central: Fz and

Oz (Donskaya et al., 2012);

 anterior temporal lobes (T3, T4), parietal lobe (P3,

P4), occipital lobe (O1, O2) (Phipps-Nelson et

al.,2009);

 motor cortex (C3, C4), midline central (Fz, Cz, Pz

and Oz) (Scheuermaier et al., 2018);

 motor cortex (C3, C4), parietal lobe (P3, P4),

occipital lobe (O1, O2) and frontal lobe (F3, F4)

(Chellappa et al., 2017);

 left frontal lobe (Fp1, F7, F3), right frontal lobe

(Fp2, F4, F8), motor cortex (C3, Cz, C4), left

temporal lobe (T3 , T5), right temporal lobe (T4,

T6), anterior temporal lobes (T3, T4), posterior

temporal lobes (T5, T6), parietal lobe (P3, Pz, P4),

occipital lobe (O1, O2), and midline central (Fz,

Cz, Pz) (Iskra-Golec et al., 2017).

Besides, sometimes the selection of electrodes is

simply based on the strongest signal (the electrodes

which provide the higher signal for particular band,

regardless of their placement on the scalp). The

significant effect of the registration channel

(electrodes) on alpha, theta and beta power was stated

by Figueiro (Figueiro et al., 2009). It means that

electrodes selected for the final analysis could

influence the interpretation of alertness. There is no

knowledge about how the selection of particular

electrodes influences the interpretation of alertness

level based on particular bands of EEG signal. It

would be interesting to find out which electrodes

(channels) seem to be best correlated with changes of

alertness level.

The aim of the article is to answer the question: is

it possible to indicate EEG electrodes for signal

processing and alertness interpretation based on alpha

and beta bands in experiments with exposure to blue

and red light? This paper presents the proposition of

electrodes selection for alertness assessment based on

results obtained during the study carried out using the

32 electrodes EEG registration with exposure to blue

and red light.

2 EXPERIMENT

2.1 Participants

Thirty three right-handed male volunteers, aged 20-

30 years (mean age 23,4, SD=2,12 years) participated

in the study: 17 of extreme morning chronotype and

16 of extreme evening chronotype, identified using

Composite Scale of Morningness – CSM (Smith et al.,

1989, Jankowski, 2015). All participants were paid

and the following exclusion criteria were applied:

sleep or mental health problems, color blindness,

glasses to work with a computer. The experimental

protocol was reviewed and approved by the Senate

Committee of Research Ethics of Józef Piłsudski

University of Physical Education in Warsaw.

Informed written consent was obtained from each

participant.

Which EEG Electrodes Should Be Considered for Alertness Assessment?

2.2 Light Exposure

The experiment was conducted in a windowless, air-

conditioned laboratory room of white walls, with

general indirect white LED lighting operated by

lighting control system. The general lighting was

used for dim light conditions i.e. < 5 lx at the eyes.

Additionally two desk LED luminaires, specially

designed for exposure to blue and red light, were

positioned, according to Alkozei et al. (Alkozei et al.,

2016), at 80 cm distance from the participant’s

nasion, with each light cantered at a 45 degree angle

from midline. The established illuminance level for

exposure was 40 lx both for red (630 nm) and blue

light (465 nm). The technical aspects of light

exposure are provided in (Wolska et al., 2018). The

study was carried out during the winter season.

2.3 Procedure

The experimental session started respectively at 7:30

am for morning chronotype and at 11:00 am for

evening chronotype, similarly to study of Maierova

(Maierova et al., 2016). The subjects were asked to

maintain a fixed regular plan sleep, lasting at least 7

hours during the week preceding the start of the

experiment. Every participant took part in two

experiments, each with exposure to different light.

The session order was counterbalanced for each

individual, to avoid the impact of familiarizing with

the procedure on results. One week interval between

the experiments was established (Wolska et al.,

2018). During each experiment participants

underwent 30 min under dim light (< 5 lx), 20 min

behavioral tests with EEG registration (Resting 1,

GoNoGo 1 and n-back 1), 30 min under blue or red

light exposure (no tests and EEG registration

performed), 70 min behavioral tests with EEG

registration (Resting 2, GoNoGo 2, n-back 2,

GoNoGo 3, n-back 3, GoNoGo 4, n-back 4, Resting

3). All EEG registrations were performed under dim-

light. The resting state (plus symbol (“+”) presented

on the screen for 3 minutes together with EEG

registration) was executed three times:

(1) after dim light and just before exposure to blue or

red light,

(2) just after exposure to particular light (acute

alerting effect) and

(3) 70 minutes after exposure (sustained acute

alerting effect).

2.4 EEG Registration

EEG measurements were taken using 256- channel

g.Hlamp amplifier (Guger Technologies, Graz,

Austria). Signal was recorded from 32 electrodes

paced according to the 10-20 International system

(Figure 1). All impedances were kept below 30kΩ

during the whole recording session. A Simulink

model (running under Matlab 2014a) was used to

control the registration of the signal. It consisted of

the building block provided by the manufacturer of

the system (Guger Technologies, Graz, Austria)

(Wolska et al., 2018).

3 DATA ANALYSIS

EEG recordings were taken from 32 electrodes and

then analyzed in order to assess the changes in

alertness level: “short term”: before exposure (1) and

after exposure to light (2), “long term”: before

exposure (1) and after exposure (3). This article

focused on changes for alpha and beta frequency

bands on all 32 electrodes – the layout in Figure 1.

The alertness level assessment based on changes in

energy in alpha and beta bands is based on the

following assumptions: the higher energy in alpha

band the lower level of alertness, the higher energy in

beta band the higher alertness level.

Figure 1: The layout of 32 electrodes used in the

experiment.

3.1 Preprocessing

The EEG recordings were filtered using the same

processing method for each registration. Band pass

FIR filtration was applied in order to eliminate

CHIRA 2019 - 3rd International Conference on Computer-Human Interaction Research and Applications

frequencies higher than 32 Hz and lower than 0.01 Hz

for all 32 channels. This had been done to ensure that

neither occasional electrical grid impact and higher

frequencies harmonics nor low frequencies events are

included in signal. Furthermore recordings were

visually checked for artifacts (due to blinking,

movements etc.) and manually marked and removed.

In case of presence of multiple artifacts in signal

interpolation (based on 2 neighboring – closest

channels) was applied. Maximum of 4 channels were

interpolated and in case of 5 or more channels

significantly affected by artifacts data set was marked

as rejected from further analysis. At the end Infinite

Component Analysis (ICA) was performed for each

data set individually. After filtration each data set was

analyzed.

3.2 Feature Extraction

The initially prepared signal was digitally processed.

To remove signals outside the alpha and beta bands

and then estimate the energy in those bands, a 4th

order lowpass Butterworth filter was applied.

Afterwards signal was divided into 5 seconds

windows. The number of windows depended on the

correctly recorded registration time (after artifacts

removal). As a result, 9 to 30 5-second windows were

obtained, which corresponds to obtaining 9 to 30

element sample sets for individual bands (alpha and

beta). For each window of signal the energy was

calculated using the variance calculation – this way

power analysis was performed. This digital

processing procedure was applied for each

participant, for each electrode of the recorded signal

and for interaction with two colours of light (red and

blue). Considering the fact, that:

 three independent signal registrations were

carried out (resting state (1), (2) and (3)) for each

experimental session,

 33 subjects were examined (morning chronotype

– with assigned codes: R01 - R17, evening

chronotype – with assigned codes: W01 - W16),

 the interaction of two color of light was used,

 signal was registered on 32 electrodes,

 EEG signal energy was calculated in two

independent alpha and beta bands,

3x33x2x32x2=12672 sets of 9 to 30 element signal

samples were obtained.

3.3 Statistical Analysis

The energy samples sets were subjected to further

statistical analysis. For each set of samples (before

exposure (1), after exposure (2) and after exposure

(3)) the mean values of energy (M1, M2, M3) and

standard deviation (STD1, STD2, STD3) were

calculated. Statistics were calculated using the

Student's t-test. The Student's t-test compares pairs of

sets between different registrations. Two types of

comparisons were made: between the registration of

resting 1 (R1) and 2 (R2) and between the

registrations resting 1 (R1) and 3 (R3) - marked as

R12 and R13 respectively. The null hypothesis H0

was assumed that the means in the compared sets of

signal samples of the analyzed resting states do not

differ statistically significant (no effect of light on

energy values) and the alternative hypothesis Ha, that

the means in the compared sets of samples differ

significantly (effect of light on energy values). The

significance level of rejecting the null hypothesis was

assumed as α=0.1. If the value of the test statistic

(Student's t-value) falls in the rejection region the null

hypothesis H0 is rejected in favor of the alternative

hypothesis Ha. The statistical analysis was carried out

using the Matlab environment (2018a). The built-in

functions mean, std, ttest2 were used for statistical

calculations.

An exemplary set of results for testing the mean

energy values of a resting signal for one electrode Oz

is presented in Table 1. Taking into account the

results of Student's t-tests, a “resultant” measure of

significance of a given difference (H12: between (1)

and (2), H13: between (1) and (3)) was introduced.

The following values for that measure were assumed:

1 – statistically significant difference, 0 – not

significant difference. T12, T13 are the values of t-

Student’s statistics t, p12, p13 are the statistical

significances.

The results of the sum of measures H12 and H13

(assigned as HS12 and HS13) indicate that after

exposure to blue light there were statistically

significant changes in energy recorded on the

electrode Oz between resting state (R1) and (R2) in 9

subjects and between resting state (R1) and (R3) in 10

subjects with evening chronotype.

The above presented analysis allows selecting the

electrodes on which the energy differences

statistically significant in the alpha or beta bands

between the R12 and R13 resting states occured most

often. For this purpose, for each electrode, for each

band and each difference (R12 and R13) a sum of

“resultant” measures of significance H12 and H13

was calculated (as HS12 and HS13 respectively) and

presented in Table 1. The higher the measure HS12

or HS13 of the individual electrode, the more the

differences in the energy of the analyzed signals of a

given electrode were statistically significant. Because

Which EEG Electrodes Should Be Considered for Alertness Assessment?

Table 1: An exemplary set of energy results obtained for the Oz electrode and alpha band (evening chronotype W and blue

light N). M1, M2, M3 – the value of energy in [V

]. STD1, STD2, STD3 – standard deviations. T12, T13 are the values of

t-Student’s statistics t, p12, p13 are the statistical significances. The significance level of rejecting the null hypothesis was

assumed as α=0.1. If p12<α , H12=1, otherwise H12=0, for p13 and H13 in the same way.

Subject H12 H13 M1 M2 M3 STD1 STD2 STD3 T12 T13 p12 p13

W01N 0 1 2.575 69.132 0.879 0.703 301.779 0.223 -1.011 10.298 0.318 0.001

W02N 1 1 1.812 2.459 1.612 0.392 1.812 0.231 -1.780 1.748 0.081 0.089

W03N 1 1 1.472 2.163 2.245 0.340 0.523 0.428 -5.779 -6.509 0.000 0.001

W04N 0 1 65.068 59.924 40.203 21.023 15.757 24.512 0.959 3.433 0.343 0.001

W05N 0 0 11.046 15.037 9.535 4.743 12.679 2.744 -1.271 1.240 0.211 0.223

W06N 1 1 15.398 11.023 11.689 3.855 3.330 3.553 4.285 3.779 0.000 0.000

W07N 1 0 1.508 3.239 1.494 0.369 1.021 0.333 -7.877 0.142 0.000 0.888

W08N 1 1 7.496 11.464 4.779 2.152 3.984 1.339 -4.639 5.462 0.000 0.001

W09N 1 0 29.627 20.426 22.842 15.384 8.437 8.974 2.582 1.660 0.013 0.104

W10N 0 0 3.154 2.992 3.052 0.802 0.690 1.420 0.667 0.329 0.508 0.743

W11N 1 0 2.073 2.434 2.256 0.380 0.682 0.305 -2.230 -1.302 0.031 0.202

W12N 1 1 46.288 60.325 61.529 12.119 16.229 14.805 -3.155 -3.916 0.003 0.001

W13N 0 1 3.855 3.292 5.616 1.983 0.687 2.340 0.866 -2.640 0.394 0.012

W15N 0 1 6.737 5.940 14.288 3.418 3.874 7.502 0.726 -4.695 0.472 0.001

W16N 1 1 5.834 4.705 3.087 2.477 2.085 1.242 1.794 5.103 0.079 0.001

Sum HS12= 9 HS13=10 -

the analyzed differences R12 and R13 are associated

with stimulation of light of a specific color, the

analysis allows selecting the electrodes on which the

response to stimulation with a specific color is the

strongest.

4 RESULTS AND DISCUSSION

The influence of blue and red light on energy

significant changes in alpha and beta bands was

observed. Only statistically significant cases of the

influence of light on energy were analyzed. It is worth

noting that the significance of the interaction of light

with signals from specific electrodes was confirmed

by the Student's t-test in all considered groups, i.e.:

the interaction of blue light (N) in the morning and

evening chronotype, the interaction of red light (C) in

the morning (R) and evening (W) chronotype. The

number of cases of significant changes in energy

(both between R12 and R13 resting states) on

individual electrodes in alpha and beta band is

presented in Table 2. The highest values of

HS12+HS13 measure were marked in red color (for

red light exposure) and blue color (for blue light

exposure) in Table 2. Also the electrodes for which

the highest values of that measure were found have

been marked in green color (for alpha band) and in

violet color (for beta band).

The difference between signal channels

(electrodes) response to blue and red color in alpha

and beta bands has been found. The visualizations of

the number of significant differences (HS12 + HS13)

in alpha and beta bands on a scalp are shown as the

impact maps in Figure 2 after blue light exposure and

in Figure 3 after red light. When analyzing impact

maps for both chronotype groups, it should be noted

that these maps are different. This means that it is

difficult to indicate one universal electrode for all

groups.

The impact maps presented in Figures 2 and 3 are

clearly asymmetrical. The maximum HS12+HS13 for

alpha bands (both for red and blue light) is shifted

from the central midline to the left motor cortex area.

CHIRA 2019 - 3rd International Conference on Computer-Human Interaction Research and Applications

Table 2: The sum of HS12 and HS13 measures for individual electrodes for both color lights stimulation among evening and

morning chronotypes, and for the whole group. The highest values of HS12+HS13 measure were marked in red color (for red

light exposure) and blue color (for blue light exposure). The electrodes for which the highest values of that measure were

found have been marked in green color (for alpha band) and in violet color (for beta band).

Electrode

HS12+HS13

Alpha band Beta band

Morning

chronot

Evening

chronot

Morning+evening

chronot

Morning

chronot

Evening

chronot

Morning+evening

chronot

Red

Blue

Red

Blue

Red

Blue

Red

Blue

Red

Blue

Red

Blue

C3 22 17 21 19 43 36 17 21 21 17 38 38

C4 20 19 19 20 39 39 12 15 19 19 31 34

CP1 19 19 18 12 37 31 20 17 22 18 42 35

CP2 18 18 18 15 36 33 16 18 16 15 32 33

CP5 17 20 19 17 36 37 19 22 26 18 45 40

CP6 19 17 20 19 39 36 19 23 20 19 39 42

CZ 21 18 14 15 35 33 20 22 19 18 39 40

F3 19 17 17 21 36 38 23 24 22 22 45 46

F4 20 21 17 17 37 38 17 20 21 19 38 39

F7 17 21 18 12 35 33 19 21 20 20 39 41

F8 22 22 17 13 39 35 18 24 24 20 42 44

FC1 17 23 17 20 34 43 16 24 18 19 34 43

FC2 21 20 15 19 36 39 16 18 18 18 34 36

FC5 17 19 16 17 33 36 22 24 18 19 40 43

FC6 21 22 16 12 37 34 22 22 24 17 46 39

FP1 16 18 18 16 34 34 21 26 25 23 46 49

FP2 14 19 16 16 30 35 17 22 22 19 39 41

FZ 19 19 19 18 38 37 15 22 18 23 33 45

O1 15 19 20 19 35 38 19 22 16 22 35 44

O2 16 20 15 19 31 39 16 22 18 18 34 40

OZ 17 22 15 19 32 41 18 22 19 18 37 40

P3 18 17 17 16 35 33 18 18 17 17 35 35

P4 18 18 16 17 34 35 15 20 19 16 34 36

P7 21 19 19 19 40 38 22 22 19 21 41 43

P8 18 17 16 20 34 37 19 25 17 21 36 46

P09 17 17 18 14 35 31 20 22 23 22 43 44

PO10 17 17 17 15 34 32 24 22 18 20 42 42

PZ 18 20 15 14 33 34 15 17 16 16 31 33

T7 16 19 14 14 30 33 24 25 19 18 43 43

T8 15 18 21 12 36 30 22 22 23 19 45 41

TP9 18 18 18 17 36 35 22 21 20 21 42 42

TP10 20 20 18 16 38 36 21 20 19 18 40 38

Which EEG Electrodes Should Be Considered for Alertness Assessment?

All participants in the study were right-handed.

However, it is difficult to draw conclusions about

lateralization of brain function influence on the brain

wave effects, although statistically confirmed cases

are considered. There were no studies carried out on

a group of left-handed. However, this indicates the

need to continue research with particular emphasis on

this aspect.

Considering the influence of visual stimuli on

alpha waves, O1 and O2 electrodes are often

indicated as the most appropriate for recording such

waves. In none of the presented results has this been

confirmed. Moreover, both for W and R chronotypes

and red light, and R chronotype and blue light, the O1

and O2 electrodes were those where the values of

HS12+HS13 measure were relatively small. Only in

the case of W chronotype and blue light exposure

values of that measure were above average.

Analyzing the presented results it would be

possible to propose the use of C3 electrode to assess

the interaction of red light and the FC1 electrode to

assess the interaction with blue light. These

electrodes are located very close together on the

scalp. Perhaps a good solution would be to use the

average signal from electrodes C3 and FC1 for

alertness assessment with blue light interaction. It

would be worthwhile to continue the tests by

analyzing the signals of other electrodes in this area

(C3, FC1). An interesting solution would be to take

Morning chronotype, alpha Evening chronotype, alpha

Morning chronotype, beta Evening chronotype, beta

Figure 2: The distribution of significant differences (HS12 + HS13) number in alpha and beta bands on a scalp after exposure.

CHIRA 2019 - 3rd International Conference on Computer-Human Interaction Research and Applications

Morning chronotype, alpha Evening chronotype, alpha

Morning chronotype, beta Evening chronotype, beta

Figure 3: The distribution of significant differences (HS12 + HS13) number in alpha and beta bands on a scalp after exposure

to red light.

into account the signals from the EEG registration

with a larger number of electrodes.

Beta band is less frequently used in alertness

analysis based on EEG registration. However, the

analysis of the presented results shows that the

maximum of significance measures HS12+HS13 for

beta waves on specific electrodes are higher than the

for alpha band. For beta waves, the electrodes from

the area around central midline do not give high

HS12+HS13 and thus seems not to be the best to the

alertness analysis caused by different colors of light.

The highest significance measures HS12+HS13 were

noted for the electrodes at the area of left frontal lobe

(F3 and FP1), both for red and blue light. That’s why

it seems reasonable to propose the use of F3 and FP1

electrodes to assess the interaction of red light and

blue light. HS12+HS13 on these electrodes are the

highest for all participants (both R and W chronotype)

and for interaction with red and blue light.

5 CONCLUSIONS

The conducted study has shown that electrodes can be

selected to assess alertness on the basis of EEG signal

analysis. In the current research, the researchers used

different approaches to the selection of electrodes.

Reasonable reasons are signal strength or its purity

(no interference or artifacts). However, signals not on

all electrodes are equally related to the effect of light

on alertness. It is worth attempting to additional

signal clearing in a situation where we can use an

Which EEG Electrodes Should Be Considered for Alertness Assessment?

electrode that collects brain waves from the area

strongly associated with the impact of the appropriate

stimulus. Research has shown that the selection of

electrodes can be made in a way that gives a higher

statistical significance of the impact. At the same

time, it is worth paying attention to the fact, that

presented in this article study is one of the first studies

of this type (if not the first one at all). Similar research

should be continued.

It is worth analyzing others, additional factors that

can affect the significance of the interaction - and thus

the selection of electrodes for signal analysis. In the

presented study, the extreme chronotypes were taken

into account. It turned out that this has a significant

impact - the interaction maps for different

chronotypes are different (Figures 3 and 4). Only

right-handed participants took part in the study. It

seems that from the point of view of the slightly

different functioning of the brain dependent on

lateralization, this is one of those factors that is worth

additional research. Attention should be paid to the

fact that while the influence of blue light on alertness

is documented in many articles, the influence of red

light is confirmed in a much smaller number of

publications. This is mainly due to the well-

documented impact of melatonin level on alertness

and the documented ability to influence light on

melatonin production.

The study described here shows that the effect on

brain waves of blue and red light is similar. What

once again confirms the possibility of interaction with

red light on alertness, although through a mechanism

other than melatonin production control.

ACKNOWLEDGEMENTS

This paper has been based on the results of a research

task carried out within the scope of the fourth stage of

the National Programme "Improvement of safety and

working conditions" partly supported in 2017–2019 -

-- within the scope of research and development ---

by the Ministry of Science and Higher Education /

National Centre for Research and Development. The

Central Institute for Labour Protection -- National

Research Institute (CIOP-PIB) is the Programme's

main co-ordinator.

REFERENCES

Alkozei, A., Smith, R., Pisner, D.A., Vanuk, J.R., Berryhill,

S.M., Fridman, A., Shane, B.R., Knight, S.A., Killgore,

W.D.S., 2016. Exposure to blue light increases

subsequent functional activation of the prefrontal

cortex during performance of a working memory task.

Sleep, vol 39 (9), 1671-1680.

http://dx.doi.org/10.5665/sleep.6090

Baek, H. Min, B.K., 2015. Blue light aids in coping with

the post-lunch dip: an EEG study. Ergonomics, 58(5),

803-810.

Bellia, L., Bisegna, F., Spada, G., 2011. Lighting in indoor

environments: Visual and non-visual effects of light

sources with different spectral power distributions.

Build Environ 46, 1984-1992. doi:

10.1016/j.buildenv.201.04.007

Cajochen, C., Munch, M., Kobialka, S., Krauchi, K.,

Steiner, R., Oelhafen, P., Eizr-Justice, A., 2005. High

sensitivity of human melatonin, alertness,

thermoregulation and heart rate to short wavelength

light. J Clin Endocrinol Metab, 90, 1311-1316.

Cajochen, C., 2007. Alerting effects of light. Sleep Med

Rev., 11, 453-464.

Cajochen, C., Chellappa, S., Schmidt, C., 2010. What keeps

us awake? The role of clocks and hourglasses, light, and

melatonin. Int Rev Neurobiol, 93, 57-90.

Chellappa S.L., Steiner R., Blattner P., Oelhafen P.,

Cajochen C. (2017) Sex differences in light sensitivity

impact on brightness perception, vigilant attention and

sleep in humans? Scientific Reports, 7: 14215, 1-9.

Donskaya, O.G., Verevkin, E.G., Putilov, A.A., 2012. The

first and second principal components of the EEG

spectrum as the correlates of sleepiness. Somnologie,

16, 69-79. Doi. 10.1007/s11818-012-0561-1.

Figueiro, M.G., 2013. Non-Visual Lighting Effects and

Their Impact on Health and Well-Being. In

Encyclopedia of Color Science and Technology.

Springer Science+Business Media New York doi:

10.1007/978-3-642-27851-8_118-4, 1-11.

Figueiro, M.G., Bierman, A., Plitnick, B., Rea, M.S., 2009.

Preliminary evidence that both blue and red light can

induce alertness at night. BMC Neurosci, 10. 10:105.

DOI: 10.1186/1471-2202-10-105

Figueiro, M.G., Rea, M.S., 2010. The effects of red and blue

light on circadian variations in cortisol, alpha amylase

and melatonin. Int J Endocrinol., Volume 2010, Article

ID 829351, 9 pages. doi:10.1155/2010/829351.

Figueiro, M.G., Sahin, L., Wood, B., Plitnick, B., 2016.

Light at night and measures of alertness and

performance: implications for shift workers. Biol Res

Nurs.. 18(1), 90-100. doi:10.1177/10998004155728 73.

Iskra-Golec, I., Golonka, I., Wyczesany, M., Smith, L.,

Siemiginowska, P., Watroba, J., 2017. Daytime effect

of monochromatic blue light on EEG activity depends

on duration and timing of exposure in young men.

Advances in Cognitive Psychology, 13(3), 241-247. doi:

10.5709/acp-0224-0.

Jankowski, K.S., 2105. Composite Scale of Morningness:

psychometric properties, validity with Munich

ChronoType Questionnare and age/sex differences in

Poland. Eur Psychiatry, 30, 166-171. doi:

10.1016/j.eurpsy.2014.01.004.

Klimesch, W., 2012. Alpha-band oscillations, attention, and

controlled access to stored information. Trends Cogn

CHIRA 2019 - 3rd International Conference on Computer-Human Interaction Research and Applications

Sci. 16(12), 606-617.

Lavoie, S., Paquet, J., Selamoui, B. Rufiange, M., Dumont,

M., 2003. Vigilance levels during and after bright light

exposure in the first half of the night. Chronobiol Int,

20(6), 1019-1038. doi: 10.1081/CBI-120025534.

Łaszewska, K., Goroncy, A., Weber, P., Pracki, T., Tafil-

Klawe, M., Pracka, D., Złomańczuk, P., 2017. Daytime

acute non-visual alerting response in brain activity

occurs as a result of short- and long wavelengths of

light. Journal of psychophysiology.

https://doi.org/10.1027/0269-8803/a000199.

Maierova, L., Borisuit, A., Scartezzini, J.L., Jaeggi, S.M.,

Schmidt, C., Münch, M., 2016. Diurnal variations of

hormonal secretion, alertness and cognition in extreme

chronotypes under different lighting conditions.

Nature. Scientific Reports, 1-10. doi:

10.1038/srep33591.

Malmivuo, J., Plonsey, R., 1995. Bioelectromagnetism.

Electroencephalography (chapter 13). In

Bioelectromagnetism – Principles and Applications of

Bioelectric and Biomagnetis Fields. Oxford University

Press. 363-374.

Okamoto, Y., Rea, M.S., Figueiro, M.G., 2014. Temporal

dynamics of EEG activity during short and long

wavelength light exposures in the early morning. BMC

Res Notes,7: 113, 1-6.

Phipps-Nelson, J., Redman, J.R., Schlangen, L.J.,

Rajaratnam, S.M., 2009. Blue light exposure reduces

objective measures of sleepiness during prolonged

night time performance testing. Chronobiol Int. 26(5):

891-912. doi: 10.1080/07420520903044364.

Plitnick, B., Figueiro, M.G., Wood, B., Rea, M.S., 2010.

The effects of red and blue light on alertness and mood

at night. Lighting Res. Technol, 42(4), 449-458. doi:

10.1177/1477153509360887

Sahin, L. and Figueiro, M.G., 2013. Alerting effects of

short-wavelengths (blue) and long – wavelengths (red)

lights in the afternoon. Physiol Behav. 116-117(5),

DOI:10.1016/j.physbeh.2013.03.014-7.

Sawicki, D., Wolska, A., Rosłon, P., Ordysiński, S., 2016.

New EEG Measure of the Alertness Analyzed by

Emotiv EPOC in a Real Working Environment. In

Proc. of the 4th International Congress on

Neurotechnology, Electronics and Informatics,

NEUROTECHNIX 2016, Porto, Portugal, 7-8 Nov.

2016, 35-42. https://doi.org/10.5220/0006041200

350042.

Scheuermaier, K., Munch, M., Ronda, J.M., Duffy, J.F.,

2018. Improved cognitive morning performance in

healthy older adults following blue – enriched light

exposure on the previous evening. Behav Brain Res.

348, 267-275, https://doi.org/10.1016/j.bbr.2018.04.021.

Smith, C.S., Reilly, C., Midkiff, K., 1989. Evaluation of

three circadian rhythm questionnaires with suggestion

for an improved measure of morningness. J Appl

Psychol., 75. 728-738.

Wolska, A., Sawicki, D., Nowak, K. Wiselka, M.,

Kołodziej, M., 2018. Method of acute alertness level

evaluation after exposure to blue and red light (based

on EEG): Technical aspects. In Proc. of the 6th

International Congress on Neurotechnology,

Electronics and Informatics, NEUROTECHNIX 2018,

Sevile, Spain,. 2018, 53-60. doi:

10.5220/0006922500530060.

Yokoi, M., Aoki, K., Shiomura, Y., Iwanga, K., Katsuura,

T., 2003. Effect of bright light on EEG activities and

subjective sleepiness to mental task during nocturnal

sleep deprivation. J. Physiol Anthropol Appl Human

Sci. 22, 257-263.

Which EEG Electrodes Should Be Considered for Alertness Assessment?