PreprintPDF Available

Using EEG to characterise drowsiness during short duration exposure to elevated indoor Carbon Dioxide concentrations

Authors:
Preprints and early-stage research may not have been peer reviewed yet.

Abstract and Figures

Drowsiness which can affect work performance, is often elicited through self- reporting. This paper demonstrates the potential to use EEG to objectively quantify changes to drowsiness due to poor indoor air quality. Continuous EEG data was recorded from 23 treatment group participants subject to artificially raised indoor CO 2 concentrations (average 2,700 ± 300 ppm) for approximately 10 minutes and 13 control group participants subject to the same protocol without additional CO 2 (average 830 ± 70 ppm). EEG data were analysed for markers of drowsiness according neurophysiological methods at three stages of the experiment, Baseline, High CO 2 and Post-Ventilation. Treatment group participants’ EEG data yielded a closer approximation to drowsiness than that of control group participants during the High CO 2 condition, despite no significant group differences in self-reported sleepiness. Future work is required to determine the persistence of these changes to EEG over longer exposures and to better isolate the specific effect of CO 2 on drowsiness compared to other environmental or physiological factors. Practical implications This study introduces EEG as a potential objective indicator of the effect of indoor environmental conditions upon drowsiness Participants exposed to 2,700 ppm for 10 minutes showed greater evidence of a progression towards drowsiness (as measured by EEG) than that of participants who received the same protocol without additional CO 2 (mean 830 ± 70 ppm), despite similar ratings of subjective sleepiness. Subjective and objectively measured indications of drowsiness were reduced following ventilation of the room. Future work could explore the potential of regular ventilation episodes in knowledge work spaces to retain alertness.
Content may be subject to copyright.
1
Long title
Short title
Using EEG to characterise drowsiness during short duration
exposure to elevated indoor Carbon Dioxide concentrations
The effect of CO2 upon drowsiness
Author names
Stephen Snow1, Amy Boyson1,2, Marco Felipe-King3, Obaid Malik1,
Louise Coutts1, Catherine J Noakes3, Hannah Gough4, Janet Barlow4, m.c.
schraefel1
Affiliations
1 Electronics and Computer Science, University of Southampton,
Southampton, SO17 1BJ, UK,
2 Psychology, University of Southampton, Southampton, SO17 1BJ, UK
3 School of Civil Engineering, University of Leeds, Woodhouse lane,
LS29JT, Leeds, UK
4 Department of Meteorology, University of Reading, Reading, UK, RG6
6AH
Corresponding
author, email
Stephen Snow, s.snow@soton.ac.uk
.CC-BY 4.0 International licenseIt is made available under a
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/483750doi: bioRxiv preprint first posted online Nov. 30, 2018;
2
1Abstract: Drowsiness which can affect work performance, is often elicited through self-
2 reporting. This paper demonstrates the potential to use EEG to objectively quantify changes to
3 drowsiness due to poor indoor air quality. Continuous EEG data was recorded from 23
4 treatment group participants subject to artificially raised indoor CO2 concentrations (average
5 2,700 ± 300 ppm) for approximately 10 minutes and 13 control group participants subject to
6 the same protocol without additional CO2 (average 830 ± 70 ppm). EEG data were analysed
7 for markers of drowsiness according neurophysiological methods at three stages of the
8 experiment, Baseline, High CO2 and Post-Ventilation. Treatment group participants’ EEG data
9 yielded a closer approximation to drowsiness than that of control group participants during the
10 High CO2 condition, despite no significant group differences in self-reported sleepiness. Future
11 work is required to determine the persistence of these changes to EEG over longer exposures
12 and to better isolate the specific effect of CO2 on drowsiness compared to other environmental
13 or physiological factors.
14
15 Keywords: EEG; drowsiness; ventilation; CO2; office; air quality
16
17
18 Practical implications:
19 This study introduces EEG as a potential objective indicator of the effect of indoor
20 environmental conditions upon drowsiness
21 Participants exposed to 2,700 ppm for 10 minutes showed greater evidence of a
22 progression towards drowsiness (as measured by EEG) than that of participants who
23 received the same protocol without additional CO2 (mean 830 ± 70 ppm), despite
24 similar ratings of subjective sleepiness.
25 Subjective and objectively measured indications of drowsiness were reduced following
26 ventilation of the room. Future work could explore the potential of regular ventilation
27 episodes in knowledge work spaces to retain alertness.
.CC-BY 4.0 International licenseIt is made available under a
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/483750doi: bioRxiv preprint first posted online Nov. 30, 2018;
3
28
29 Introduction
30 Being a product of human respiration, carbon dioxide (CO2) increases in indoor spaces when
31 ventilation of the space is insufficient to replace stale air [1,2]. CO2 is thus a useful indicator
32 of ventilation and, by extension air quality indoors, in occupied spaces [3,4]. A large body of
33 literature exists relating poor ventilation to mild health symptoms [2,5–7] and lowered
34 cognitive performance [4,8–10]. Office-realistic levels of CO2 are reported to be typically <
35 3,000 ppm, but whether CO2 itself negatively impacts cognitive performance, or whether other
36 pollutants such as volatile organic compounds (VOCs), including human bio-effluents, are
37 responsible, is still unclear [11,12]. Human performance effects have been recorded in studies
38 both where CO2 is accompanied by human bio-effluents (e.g. the CO2 concentration is a
39 product of poor ventilation in occupied spaces) [4,13,14] and where pure CO2 gas is added to
40 a room to achieve steady-state concentrations [12,13,15–18].
41 At a room concentration of 3,000 ppm, human bio-effluents are found to cause an increase in
42 respired -end-tidal- CO2 (ETCO2), increased blood pressure, and seemingly increased
43 stress/arousal, as well as reduced cognitive performance [19]. Zhang et al. proposes CO2 with
44 bio-effluents affects cognitive performance through either (1) stress/arousal or (2)
45 physiological factors such as an increase in ETCO2 and reduced nasal peak flow, triggering
46 symptoms such as subjective (self-reported) sleepiness, tiredness and headache [19]. One study
47 found that four hours of exposure to non-ventilated conditions, with average CO2
48 concentrations above 2,700 ppm, resulted in significantly increased blood-CO2, heart rate
49 variability, and increased peripheral blood flow, as well as increased prevalence of health
50 symptoms and self-reported sleepiness [14]. The study reports that the high CO2 concentration
51 itself (separate to bio-effluents) may be a parameter affecting physiology which can lower
52 functional ability and increase (self-reported) sleepiness [14]. Given findings that 1,400 ppm
.CC-BY 4.0 International licenseIt is made available under a
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/483750doi: bioRxiv preprint first posted online Nov. 30, 2018;
4
53 [15] and 2,500 ppm [16] of CO2 achieved by introducing pure gas into a room correlates to
54 lower decision making capability, cumulatively, there is some evidence that CO2 itself,
55 independent of other indoor pollutants, may play a role in detrimentally affecting aspects of
56 work performance [15–17].
57 Drowsiness and fatigue are recognised as important parameters affecting office work and
58 productivity [14,20]. In this study we focus on drowsiness, (i.e. lethargy or wish to sleep) [21–
59 23], rather than mental fatigue (i.e. exhaustion or lack of motivation for task(s) due to extended
60 work effort) [24]. Sub-optimal air quality (i.e. poor ventilation/high CO2) is correlated to
61 increased self-reported sleepiness and fatigue [14]. Yet factors such as sleepiness, drowsiness
62 and fatigue, when reported in studies assessing the effect of indoor conditions on humans, are
63 often elicited subjectively through questionnaires only [10,14,18,25]. One study uses voice
64 analysis to as a means of objectively measuring fatigue [20], but this method has not been
65 widely adopted. The lack of objective measurement of drowsiness or fatigue may be
66 problematic, given that self-reporting is identified as a less reliable measurement than objective
67 measurement [26,27]. On the other hand, fields such as Neurophysiology, have a long history
68 of objectively measuring sleep, and wakeful sleepiness/drowsiness using
69 electroencephalogram (EEG). EEG records electrical activity in the brain using electrodes
70 fitted to a cap, or placed on the scalp directly [28]. EEG data can be analysed to: (a) detect
71 specific events (event-related potential) or (b) time-averaged power in different frequency
72 bands [28]. A dominance of low frequency power is typically associated with lower
73 neurological arousal (delta, theta) [22].
74 The impact of office-realistic concentrations of CO2 upon objectively measured drowsiness is
75 a knowledge gap in the literature. Temperature effects on drowsiness using EEG find lower
76 temperatures are correlated to reduced drowsiness [29]. EEG research to date concentrates on
.CC-BY 4.0 International licenseIt is made available under a
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/483750doi: bioRxiv preprint first posted online Nov. 30, 2018;
5
77 neurological effects of much higher concentrations of CO2 than is likely to occur in indoor
78 spaces, e.g. 5% CO2/air mixture (50,000 ppm) [30–32], or 10% (100,000 ppm) [33] and the
79 resultant hypercapnia (elevated blood CO2) [30,31,33]. In these studies EEG results are
80 assessed according to arousal state (i.e. overall changes to low-frequency parameters), but not
81 drowsiness specifically. Xu et al.[31] found inhalation of a 5% CO2/air mixture (50,000 ppm)
82 caused transition to a lower (brain) arousal state, characterised by a relative increase in delta
83 power and corresponding decrease in alpha power. Bloch-Salisbury [33] subjected participants
84 to 10% CO2 (10,000 ppm) through direct inhalation, finding a significant decrease in both
85 overall power and a movement of the centroid frequency (i.e. the centrepoint of the mass of
86 frequencies observed) toward lower frequencies.
87 In summary, (1) findings are mixed as to whether CO2 is a pollutant affecting cognitive
88 performance in its own right, with some studies finding evidence that CO2 affects cognitive
89 performance [15–17], while others find no evidence of this relationship [10,13,18]. (2) Poor
90 indoor air quality is correlated to increased subjective drowsiness [14], yet drowsiness is
91 typically elicited through self-reporting [10,14,18,25], which is less reliable than objective
92 measurement [26,27]. (3) the field of neurophysiology offer methods of objectively measuring
93 drowsiness (a precursor to sleepiness) using EEG [23,34], yet these methods have not yet been
94 applied to office-realistic CO2 concentrations. (4) Literature on the effect of CO2 on resting
95 EEG is presently limited to the human effects of much higher levels of CO2 [31,33] than could
96 realistically be achieved indoors through human respiration. Comparable studies of office-
97 realistic concentrations of CO2 are not yet available, providing impetus for this present paper.
98 This paper details the novel application of using electroencephalogram (EEG) as a means of
99 objectively measuring the effect of CO2 on drowsiness at office-realistic concentrations.
100 Resting EEG and other physiological and subjective parameters were recorded from
.CC-BY 4.0 International licenseIt is made available under a
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/483750doi: bioRxiv preprint first posted online Nov. 30, 2018;
6
101 participants exposed to 2,700 ± 300 ppm of CO2 in an office for 10 minutes, as a means of
102 determining the physiological changes of a short-duration exposure to elevated CO2
103 concentration and testing for EEG data indicative of a progression towards drowsiness. A key
104 aim of the paper is to explore the effect of CO2 on drowsiness, given that drowsiness is a
105 determinant of human work performance [20,35] and compare results to both cognitive science
106 literature on the cognitive performance effects of office-realistic concentrations of CO2
107 [4,8,10,12,15,16] and neurophysiology literature on the neurological effects of much higher
108 concentrations of CO2 [30–33].
109 Materials and methods
110
111 Rationale for study design
112 Our chosen target for CO2 concentration (2,700ppm) reflects a high, but realistic level achieved
113 in occupied spaces when windows and doors are closed [2,14]. In a meta-review of classroom
114 ventilation, Fisk [2] found six studies of 20 or more classrooms recorded average or median
115 CO2 concentrations between 2,000 and 3,000 ppm. The target concentration is chosen to be
116 comparable with other studies assessing the human performance effects of indoor CO2
117 concentration, e.g. 2,260 ppm [4] 2,500 ppm [16] or 3,000 ppm [10,17,19]. The duration of
118 exposure to elevated CO2 concentration in our study is shorter compared to others [4,14,16],
119 and relates to our aim to record and analyse EEG continuously throughout the experiment to
120 provide a novel focus on immediate-term physiological effects of CO2. Continuous EEG
121 recording is less practicable over extended study durations due to the need for participants to
122 remain still during EEG recordings to ensure clean data [28]. The need to remain still over
123 extended durations, when combined with a lack of stimulation may produce a tendency to
124 fidget, which may in turn affect measured EEG parameters, or potentially cause boredom/
125 drowsiness itself, which could confound determination of drowsiness as caused through
126 changing indoor environment parameters.
.CC-BY 4.0 International licenseIt is made available under a
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/483750doi: bioRxiv preprint first posted online Nov. 30, 2018;
7
127
128 Participants
129 A total of 47 subjects were recruited and participated in the study between October 2016 and
130 February 2017. Usable EEG data was available from 36 of the 47 participants, reflective of the
131 sensitivity of EEG to movement artefacts and the researchers’ wish for data reliability. The
132 study protocol and conditions of participation were approved by the University of Southampton
133 Ethical Research Governance Office (ERGO# 30443). Sampling was achieved by advertising
134 the study on billboards throughout the University, a local supermarket and a departmental
135 mailing list. Convenience sampling was used for contacts of the research team who were
136 unaware of the study protocol. The final sample was comprised mostly of students and staff
137 from the University. Written consent was gathered from each participant prior to their
138 participation in the study. Exclusion criteria for the study were adapted from those used by
139 Garner et al. [36], a study where participants were subjected to 7.5% CO2 (75,000 ppm) level
140 of CO2. Exclusion criteria included current or historic drug/alcohol abuse or panic attacks,
141 current treatment for migraine headaches, pregnant, current neurological conditions (e.g.
142 epilepsy), and recent severe illness. Participants were compensated £10 in vouchers for an
143 online retailer for their participation.
144 Participants were split into two groups. Of the participants with usable EEG data, this involved:
145 23 participants in the “treatment group” (TG) who received artificially raised CO2
146 concentrations and 13 participants in the “control group” (CG) for whom CO2 concentrations
147 were not artificially raised (Table 1). The variance in the size of the groups is due to which of
148 the participants had sufficiently clean EEG for inclusion and the difficulty in recruiting a larger
149 sample.
150
151 Table 1- Participant attributes
Treatment group
Control group
.CC-BY 4.0 International licenseIt is made available under a
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/483750doi: bioRxiv preprint first posted online Nov. 30, 2018;
8
Male
12
9
Female
11
4
Median age (years)
23.0
24.5
152
153 Statistical power analysis was calculated a-priori using G*Power software [37]. Effect size was
154 estimated at 0.4 based on similar experiments [12], number of groups = 2 (treatment, control),
155 number of measurements = 3 (Baseline, High-CO2, Post ventilation- defined below),
156 significance level 0.05. This gave a between factors recommendation for 58 participants, and
157 recommendations for both within-factors and within-between factors of 18 participants. In this
158 paper we concentrate on within-factors analysis.
159
160 Study room
161 A motivation for the study was to replicate office-realistic scenarios. All experiments took
162 place in a small, carpeted, naturally ventilated office of dimensions 4,000 mm by 3,400 mm
163 (floor area) by 3,050 mm (high) (Figure 1). The office was on the fourth floor, on the northern
164 end of a large building in the south of England. The office had two high windows on the north
165 and west corner of the room. Only the western window could be opened, and is visible behind
166 the participant in Figure 1. The CO2 cylinder was positioned directly in front of the openable
167 window. The door of the room led to a larger reception office which was occupied by one staff
168 member during some but not all of the experiments. The numbered arrows in Figure 1 point to
169 the location of the CO2 loggers.
170
171 Figure 1- Study room showing participant with EEG cap, location of loggers, window and CO2 cylinder
172
173 The infiltration rate of the study room with the windows closed was calculated according to
174 Laussmann et al. [38] using a tracer-gas decay method overnight, with the researcher ensuring
.CC-BY 4.0 International licenseIt is made available under a
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/483750doi: bioRxiv preprint first posted online Nov. 30, 2018;
9
175 the mixing of CO2 in the room by observing the range of the readings from the three CO2
176 monitors and ensuring all were within instrument error before leaving the room overnight. This
177 method gave an infiltration rate of 0.078 ± 0.002 (R2 = 0.91) air changes per hour, consistent
178 with the rubber-sealed windows and minimal air gaps around the door. The value is
179 approximate, given air exchange rates can differ over time due to differences in temperature,
180 wind direction and wind speed [39].
181 Carbon dioxide was introduced using a cylinder of ultrapure CO2 (greater than 99.99% purity)
182 located in the corner of the room with the outlet attached to pedestal fan to achieve mixing.
183 The fan was pointed away from the participant and in operation only for the duration of
184 Condition 3 (see Table 2), when CO2 was being released, in order to minimise any influence
185 of air movement on perception or produce possible thermal comfort effects during subsequent
186 conditions. The target CO2 concentration once mixed was 2,700 ppm (mean: 2,700 ± 300 ppm
187 for the duration of Condition 5). Participants were instructed to sit at the table in the middle of
188 the room while the researcher operated the computer and the gas cylinder behind the
189 participant. In this way participants were aware the air quality was going to be changed
190 somehow during the experiment, but were not aware how.
191
192 Experimental Procedure
193 The experimental protocol took place in the one study room (Figure 1). The study protocol is
194 summarised below for TG participants (Table 2). CG participants experienced the same
195 protocol to that of TG participants, except that the CO2 concentration of the room was not
196 modified using the cylinder. Instead a pre-recorded and equalized sound was used in place of
197 the CO2 gas being released throughout Condition 3 to mimic the sound of the gas release. When
198 questioned, no CG participant identified the sound as audio playback and thus every participant
199 assumed their air quality was being modified.
.CC-BY 4.0 International licenseIt is made available under a
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/483750doi: bioRxiv preprint first posted online Nov. 30, 2018;
10
200
201 Table 2- Experimental protocol
Condition
number
Duration
Pre-start
Pre-start
1
2 minutes
2
5 minutes
3
2-3 minutes
(dependent on
CO2 mixing)
4
2 minutes
5
8 minutes
6
5 minutes
7
5 minutes
202
203 For comparative data analysis, three two-minute segments were selected for comparison, (1)
204 Baseline the first two minutes of Condition 2; before the environmental conditions were
205 changed, (2) High-CO2 – The last two minutes of Condition 5; beginning when TG participants
206 had been exposed to the higher CO2 concentration for 8 minutes; and (3) Post-Ventilation - last
207 two minutes of Condition 7, beginning after 8 minutes of room ventilation. The location of
208 these analysis segments within the context of the study protocol are shown in Figure 2.
209
210 Figure 2- Study protocol with indicative CO2 level showing location of Baseline, High-CO2 and Post-Ventilation segments
211
212 Measurement
213 Three factory calibrated Rotronic CL11 (BSRIA, Bracknell, UK) environmental loggers
214 measured temperature, humidity and CO2 concentration throughout each experiment. The
.CC-BY 4.0 International licenseIt is made available under a
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/483750doi: bioRxiv preprint first posted online Nov. 30, 2018;
11
215 loggers were positioned approximately equidistant around the room and are labelled 1, 2 and 3
216 in Figure 1. The loggers were positioned so as to avoid influence from direct respiration. The
217 heights of the loggers from the floor were 720 mm (logger 1), 1,545 mm (logger 2) and 1,995
218 mm (logger 3). The distance from logger 2 to logger 3 was 2,100 mm and logger 1 was
219 approximately 1,300 mm perpendicular to the participant’s heads (Figure 1). Instrument
220 accuracies for the CL11 are ± 0.3 ° C (temperature), < 2.5% RH (humidity) and ± 30 ppm ±
221 5% of the measured value. The logging frequency of the CL11 monitors was set to 10 seconds
222 throughout the experiments. The CL11’s display updates approximately once per second,
223 enabling the researcher to monitor and control the release of CO2 in the room to a reasonable
224 granularity. The length of Condition 3 (adding CO2) was varied according to the time taken to
225 achieve mixing (Table 2), to enable confidence in the mixing of the room by the start of
226 Condition 4.
227 EEG data was gathered from each participant using a Neuroelectrics ENOBIO 20 dry electrode
228 wearable wireless EEG cap (19 channel, 10-20 placement, 500 Hz sampling rate). Two
229 reference electrodes (DLR, CRL) were positioned on the participants’ mastoid muscle. EEG
230 was gathered continuously throughout each of the experimental conditions (Table 2, Figure 2).
231 In order to minimise movement artefacts in the EEG, participants were asked to sit quietly and
232 remain still throughout the experiments except during the short break for the questionnaire
233 following Condition 5 (refer Table 2, Figure 2).
234 Subjective responses were gathered in relation to experience of sick building symptoms (e.g.
235 irritated eyes, sore throat, congested nose) [40], positive/negative affect (PANAS) [41],
236 Stanford Sleepiness Scale [42] and thermal comfort (ASHRAE 7 point scale) [43] were
237 gathered from participants at Baseline, High- CO2 and Post-Ventilation segments.
.CC-BY 4.0 International licenseIt is made available under a
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/483750doi: bioRxiv preprint first posted online Nov. 30, 2018;
12
238 As a proof-of-concept, this paper focuses specifically on EEG results and the Stanford
239 Sleepiness Scale.
240
241 Analysis
242
243 Environmental measurements
244 Data from the Rotronic CL11 environmental monitors was downloaded and condition timings
245 entered retrospectively for analysis. Due to the difference in logging frequency of the CL11s
246 (10 sec) compared to the EEG measurements (500 Hz), the error on the readings versus that of
247 the condition timings is expected to be approximately ± 20 seconds. This error was considered
248 acceptable given the gradual changes in temperature/humidity and the mixing behaviour of the
249 CO2 in the room.
250
251 EEG pre-processing
252 EEG data were filtered using a Butterworth filter; low pass at 45 Hz and high pass at 0.15 Hz.
253 Artefact rejection was implemented in two stages. The first used the artefact rejection algorithm
254 WPT-EMD [44,45], which uses a sample of minimum variance EEG taken from Condition 2.
255 The second stage of artefact rejection involved an amplitude threshold cut-off of ±100 µV, and
256 replacing outlying data with a 10-second moving median around the extreme value. Electrodes
257 showing consistent noise or flat-lined output were deleted from the dataset. As mentioned, of
258 the total 47 participants, 36 participants had sufficiently clean data throughout the experiment
259 and sufficient representation of clean electrodes in each brain region (frontal, central, temporal,
260 parietal, occipital) to warrant further analysis.
261 Bandpower was extracted from the pre-processed continuous EEG for delta (0.15-3 Hz), theta
262 (4-7 Hz), alpha (8-13 Hz), beta (14-35 Hz), and gamma (> 35 Hz) frequency bands, over one
263 second windows. Average bandpower was computed for frontal (F3, Fz, F4, FP1, FP2), central
264 (C3, Cz, C4), parietal (P7, P3, Pz, P4, P8), temporal (T7, T8), and occipital (O1, O2) electrodes
.CC-BY 4.0 International licenseIt is made available under a
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/483750doi: bioRxiv preprint first posted online Nov. 30, 2018;
13
265 for each analysis segment (Baseline, High-CO2, Post-Ventilation). Gamma was excluded from
266 further analysis owing to the focus of the study protocol on low frequency behaviour and
267 because gamma represented < 1% of total power at each analysis segment. Post-hoc analysis
268 found the lowest delta component (0.15-1.5 Hz) to be contaminated with eye movement
269 artefacts and was subsequently rejected from analysis. Rather than excluding delta from
270 analysis completely, and given eye movement artefacts typically occur at approximately 1 Hz
271 [46], we instead report on high-delta (2-3 Hz) and exclude only low-delta (i.e. all frequencies
272 < 2 Hz).
273 Mixed model ANOVAs were conducted with factors including electrode region, analysis
274 segment, group, and frequency to investigate electrophysiological markers of drowsiness
275 consistent with the literature (detailed below).
276
277 EEG- drowsiness characterisation
278 Our characterisation of drowsiness applied to the EEG results is grounded in relevant literature:
279 A meta-review of the psychophysiology of automobile driver fatigue finds changes in delta and
280 theta strongly linked to the transition towards fatigue [21]. Tired wakefulness among sleep
281 deprived participants produces an EEG with enhanced power in the low frequency range 1-8
282 Hz (delta and theta) [22,47]. Providing a greater topographical specificity than previous studies,
283 Gorgoni et al. finds sleep deprived participants exhibit an EEG involving global increases in
284 delta and theta (i.e. registered in multiple areas of the brain) [23]. Thus in this study, drowsiness
285 is characterised by post-hoc analysis of the cleaned EEG data according to an increase in delta
286 and theta, particularly if these increases are found at multiple brain electrode regions.
287
.CC-BY 4.0 International licenseIt is made available under a
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/483750doi: bioRxiv preprint first posted online Nov. 30, 2018;
14
288 Results
289 All statistical analyses conducted and reported in this section relate to data from the three
290 analysis segments of Baseline, High-CO2 and Post-Ventilation. Additionally, all analyses and
291 data reported below relate to the 36 participants with usable EEG data.
292
293 Indoor conditions by analysis segment
294 Table 3 below summarises the measured indoor environment parameters at each of the two-
295 minute analysis segments: Baseline, High- CO2 and Post-Ventilation (Figure 2), for TG and
296 CG participants:
297
298 Table 3: Indoor parameters by analysis segment, TG and CG participants
Treatment group participants
CO2 (ppm)
Temp (°C)
RH1 (%)
Baseline
670 ± 80
21.8 ± 2.3
44.1 ± 8.2
High-CO2
2750 ± 160
22.2 ± 2.5
44.6 ± 7.9
Post-Ventilation
850 ± 210
21.5 ± 2.4
43.9 ± 7.5
Control group participants
CO2 (ppm)
Temp (°C)
RH1 (%)
Baseline
660 ± 40
23.6 ± 1.8
37.7 ± 7.6
High-CO2
860 ± 50
24.3 ± 2.0
37.8 ± 7.3
Post-Ventilation
680 ± 80
23.4 ± 1.8
37.8 ± 7.8
299
300 The mean CO2 values for the two minute segments of Baseline and High-CO2 correspond
301 closely to the mean CO2 values for TG and CG participants for entire five minute duration of
302 Condition 2 (650 ± 80 ppm TG, 640 ± 50 ppm CG) and eight minute duration of Condition 5
303 (2,700 ± 300 ppm in TG, 830 ± 70 in CG). With reference to Table 3, TG participants were
304 exposed on average to an additional 1,898 ppm of pure CO2 to that generated by human
305 respiration alone.
.CC-BY 4.0 International licenseIt is made available under a
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/483750doi: bioRxiv preprint first posted online Nov. 30, 2018;
15
306 To control for possible temperature effects, all participants were able to adjust clothing as they
307 wished prior to the experiment to ensure comfort. A 3 (analysis segment) by 2 (group) mixed
308 model ANOVA was run to assess temperature fluctuations. Results show that CG participants
309 were tested at a significantly higher temperature than TG participants (see Table 3 and Section
310 0; F (1, 34) = 6.30, p = .02,
η
p2 = .16). This was due to the majority of CG participants being
311 tested following the activation of the building’s heating systems. Results also showed that
312 temperature varied significantly between each of the analysis segments irrespective of group
313 (F (1.46, 49.55) = 50.75, p < .001,
η
p2 = .60; Sidak post-hoc p’s < .02). Temperature was higher
314 on average for both groups at High CO2 relative to the other conditions, due to the doors and
315 windows remaining closed; additionally, Post-Ventilation was colder than both High CO2 and
316 Baseline for both groups due to the windows being open throughout the condition and the
317 cooler outside air due to the season. However, the difference in temperature between analysis
318 segments (i.e. Baseline vs High CO2 vs Post-Vent) did not greatly exceed instrument accuracy
319 (0.3 °C).
320 The period of ventilation (including the Post-Ventilation analysis segment) was uncontrolled.
321 During this period, CO2 concentration (Table 3), as well as air change rate, indoor air velocity
322 and external noise was variable between participants, depending on external factors such as
323 wind direction, wind speed and traffic. We did not attempt to isolate, measure or control for
324 these variables, and include the Post-Ventilation segment in our analysis simply as a reference
325 period of increased fresh air and sensory disturbance.
326
327 EEG results
328 To test for the effect of elevated CO2 concentration upon participants’ EEG, a 4 (frequency) by
329 5 (electrode region) by 3 (analysis segment) by 2 (group) mixed model ANOVA was run.
.CC-BY 4.0 International licenseIt is made available under a
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/483750doi: bioRxiv preprint first posted online Nov. 30, 2018;
16
330 Results found a main effect of frequency (F (1.08, 36.58) = 89.62, p < .001,
η
p2 = .73), electrode
331 region (F (1.50, 51.13) = 50.52, p < .001,
η
p2 = .60), and analysis segment (F (2, 68) = 7.98, p
332 = .001,
η
p2 = .19). In addition significant interactions were also found for frequency by region
333 (F (1.72, 58.56) = 34.57, p < .001,
η
p2 = .50), frequency by analysis segment (F (2.09, 70.95)
334 = 9.16, p < .001,
η
p2 = .21), region by analysis segment (F (2.98, 101.29) = 7.61, p < .001,
η
p2
335 = .18), and frequency by region by analysis segment (F (3.73, 126.84) = 4.91, p = .001,
η
p2 =
336 .13). There was no main effect of group, and no significant group interactions.
337 Post-hoc analysis of the main effects (Sidak) showed that each frequency significantly differed
338 from the others (p’s < .004) such that high-delta had the highest power, followed by theta, then
339 alpha, then beta. Frontal electrodes had greater power than all other regions (p’s < .001).
340 Central and temporal electrodes did not differ from each other and neither did parietal and
341 occipital electrodes. Frequency power during Baseline was significantly lower than during the
342 High-CO2 (p = .001) analysis segment, but did not differ from Post-Ventilation. There was a
343 trend toward the Post-Ventilation analysis segment having a lower overall power than the High-
344 CO2 segment (p = .09).
345 To investigate the significant interactions, paired-sample t-tests were computed between the
346 Baseline and High-CO2 analysis segments and the High-CO2 and Post-Ventilation analysis
347 segments for each brain region and frequency, overall and for the TG and CG participants
348 separately (Table 4).
349
350 Table 4- Overall power, within measures, comparison of changes in power by analysis segment for each group. p-values
351 derived from paired sample post-hoc t-tests
High-CO2 vs Baseline
Post-Vent vs High-CO2
Overall power, within-
measures
Overall
Treatment
group
Control
group
Overall
Treatment
group
Control
group
h-delta
p < .001
p = .01
p = .003
p = .004
p = .07 a
p = .02
Frontal
theta
p < .001
p = .004
p < .001
p = .07 a
p = .77
p = .53
.CC-BY 4.0 International licenseIt is made available under a
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/483750doi: bioRxiv preprint first posted online Nov. 30, 2018;
17
alpha
p = .07 a
p = .31
p = .11
p = .73
p = .83
p = .53
beta
p = .003
p = .09 a
p = .007
p = .47
p = .79
p = .48
h-delta
p = .002
p = .02
p = .05 a
p = .04
p = .38
p = .04
theta
p = .14
p = .02
p = .89
p = .36
p = .52
p = .53
Central
alpha
p = .40
p = .31
p = .98
p = .32
p = .38
p = .65
beta
p = .36
p = .43
p = .64
p = .35
p = .57
p = .45
h-delta
p = .02
p = .04
p = .27
p = .35
p = .55
p = .48
theta
p = .01
p = .006
p = .55
p = .16
p = .37
p = .26
Parietal
alpha
p = .03
p = .001
p = .92
p = .43
p = .32
p = .88
beta
p = .03
p = .02
p = .60
p = .46
p = .29
p = .93
h-delta
p = .13
p = .33
p = .22
p = .34
p = .50
p = .52
theta
p = .38
p = .62
p = .38
p = .67
p = .31
p = .70
Temporal
alpha
p = .77
p = .87
p = .80
p = .81
p = .58
p = .36
beta
p = .67
p = .86
p = .64
p = .68
p = .79
p = .75
h-delta
p = .009
p = .03
p = .14
p = .07 a
p = .15
p = .31
theta
p = .008
p = .03
p = .16
p = .16
p = .61
p = .08 a
Occipital
alpha
p = .20
p = .18
p = .53
p = .03
p = .26
p = .02
beta
p = .04
p = .14
p = .16
p = .21
p = .73
p = .09 a
h-delta
p < .001
p = .007
p = .009
p = .01
p = .11
p = .04
theta
p < .001
p = .003
p = .006
p = .08 a
p = .53
p = .03
Overall
alpha
p = .008
p = .11
p = .03
p = .20
p = .17
p = .65
beta
p = .01
p = .12
p = .03
p = .35
p = .55
p = .49
352 a Trend (p < .10). Italics denotes significant p-values
353
354
355 Overall results, irrespective of group, show no changes in the temporal electrode region for any
356 frequency. The strongest effects from Baseline to High-CO2 are an increase of frontal high-
357 delta, theta and beta, central high-delta, and occipital high-delta and theta, as well as global
358 increases in high-delta, theta, and alpha. Despite a lack of significant group effects in the overall
359 model, the data presented in Table 4 show a clear difference in the pattern of frequency power
360 changes across the brain in the two groups. According to the definition of drowsiness employed
361 (Section 0), the results show the EEG of the TG shows a closer approximation to drowsiness
362 compared to that of the CG, considering: (a) the increase in delta and theta is more global than
363 the CG and (b) CG also has a significant overall increase in alpha and beta, while TG increase
364 is theta and high-delta only.
365
.CC-BY 4.0 International licenseIt is made available under a
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/483750doi: bioRxiv preprint first posted online Nov. 30, 2018;
18
366 Relationship between EEG and temperature
367 In order to assess whether any relationship existed between the temperature in the room and
368 the EEG, Pearson correlations were run for each analysis segment. The results show no
369 significant correlation between the average temperature during the segment and the global EEG
370 power of each frequency recorded during that time period. Correlations were also run for each
371 electrode region. This analysis found a significant negative relationship for alpha power in the
372 temporal region and temperature during Baseline only (r = -.34, p = .04).
373
374 Self-reported sleepiness (Effect of analysis segment, treatment group, within
375 measures)
376 Analysis of questionnaire data on subjective sleepiness found a significant main effect of
377 analysis segment on self-reported sleepiness, χ2 (2) = 22.84, p < .001 (Friedman’s ANOVA).
378 Wilcoxon matched pairs post-hoc comparisons show that participants at High-CO2 had
379 significantly higher ratings of sleepiness than both Baseline (p < .001) and Post-Ventilation (p
380 = .01). The Post-Ventilation segment also showed significantly higher ratings of sleepiness
381 than Baseline (p = .01) (Table 5). These p-values remained significant when analysed using
382 parametric statistics (3-way ANOVA).
383
384 Table 5- Self reported sleepiness, average rating with SD, within measures, TG and CG participants
Self-reported Sleepiness, Average Rating ± SD, within-measures
Treatment group
Control group
Baseline
2.2 ± .7
2.2 ± .8
High-CO2
3.2 ± 1.1
3.7 ± 1.0
Post-Ventilation
2.7 ± 1.2
2.6 ± 1.0
385 Stanford Sleepiness index: Likert scale from 1 (wide awake) to 7 (sleep onset soon).
386
387 The average sleepiness ratings are similar for both TG and CG participants; p > .05 for both
388 parametric and non-parametric comparisons (Table 5), indicating that subjective sleepiness
.CC-BY 4.0 International licenseIt is made available under a
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/483750doi: bioRxiv preprint first posted online Nov. 30, 2018;
19
389 was not affected by the changes in CO2 concentration. None of the group comparisons for
390 sleepiness approach significance.
391 Discussion
392
393 The effect of office-realistic changes to CO2 on resting EEG represent a knowledge gap in the
394 literature to date. This study tests the effect of a 2,700 ppm concentration of CO2 in an office
395 on resting EEG, analysing EEG results for indicators of a progression towards drowsiness. Data
396 was analysed at three segments of each experiment; Baseline, High-CO2 and Post-Ventilation.
397 This study supports the role of EEG as a means of objectively measuring drowsiness in humans
398 when affected by changes to the indoor climate.
399
400 Evidence for the effect of CO2 on drowsiness- Relationship between TG and CG
401 participants’ EEG
402
403 Results from this study provide an indication that the indoor CO2 concentration of 2,700 ppm
404 had an effect on the EEG indicative of a progression towards drowsiness, when drowsiness is
405 characterised by a global increase in delta and theta [22,23]. Despite the lack of a significant
406 effect of group in the overall model, and both groups showing some evidence of a progression
407 towards drowsiness, the evidence of drowsiness is stronger for the TG (Table 4). A distinct
408 trend observed among TG participants is the global nature of the high-delta and theta increases
409 from Baseline to High-CO2 among TG participants relative to the only frontal increase in these
410 parameters among CG participants. The findings of this paper reinforce calls for sufficient
411 ventilation in knowledge work spaces [2] and greater occupant awareness of indoor CO2
412 concentration in these spaces [48].
413 The Post-Ventilation findings show further differences between the TG and CG, where the CG
414 participants appeared better able to overcome the increased (EEG-assessed) drowsiness
415 experienced in the High-CO2 analysis segment. This may imply that the increased CO2
.CC-BY 4.0 International licenseIt is made available under a
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/483750doi: bioRxiv preprint first posted online Nov. 30, 2018;
20
416 experienced by TG participants affected the return of the EEG signals to Baseline levels.
417 However given the difference in sample size between the groups, caution must be taken when
418 looking at any potential group differences until further research is conducted with larger, more
419 equal group sizes.
420
421 Relationship between self-reported and EEG-measured drowsiness
422 The EEG of the TG more closely approximates drowsiness at High CO2 compared to the CG.
423 Yet the difference between average self-reported sleepiness ratings at High CO2 between CG
424 and TG is minimal (half the standard deviation), and is not significantly different between
425 groups (p > 0.5), (Table 5). Longer exposures to comparable concentrations of CO2 with bio-
426 effluents are found to affect (subjectively assessed) drowsiness: 255 minutes exposure to 3,000
427 ppm with bio effluents increased subjective sleepiness and difficulty in thinking clearly [10];
428 235 minutes exposure to 2,260 ppm affected perceived fatigue and perceived lack of energy
429 [4] and four hours’ exposure to CO2 above 2,700 ppm resulted in increased subjective
430 sleepiness [14]. The duration of this present study is much shorter than other studies and
431 subjective sleepiness between groups was unaffected. Given the short duration of the study and
432 the similarity of subjective sleepiness between groups, a possible explanation here is that both
433 groups self-report higher feelings of sleepiness simply as a function of time (being sat still in
434 the same room with no stimulation).
435 Further work is required to determine whether the objectively measured drowsiness indicated
436 in the EEG results persist over longer timescales, whether self-reported drowsiness is better
437 correlated to EEG over time, and whether EEG may be used as something of an early warning
438 system for drowsiness. Small changes in CO2 can quickly affect blood pH [31], and owing to
439 the short duration of the experiment, it is possible that EEG results may provide a more timely
440 indication of physiological changes than subjective sleepiness, though this suggestion needs to
.CC-BY 4.0 International licenseIt is made available under a
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/483750doi: bioRxiv preprint first posted online Nov. 30, 2018;
21
441 be corroborated. Additionally, because both subjectively and objectively measured indications
442 of drowsiness were reduced following ventilation of the room future work could additionally
443 explore the potential of regular ventilation episodes in knowledge work spaces to retain
444 alertness.
445
446 Relationship between EEG and temperature
447 Results also show a significant effect of temperature with CG participants, completing the
448 experiment at a slightly higher temperature than TG participants. Temperature in both groups
449 increased from Baseline to High-CO2 before dropping to below baseline levels as a result of
450 the ventilation of the room. Related literature finds lower temperatures (without increased CO2)
451 are correlated to decreased drowsiness as measured by EEG [29], and increasing indoor
452 temperatures (i.e. warm discomfort) is correlated to difficulty concentrating [49]. These
453 findings might explain the higher subjective sleepiness experienced by the CG at High CO2;
454 however, as mentioned, the subjective sleepiness ratings were small and not statistically
455 significant and all participants were invited to modify their clothing if required in order to
456 remain thermally comfortable throughout the experiment. Conversely, the TG had a higher
457 objective indication of drowsiness but were subject to cooler temperatures than the CG,
458 potentially suggestive that (1) the effects on the EEG of the TG in this study may be attributable
459 to CO2 rather than temperature and (2) that subjective and objective determinations of
460 drowsiness may not be correlated over short timescales. Future research could better control
461 the temperature of the environment to remove this variable as a potential confound.
462 Additionally, the correlation between objectively and subjectively measured drowsiness due to
463 changed CO2 conditions needs to be further explored, e.g. the potential for EEG to act as an
464 early warning system for drowsiness.
465
.CC-BY 4.0 International licenseIt is made available under a
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/483750doi: bioRxiv preprint first posted online Nov. 30, 2018;
22
466 Limitations and confounding factors
467 The results of this study should be viewed in light of its limitations: (1) The duration of
468 exposure in this study is much shorter than comparable studies of office-realistic CO2
469 concentrations on humans [8,10,14,16,50], and future work is required to determine whether
470 the changes in EEG with respect to drowsiness are momentary or sustained. (2) Accordingly,
471 changes in the EEG of the TG should be considered as indicative of a neurological progression
472 towards drowsiness, rather than definitive drowsiness. (2) While the CO2 outlet was attached
473 to a fan, mixing may not have been as effective as is possible in a climate chamber. (3) All
474 participants assumed that gas was released into the room during the experiment, as the CG
475 participants were exposed to a pre-recorded and equalized sound to mimic the CO2 gas being
476 released throughout Condition 3. Thus the participants were blind to the conditions, but were
477 not blinded to the fact that the air in the room was (supposedly) being modified. Thus it cannot
478 be ruled out that some CG may have experienced a placebo reaction. (4) The treatment and
479 control groups differ in sample size and the study is underpowered with respect to between-
480 groups analysis (a-priory power analysis N = 58, i.e. 29 per group), potentially explaining the
481 lack of group effects found in the overall ANOVA. However, even after discarding participants
482 with poor EEG data, the study is still well powered to make conclusions based on the within
483 subjects analysis (a-priori power analysis n = 18) of the whole sample, and for the TG. As such,
484 we are confident in our conclusion that the pattern of results found for this group more closely
485 approximates drowsiness. The study is only slightly under powered with regards to within
486 subjects analysis for the CG group only.
487
488 Future work
489
490 To corroborate our findings, future work using EEG as an objective indicator of the effects of
491 changes to indoor air quality would be helpful. To better isolate CO2 as a variable in future
492 studies, we suggest a within subjects study design for future work in order to ensure equal
.CC-BY 4.0 International licenseIt is made available under a
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/483750doi: bioRxiv preprint first posted online Nov. 30, 2018;
23
493 representation in the high and “sham” CO2 groups. Such a design would control for any
494 individual differences between the groups. Fully blinding participants to experimental
495 conditions might also be beneficial. In addition, there are personal factors not controlled for in
496 this study which could feasibly influence drowsiness, such as number of hours sleep, amount
497 of time since their last meal, their previous activity before experiment. Future studies should
498 account for such factors. Given our finding that a 10 minute ventilation period appeared to
499 reverse the trend towards drowsiness (Post-Vent versus High CO2), we suggest further work
500 investigates the acceptability of periodic drafts in naturally ventilated workplaces as a means
501 of maintaining vigilance and concentration.
502 Conclusion
503 Drowsiness represents an important factor affecting office work and productivity [14,20], yet
504 many studies assessing the effects of poor indoor environment quality on humans gather only
505 subjective data for factors potentially affecting work performance such as drowsiness or mood.
506 In this study we have demonstrated the potential for EEG to be used as an objective
507 measurement of drowsiness to determine the effect of elevated levels of indoor CO2. Results
508 indicate that even short exposure to elevated levels of CO2 indoors (TG) can produce EEG
509 indicative of a progression towards drowsiness. Further work is necessary to corroborate these
510 findings.
511 Priorities for further work have been outlined including: longer-duration studies using EEG,
512 full blinding to test conditions, accounting for other potential physiological factors which may
513 affect drowsiness (e.g. including time since last meal, hours of sleep), and the acceptability of
514 periodic drafts in naturally ventilated workplaces as a means of maintaining vigilance and
515 concentration.
.CC-BY 4.0 International licenseIt is made available under a
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/483750doi: bioRxiv preprint first posted online Nov. 30, 2018;
24
516 Acknowledgements
517 This work was carried out as part of the Refresh project funded by the Engineering and Physical
518 Sciences Research Council (EPSRC) Grants: EP/K021907/1, EP/K021834/1 and
519 EP/K021893/1. Thanks are required for John Lally for technical assistance. All data supporting
520 this study are openly available from the University of Southampton repository: (DOI TO BE
521 INSERTED ONCE PUBLISHED).
522
523 References
524
525 1. Cui S, Cohen M, Stabat P, Marchio D. CO2 tracer gas concentration decay method for
526 measuring air change rate. Build Environ. 2015;84:162–9.
527 2. Fisk WJ. The ventilation problem in schools: literature review. Vol. 27, Indoor Air.
528 2017. p. 1039–51.
529 3. Etheridge D. A perspective on fifty years of natural ventilation research. Build Environ.
530 2015;91:51–60.
531 4. Maula H, Hongisto V, Naatula V, Haapakangas A, Koskela H. The effect of low
532 ventilation rate with elevated bioeffluent concentration on work performance, perceived
533 indoor air quality, and health symptoms. Indoor Air. 2017;27(6):1141–53.
534 5. Mendell MJ, Heath GA. Do indoor pollutants and thermal conditions in schools
535 influence student performance? A critical review of the literature. Indoor Air.
536 2005;15(1):27–52.
537 6. Wargocki P, Sundell J, Bischof W, Brundrett G, Fanger PO, Gyntelberg F, et al.
538 Ventilation and health in non-industrial indoor environments: Report from a European
539 Multidisciplinary Scientific Consensus Meeting (EUROVEN). Vol. 12, Indoor Air.
540 2002. p. 113–28.
541 7. Wyon DP. The effects of indoor air quality on performance and productivity. Indoor
.CC-BY 4.0 International licenseIt is made available under a
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/483750doi: bioRxiv preprint first posted online Nov. 30, 2018;
25
542 Air. 2004;14 Suppl 7(Suppl 7):92–101.
543 8. Maddalena R, Mendell MJ, Eliseeva K, Chan WR, Sullivan DP, Russell M, et al. Effects
544 of ventilation rate per person and per floor area on perceived air quality, sick building
545 syndrome symptoms, and decision-making. Indoor Air. 2015;25(4):362–70.
546 9. Fang L, Wyon DP, Clausen G, Fanger PO. Impact of indoor air temperature and
547 humidity in an office on perceived air quality, SBS symptoms and performance. Indoor
548 Air, Suppl. 2004;14(SUPPL. 7):74–81.
549 10. Zhang X, Wargocki P, Lian Z, Thyregod C. Effects of exposure to carbon dioxide and
550 bioeffluents on perceived air quality, self-assessed acute health symptoms, and cognitive
551 performance. Indoor Air. 2017;27(1):47–64.
552 11. Wargocki P, Wyon DP. Ten questions concerning thermal and indoor air quality effects
553 on the performance of office work and schoolwork. Build Environ. 2017;112:359–66.
554 12. Zhang X, Wargocki P, Lian Z. Human responses to carbon dioxide, a follow-up study
555 at recommended exposure limits in non-industrial environments. Build Environ.
556 2016;100:162–71.
557 13. Zhang X, Wargocki P, Lian Z. Effects of Exposure to Carbon Dioxide and Human
558 Bioeffluents on Cognitive Performance. Procedia Eng. 2015;121:138–42.
559 14. Vehviläinen T, Lindholm H, Rintamäki H, Pääkkönen R, Hirvonen A, Niemi O, et al.
560 High indoor CO 2 concentrations in an office environment increases the transcutaneous
561 CO 2 level and sleepiness during cognitive work. J Occup Environ Hyg. 2016;13(1):19–
562 29.
563 15. Allen JG, MacNaughton P, Satish U, Santanam S, Vallarino J, Spengler JD.
564 Associations of cognitive function scores with carbon dioxide, ventilation, and volatile
565 organic compound exposures in office workers: A controlled exposure study of green
566 and conventional office environments. Environ Health Perspect. 2016;124(6):805–12.
.CC-BY 4.0 International licenseIt is made available under a
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/483750doi: bioRxiv preprint first posted online Nov. 30, 2018;
26
567 16. Satish U, Mendell MJ, Shekhar K, Hotchi T, Sullivan D, Streufert S, et al. Is CO2 an
568 indoor pollutant? Direct effects of low-to-moderate CO2 concentrations on human
569 decision-making performance. Environ Health Perspect. 2012 Sep 20;120(12):1671–7.
570 17. Kajtár L, Herczeg L. Influence of carbon-dioxide concentration on human well-being
571 and intensity of mental work. Idojaras. 2012;116(2):145–69.
572 18. Liu W, Zhong W, Wargocki P. Performance, acute health symptoms and physiological
573 responses during exposure to high air temperature and carbon dioxide concentration.
574 Build Environ. 2017;114:96–105.
575 19. Zhang X, Wargocki P, Lian Z. Physiological responses during exposure to carbon
576 dioxide and bioeffluents at levels typically occurring indoors. Indoor Air.
577 2017;27(1):65–77.
578 20. Tanabe S, Nishihara N. Productivity and fatigue. Indoor Air, Suppl. 2004 Aug
579 1;14(SUPPL. 7):126–33.
580 21. Lal SKL, Craig A. Driver fatigue: Electroencephalography and psychological
581 assessment. Psychophysiology. 2002;39(3):313–21.
582 22. De Gennaro L, Marzano C, Veniero D, Moroni F, Fratello F, Curcio G, et al.
583 Neurophysiological correlates of sleepiness: A combined TMS and EEG study.
584 Neuroimage. 2007;39(3):313–21.
585 23. Gorgoni M, Ferlazzo F, Ferrara M, Moroni F, D’Atri A, Fanelli S, et al. Topographic
586 electroencephalogram changes associated with psychomotor vigilance task performance
587 after sleep deprivation. Sleep Med. 2014;15(9):1132–9.
588 24. Trejo LJ, Kubitz K, Rosipal R, Kochavi RL, Montgomery LD, Rosepal R, et al. EEG-
589 Based Estimation and Classification of Mental Fatigue. Psychology. 2015;(April):1–44.
590 25. Wargocki P, Wyon DP, Sundell J, Clausen G, Fanger PO. The Effects of Outdoor Air
591 Supply Rate in an Office on Perceived Air Quality, Sick Building Syndrome (SBS)
.CC-BY 4.0 International licenseIt is made available under a
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/483750doi: bioRxiv preprint first posted online Nov. 30, 2018;
27
592 Symptoms and Productivity. Indoor Air. 2000;10(4):222–36.
593 26. Donaldson SI, Grant-Vallone EJ. Understanding self-report bias in organizational
594 behavior research. J Bus Psychol. 2002;17(2):245–60.
595 27. Rasheed EO, Byrd H. Can self-evaluation measure the effect of IEQ on productivity? A
596 review of literature. Facilities. 2017;35(11/12):601–21.
597 28. Niedermeyer E, Silva FHL Da. Electroencephalography: Basic Principles, Clinical
598 Applications, and Related Fields. 1st ed. Lippincott Williams and Wilkins. Lippinciott
599 Williams and Wilkins; 2004. 1309 p.
600 29. Landstrom U, Englund K, Nordstrom B, Stenudd A. Laboratory studies on the effects
601 of temperature variations on drowsiness. Percept Mot Skills. 1999;83(3):1217–29.
602 30. Wang D, Yee BJ, Wong KK, Kim JW, Dijk DJ, Duffin J, et al. Comparing the effect of
603 hypercapnia and hypoxia on the electroencephalogram during wakefulness. Clin
604 Neurophysiol. 2015;126(1):103–9.
605 31. Xu F, Uh J, Brier MR, Hart J, Yezhuvath US, Gu H, et al. The influence of carbon
606 dioxide on brain activity and metabolism in conscious humans. J Cereb Blood Flow
607 Metab. 2011;31(1):58–67.
608 32. Thesen T, Leontiev O, Song T, Dehghani N, Hagler DJ, Huang M, et al. Depression of
609 cortical activity in humans by mild hypercapnia. Hum Brain Mapp. 2012;33(3):715–26.
610 33. Bloch-Salisbury E, Lansing R, Shea S a. Acute changes in carbon dioxide levels alter
611 the electroencephalogram without affecting cognitive function. Psychophysiology.
612 2000;37(4):418–26.
613 34. Iber C, Ancoli-Israel S, Chesson AL Jr. QS for the AA, Medicine. of S, Iber CC, Ancoli-
614 Israel S, Chesson AL, Quan SF, et al. The AASM Manual for the Scoring of Sleep and
615 Associated Events: Rules, Terminology and Technical Specifications. In: Berry, R.B.,
616 Brooks, R., Gamaldo, C.E., Harding, S.M., Marcus, C.L. and Vaughn BV, editor.
.CC-BY 4.0 International licenseIt is made available under a
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/483750doi: bioRxiv preprint first posted online Nov. 30, 2018;
28
617 AASM Manual for Scoring Sleep. Westchester, IL; 2007.
618 35. Lamb S, Kwok KCS. A longitudinal investigation of work environment stressors on the
619 performance and wellbeing of office workers. Appl Ergon. 2016;52:104–11.
620 36. Garner M, Attwood A, Baldwin DS, Munafò MR. Inhalation of 7.5% carbon dioxide
621 increases alerting and orienting attention network function. Psychopharmacology (Berl).
622 2012;223(1):67–73.
623 37. Faul F, Erdfelder E, Lang A-G, Buchner A. G*Power 3: A flexible statistical power
624 analysis program for the social, behavioral, and biomedical sciences. Behav Res
625 Methods. 2007;39(2):175–91.
626 38. Laussmann D, Helm D. Air Change Measurements Using Tracer Gases. Chem Emiss
627 Control Radioact Pollut Indoor Air Qual. 2004;1:365–404.
628 39. Gough HL, Luo Z, Halios CH, King MF, Noakes CJ, Grimmond CSB, et al. Field
629 measurement of natural ventilation rate in an idealised full-scale building located in a
630 staggered urban array: Comparison between tracer gas and pressure-based methods.
631 Build Environ. 2018;137:246–56.
632 40. Hedge A, Erickson WA. A study of indoor environment and sick building syndrome
633 complaints in air-conditioned offices: benchmarks for facility performance. Int J Facil
634 Manag. 1997;1(4):185–92.
635 41. Watson D, Clark LA, Tellegen A. Development and validation of brief measures of
636 positive and negative affect: The PANAS scales. J Pers Soc Psychol. 1988;54(6):1063–
637 70.
638 42. Quest S. Stanford Sleepiness Scale [Internet]. Stanford Sleepiness Scale. 2000 [cited
639 2017 Jan 31]. p. 1. Available from: https://web.stanford.edu/~dement/sss.html
640 43. American Society of Heating Refrigerating and Air Conditioning Engineers (ASHRAE).
641 ANSI/ASHRAE Standard 55: Thermal Environmental Conditions for Human
.CC-BY 4.0 International licenseIt is made available under a
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/483750doi: bioRxiv preprint first posted online Nov. 30, 2018;
29
642 Occupancy. [Internet]. ASHRAE. 2013. Available from:
643 https://www.ashrae.org/technical-resources/bookstore/thermal-environmental-
644 conditions-for-human-occupancy
645 44. Biswas D, Bono V, Scott-South M, Chatterjee S, Soska A, Snow S, et al. Analysing
646 wireless EEG based functional connectivity measures with respect to change in
647 environmental factors. In: 3rd IEEE EMBS International Conference on Biomedical and
648 Health Informatics, BHI 2016. 2016. p. 599–602.
649 45. Bono V, Jamal W, Das S, Maharatna K. Artifact reduction in multichannel pervasive
650 EEG using hybrid WPT-ICA and WPT-EMD signal decomposition techniques. In:
651 ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing -
652 Proceedings. 2014. p. 5864–8.
653 46. Tatum W, Husain A, Benbadis S, Kaplan P. Handbook of EEG Interpretation. Medicine.
654 New York: Demos Medical Publishing; 2008.
655 47. Tinguely G, Finelli LA, Landolt HP, Borbély AA, Achermann P. Functional eeg
656 topography in sleep and waking: State-dependent and state-independent features.
657 Neuroimage. 2006;32(1):283–92.
658 48. Wargocki P, Da Silva NAF. Use of visual CO2 feedback as a retrofit solution for
659 improving classroom air quality. Indoor Air. 2015 Feb 1;25(1):105–14.
660 49. Mendell MJ, Mirer AG. Indoor thermal factors and symptoms in office workers:
661 findings from the us epa base study. Indoor Air. 2009;19(4):291–302.
662 50. Maula H, Hongisto V, Koskela H, Haapakangas A. The effect of cooling jet on work
663 performance and comfort in warm office environment. Vol. 104, Building and
664 Environment. 2016. p. 13–20.
665
.CC-BY 4.0 International licenseIt is made available under a
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/483750doi: bioRxiv preprint first posted online Nov. 30, 2018;
.CC-BY 4.0 International licenseIt is made available under a
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/483750doi: bioRxiv preprint first posted online Nov. 30, 2018;
.CC-BY 4.0 International licenseIt is made available under a
(which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint. http://dx.doi.org/10.1101/483750doi: bioRxiv preprint first posted online Nov. 30, 2018;
... Waard and Brookhuis [25] showed that the relative energy parameter ((α+θ)/β) of the driver decreased as the driving task continued. Snow et al. [14,26] adopted the EEG band powers as an objective measurement of drowsiness to study the CO 2 effects on human responses. They [26] reported that EEG was more closely resembled drowsiness during a short duration exposure at the CO 2 concentration of 2700 ppm compared to 830 ppm, but the CO 2 -induced drowsiness was only observed among the participants who had slept less in previous night [14]. ...
... Snow et al. [14,26] adopted the EEG band powers as an objective measurement of drowsiness to study the CO 2 effects on human responses. They [26] reported that EEG was more closely resembled drowsiness during a short duration exposure at the CO 2 concentration of 2700 ppm compared to 830 ppm, but the CO 2 -induced drowsiness was only observed among the participants who had slept less in previous night [14]. Moreover, Bloch-Salisbury et al. [27] found that the acute changes in CO 2 concentration altered the latency or amplitude of the N1, P2, or P3 event-related potential (ERP) without affecting the cognitive function. ...
Article
Fifteen healthy male college-age students were exposed in an enclosed environmental chamber to investigate their physiological responses to elevated carbon dioxide (CO2) concentration and mental workload. During each CO2 exposure condition (1626 ± 306 ppm, 3562 ± 259 ppm and 5087 ± 318 ppm), a resting test was conducted first, and then the multi-attribute tasks with three different mental workload levels were performed in the order designed by the Latin square method. The breathing, electrocardiogram (ECG) and electroencephalogram (EEG) data were collected overlapping the period when participants were in the resting and task tests. The statistical analyses indicated that the elevated CO2 concentration had positive correlations with the breathing wave amplitude, the percentage of adjacent normal-to-normal intervals that differ from each other by more than 50 milliseconds (pNN50), and the beta relative power of EEG during the task tests, but no significant response was observed during the resting test. The elevated mental workload was associated with increased arousal level as indicated by the decreased standard deviation of the normal-to-normal interval (SDNN), increased respiration rate and higher beta relative power of EEG, which could help explain the inverted U-shaped effect of mental workload on task performance. The physiological responses indicated that more effort was required to accomplish tasks when participants were exposed to high CO2 concentration or heavy mental workload, which led to a direct impact on task performance. The order of CO2 exposures was not balanced among the participants, which should be considered as a major limitation of this study.
... The coefficient value is −2.074 and the p value is lower than 0.05. This could be due to the possible appearance of drowsiness [96]. ...
Article
Full-text available
Globalization has increased the number of road trips and vehicles. The result has been an intensification of traffic accidents, which are becoming one of the most important causes of death worldwide. Traffic accidents are often due to human error, the probability of which increases when the cognitive ability of the driver decreases. Cognitive capacity is closely related to the driver’s mental state, as well as other external factors such as the CO2 concentration inside the vehicle. The objective of this work is to analyze how these elements affect driving. We have conducted an experiment with 50 drivers who have driven for 25 min using a driving simulator. These drivers completed a survey at the start and end of the experiment to obtain information about their mental state. In addition, during the test, their stress level was monitored using biometric sensors and the state of the environment (temperature, humidity and CO2 level) was recorded. The results of the experiment show that the initial level of stress and tiredness of the driver can have a strong impact on stress, driving behavior and fatigue produced by the driving test. Other elements such as sadness and the conditions of the interior of the vehicle also cause impaired driving and affect compliance with traffic regulations.
... Health effects of carbon dioxide exposure[36][37][38][39][40][41][42][43][44][45]. ...
Article
Full-text available
The current studies on carbon dioxide (CO2) emissions and the impacts on public health focus mainly on evaluating CO2 emissions from two types of emission sources. The first is a fixed source type from industrial plants, which can be controlled or easily evaluated. The second is a mobile source type from the transport sector, especially from medium- and heavy-duty vehicles (MHDVs), which produce high emissions. However, the common methods of evaluation of the average value of CO2 emissions per kilometer of a vehicle use a general Intergovernmental Panel on Climate Change (IPCC) model that does not consider the topography or road conditions. This affects the accuracy of CO2 emission assessments and, in turn, affects the accuracy of any analysis needed to establish health policies and the management of public health within the affected area. In this paper, therefore, we present the development of emission coefficient calculations with varying topography conditions for MHDVs with consideration of the health effects on the populace. The study area was the Asian highway network in Thailand that passes through all regions and is geographically diverse. Data were collected from the Department of Highway’s records, global positioning system (GPS) and electronic control unit (ECU) with data consisting of road data, slope, distance, traffic level and vehicle weight, as well as fuel consumption along the transportation route. To analyze and map out the correlation of the CO2 emission coefficients for each road slope, we determined the coefficient of the CO2 emissions using multiple linear regression analysis and validated this using Huber–White robust standard errors for heteroscedasticity. This method was more cost-efficient and time-efficient compared to the conventional approaches. We also created CO2 emission maps with risk points for health effects, and we propose policy designs and plans to manage the traffic level in each of these areas prone to higher levels of such emissions.
... 2018). Symptoms such as fatigue and drowsiness caused directly by CO 2 have been demonstrated by the use of electroencephalogram (EEG) techniques (Snow et al. 2018). In a study of office workers, a 20% increase in blood CO 2 , to significantly above normal levels, was measured along with sleepiness, headaches, heart rate variation and poor concentration in air that averaged 2,800 ppm CO 2 (Vehviläinen et al. 2016) while lowered heart rate and arousal level (fatigue) is clear at 4000 ppm (Xia et al. 2020). ...
Working Paper
Full-text available
As atmospheric levels of carbon dioxide continue to escalate and drive climate change, the safe level for breathing remains unclear. The toxicity of CO2 for breathing has been defined for high concentrations but it is unknown what levels will compromise human health when individuals are perpetually exposed for their lifetime. There is now substantial evidence that permanent exposure, to CO2 levels predicted by the end of the century, will have significant effects on humans. Unhealthy blood CO2 concentrations have been measured from people in common indoor environments where reduced thinking ability and health symptoms have been observed at levels of CO2 above 600 ppm for relatively short-term exposures. Although humans and animals are able to deal with elevated levels of CO2 in the short-term due to various compensation mechanisms in the body, the persistent effects of these mechanisms may have severe consequences in a perpetual environment of elevated CO2. These include threats to life such as chronic inflammation, kidney failure, bone atrophy and loss of brain function. Human tissue calcification associated with carbonic anhydrase, the enzyme that converts CO2 in the body, along with other protein malfunctions, may be an existential threat. Existing research indicates that as ambient CO2 increases in the near-future, there will be an associated increase in cancers, neurological disorders and other conditions. Research is urgently required to clearly identify the severity and proximity of this risk, associated with the primary human function of breathing, being a major aspect of climate change.
Article
Environmental carbon dioxide (CO2) could affect various mental and physiological activities in humans, but its effect on daytime sleepiness is still controversial. In a randomized and counterbalanced crossover study with twelve healthy volunteers, we applied a combinational approach using classical frequentist and Bayesian statistics to analyze the CO2 exposure effect on daytime sleepiness and electroencephalogram (EEG) signals. Subjective sleepiness was measured by the Japanese Karolinska Sleepiness Scale (KSS‐J) by recording EEG during CO2 exposure at different concentrations: Normal (C), 4000 ppm (Moderately High: MH), and 40 000 ppm (high: H). The daytime sleepiness was significantly affected by the exposure time but not the CO2 condition in the classical statistics. On the other hand, the Bayesian paired t‐test revealed that the CO2 exposure at the MH condition might induce daytime sleepiness at the 40‐min point compared with the C condition. By contrast, EEG was significantly affected by a short exposure to the H condition but not exposure time. The Bayesian analysis of EEG was primarily consistent with results by the classical statistics but showed different credible levels in the Bayes’ factor. Our result suggested that the EEG may not be suitable to detect objective sleepiness induced by CO2 exposure because the EEG signal was highly sensitive to environmental CO2 concentration. Our study would be helpful for researchers to revisit whether EEG is applicable as a judgment indicator of objective sleepiness.
Article
Full-text available
In this paper, measurements of CO2 rate and its relation to temperature, humidity and wind speed were carried out at 13 different locations in Erbil city during March [ 2017 to 2018], it was found that CO2 rate is the same as global level (418ppm) for outdoor. Measurements of CO2 rates for indoor environment showed that using kerosene heaters (during winter season) caused a rapid increase of CO2 rate in a short time, which has harmful side effect on human health.
Article
Full-text available
Growing evidence suggests that environmentally relevant elevations in CO2 (<5,000 ppm) may pose direct risks for human health. Increasing atmospheric CO2 concentrations could make adverse exposures more frequent and prolonged through increases in indoor air concentrations and increased time spent indoors. We review preliminary evidence concerning the potential health risks of chronic exposure to environmentally relevant elevations in ambient CO2, including inflammation, reductions in higher-level cognitive abilities, bone demineralization, kidney calcification, oxidative stress and endothelial dysfunction. This early evidence indicates potential health risks at CO2 exposures as low as 1,000 ppm—a threshold that is already exceeded in many indoor environments with increased room occupancy and reduced building ventilation rates, and equivalent to some estimates for urban outdoor air concentrations before 2100. Continuous exposure to increased atmospheric CO2 could be an overlooked stressor of the modern and/or future environment. Further research is needed to quantify the major sources of CO2 exposure, to identify mitigation strategies to avoid adverse health effects and protect vulnerable populations, and to fully understand the potential health effects of chronic or intermittent exposure to indoor air with higher CO2 concentrations. SharedIt Link: https://rdcu.be/bLpLa
Article
Full-text available
Currently, no clear standards exist for determining urban building natural ventilation rates, especially under varying realistic meteorological conditions. In this study, ventilation rates are determined using tracer gas decay and pressure-based measurements for a full-scale (6 m tall) cube. The cube was either isolated (2 months of observations) or sheltered within a staggered array (7 months), for both single-sided and cross ventilation (openings 0.4 × 1 m). Wind speeds at cube height ranged between 0.04 m s¯¹ and 13.1 m s¯¹. Errors for both ventilation methods are carefully assessed. There is no discernible linear relation between normalised ventilation rates from the two methods, except for cross ventilation in the array case. The ratio of tracer gas and pressure derived ventilation rates is assessed with wind direction. For single-sided (leeward opening) cases it approached 1. For cross ventilation the ratio was closer to 1 but with more scatter. One explanation is that agreement is better when internal mixing is less jet-dominated, i.e. for oblique directions in the isolated case and for all directions for unsteady array flows. Sheltering may reduce the flushing rate of the tracer gas from the cube relative to internal mixing rate. This new dataset provides an extensive range of conditions for numerical model evaluation and for understanding uncertainty of ventilation rates. Knowledge of the latter is critical in building design.
Article
Purpose: The purpose of this literature review is to investigate the reliability of self-evaluation as a method for measuring the effect(s) of indoor environment quality (IEQ) on the productivity of office workers. The aim of this review is to identify the various constraints to its adequacy in measuring productivity. Design/methodology/approach: Thirty studies were selected from peer-reviewed sources and reviewed on their method of measuring productivity. These studies used self-evaluation (questionnaires or interview) as the sole method of measuring the effect of IEQ on productivity/performance. Findings: This review provides insight on the insufficiencies and biases prevalent in self-evaluation. Various issues that compromised the reliability of self-evaluation results in an office environment were discussed. It was concluded that self-evaluation is not reliable and does not accurately measure occupant productivity. Research limitations/implications: This study has been a review of past studies and their findings. Further studies that will provide empirical evidence are required to solely test the reliability of self-evaluation in measuring productivity and the effect of factors such as IEQ on it. Practical implications: The paper calls for further debate on occupant productivity measurement and how the various factors that affect it can be quantified into measurable entities. Originality/value: This paper fulfils an identified need to revisit the technique of self-evaluation as a method for measuring occupant productivity.
Conference Paper
In order to reduce the muscle artifacts in multi-channel pervasive Electroencephalogram (EEG) signals, we here propose and compare two hybrid algorithms by combining the concept of wavelet packet transform (WPT), empirical mode decomposition (EMD) and Independent Component Analysis (ICA). The signal cleaning performances of WPT-EMD and WPT-ICA algorithms have been compared using a signal-to-noise ratio (SNR)-like criterion for artifacts. The algorithms have been tested on multiple trials of four different artifact cases viz. eye-blinking and muscle artifacts including left and right hand movement and head-shaking.
Article
Human subjects were exposed for 3 h in a climate chamber to the air temperature of 35 °C that is an action level, at which the working time needs to be diminished in China. The purpose was to put this action level to test by measuring physiological responses, subjective ratings and cognitive performance, and compare them with responses at temperature of 26 °C (reference exposure). Moreover, CO2 was increased to 3000 ppm (CO2 exposure) at 35 °C to further examine, whether this change will have any effect on the measured responses. Compared with the reference exposure, exposure to 35 °C caused subjects to report feeling uncomfortably warm, to rate the air quality as worse, to report increased sleepiness and higher intensity of several acute health symptoms. Eardrum temperature, skin temperature, heart rate and body weight loss all increased significantly at this exposure, arterial oxygen saturation decreased significantly, while the percentage of adjacent inter-beat cardiac intervals differing by > 50 m (pNN50) decreased significantly, indicating elevated stress. The performance of addition and subtraction tasks decreased significantly during this exposure, as well. Increasing CO2 to 3000 ppm at 35 °C caused no significant changes in responses. Present results reaffirm the selection of 35 °C as an action level, and show that concurrently occurring high CO2 levels should not exacerbate the hazards.
Conference Paper
In this paper we present a systematic exploration to formulate a predictive model of the human cognitive process with the changing environmental conditions at workplace. We select six different environmental conditions with small change in temperature/ventilation representative of realistic work environment having manual control. EEG data were acquired through 19-channel wireless system from three participants and CO2, Temperature, Relative humidity were recorded throughout the six conditions. The EEG data was pre-processed using an artifact reduction algorithm and 129 neurophysiological features were extracted from functional connectivity measures using complex network analysis. The environmental data were processed to generate 15 time/frequency domain features. Five best features selected through a ranking algorithm for all the variables across the six conditions were processed to formulate a model (environmental parameters as predictors) using retrospective 10-fold crossvalidation in conjunction with multiple linear regression. The model was prospectively evaluated over 10 runs on a test set to predict the EEG variable across the six conditions and parameters corresponding to the run producing least root mean square error were reported. Our exploration shows that the condition having no modulation of the ambient environmental parameters reflects the optimum condition for predicting the EEG features using the examined environmental parameters.
Article
The aim of our study was to determine the effect of a cooling jet on performance and comfort in warm office environment. We compared cognitive performance, subjective workload, cognitive fatigue, thermal comfort, symptoms, perceived working conditions and perception of airflow in warm temperature (29.5 °C) in two conditions: with and without the jet. Twenty-nine students participated in the experiment in which a repeated measures design was employed. The jet improved the speed of response in a working memory task with increasing exposure time but did not affect other performance measures. Self-rated performance was higher and frustration was lower with the jet. Tiredness increased with increasing exposure time without the jet, but remained constant with the jet. Thermal comfort and perceived working conditions were improved, and indoor air was perceived fresher with the jet. Eye symptoms increased over time with the jet. The results support the use of cooling jet in offices with high thermal loads where individual control over the air temperature is not possible via air conditioning. It seems, however, that there is a need for individual control over the jet already when the target velocity is set to the upper limit according to ASHRAE standard.
Article
To extend the results of a previous study on the effects of carbon dioxide (CO2) and bioeffluents on humans, the new study reported in this paper was carried out. The purpose of this study was to examine, whether exposure to CO2 at 5,000 ppm would cause sensory discomfort, evoke acute health symptoms, reduce the performance of cognitive tasks, or result in changes in physiological responses. The outdoor air supply rate was set high enough in a low-emission stainless-steel climate chamber to create a reference condition with CO2 at 500 ppm when subjects were present, and chemically pure CO2 was added to the supply air to create an exposure condition with CO2 at 5,000 ppm (the measured exposure level was ca. 4,900 ppm). Ten healthy college-age students were exposed twice to each of the two conditions for 2.5 hours in a design balanced for order of presentation. The raised CO2 concentration had no effect on perceived air quality or physiological responses except for end-tidal CO2 (ETCO2), which increased more (to 5.3 kPa) than it was in the reference condition (5.1 kPa). Other results indicate additionally that a 2.5-hour exposure to CO2 up to 5,000 ppm did not increase intensity of health symptoms reported by healthy young individuals and their performance of simple or moderately difficult cognitive tests and some tasks resembling office work. These results accord well with the current occupational exposure limit recommendation for CO2 and with many other reports published in the literature.
Article
Twenty-five subjects were exposed to different levels of carbon dioxide (CO2 ) and bioeffluents. The ventilation rate was set high enough to create a reference condition of 500 ppm CO2 with subjects present; additional CO2 was then added to supply air to reach levels of 1,000 or 3,000 ppm, or the ventilation rate was reduced to allow metabolically-generated CO2 to reach the same two levels (bioeffluents increased as well). Heart rate, blood pressure, End-Tidal CO2 (ETCO2 ), oxygen saturation of blood (SPO2 ), respiration rate, nasal peak flow and forced expiration were monitored and the levels of salivary α-amylase and cortisol were analyzed. The subjects performed a number of mental tasks during exposures and assessed their levels of comfort and the intensity of their acute health symptoms. During exposure to CO2 at 3,000 ppm, when CO2 was added or ventilation was restricted, ETCO2 increased more and heart rate decreased less than the changes that occurred in the reference condition. Exposure to bioeffluents, when metabolically-generated CO2 was at 3,000 ppm, significantly increased diastolic blood pressure and salivary α-amylase level compared with pre-exposure levels, and reduced the performance of a cue-utilisation test: these effects may suggest higher arousal/stress. A model is proposed describing how mental performance is affected by exposure to bioeffluents. This article is protected by copyright. All rights reserved.