Acute changes in carbon dioxide levels alter the electroencephalogram without affecting cognitive function
Abstract
The partial pressure of carbon dioxide in the arterial blood (PaCO2) is usually tightly regulated, yet it varies among healthy people at rest (range approximately 32-44 mmHg) as well as within an individual during many natural life situations. The present study examined whether modest changes in end-tidal PCO2 (PETCO2; a noninvasive measure of PaCO2) affect electroencephalographic (EEG) activity, cognitive function, and vigilance. Nine adults were ventilated mechanically using a mouthpiece; respiratory rate and breath size were held constant while PETCO2 was set to levels that produced minimal discomfort. Despite discrete changes in EEG, neither acute PETCO2 increases (mean = 47 mmHg) nor decreases (mean = 30 mmHg) from resting levels (mean = 38 mmHg) affected performance on cognitive tasks, latency or amplitude of the N1, P2, or P3 event-related potential, or alertness. Modest changes in PETCO2 may cause significant alterations in the EEG without disturbing cognitive function.
Acute changes in carbon dioxide levels alter
the electroencephalogram without affecting
cognitive function
ELISABETH BLOCH-SALISBURY,
a
ROBERT LANSING,
b
and STEVEN A. SHEA
c
a
Physiology Program, Harvard School of Public Health, Boston, Massachusetts, USA
b
Department of Psychology, University of Arizona, Tucson, USA
c
Harvard Medical School, Brigham and Women’s Hospital, Boston, Massachusetts, USA
Abstract
The partial pressure of carbon dioxide in the arterial blood ~PaCO
2
! is usually tightly regulated, yet it varies among
healthy people at rest ~range ;32–44 mmHg! as well as within an individual during many natural life situations. The
present study examined whether modest changes in end-tidal PCO
2
~Pet
CO
2
; a noninvasive measure of PaCO
2
! affect
electroencephalographic ~EEG! activity, cognitive function, and vigilance. Nine adults were ventilated mechanically
using a mouthpiece; respiratory rate and breath size were held constant while Pet
CO
2
was set to levels that produced
minimal discomfort. Despite discrete changes in EEG, neither acute Pet
CO
2
increases ~mean5 47 mmHg! nor decreases
~mean 5 30 mmHg! from resting levels ~mean 5 38 mmHg! affected performance on cognitive tasks, latency or
amplitude of the N1, P2, or P3 event-related potential, or alertness. Modest changes in Pet
CO
2
may cause significant
alterations in the EEG without disturbing cognitive function.
Descriptors: Hypercapnia, Hypocapnia, ERPs, AAT
In healthy humans the partial pressure of carbon dioxide in the
arterial blood ~PaCO
2
! is normally tightly regulated ~for review
see Cunningham, Robbins, & Wolff, 1986!. Chemoreceptors within
the carotid bodies and brainstem are sensitive to changes in PCO
2
and provide a reflex control of breathing that adjusts gas exchange
in the lungs to maintain a relatively constant level of PaCO
2
.
Despite this automatic control, many natural life situations such as
speaking, singing, playing wind instruments, living at altitude,
breath holding, or breathing in confined spaces ~e.g., manned space
flights and space stations! can acutely increase or decrease PaCO
2
levels as much as 5 mmHg above or below a person’s average level
~e.g., Manzey & Lorenz, 1998; Shea, 1997!. The PaCO
2
also in-
creases during sleep ~e.g., Phillipson & Bowes, 1986! and through-
out the menstrual cycle ~e.g., Lyons, 1976!. In addition, variations
have been measured in the average level of PaCO
2
among healthy
individuals under resting conditions ranging from 32 to 44 mmHg
~e.g., Shea, Walter, Murphy, & Guz, 1987!. Among patients with
respiratory problems ~e.g., emphysema, hyperventilation syn-
drome, paralysis, patients on mechanical ventilation! the range can
be much greater, both within an individual and across patient pop-
ulations ~e.g., Bloch-Salisbury, Shea, Brown, Evans, & Banzett,
1996!. The central nervous system ~CNS! effects of such modest
acute changes in PaCO
2
and how these changes affect brain elec-
trical activity, cognitive function, and vigilance are largely unknown.
A variety of CNS effects have been demonstrated with marked
increases in PaCO
2
~hypercapnia! or decreases in PaCO
2
~hypo-
capnia!. There is a direct linear relationship between cerebral blood
flow and PaCO
2
in the range of 20–80 mmHg ~Michenfelder,
1988!. Vasodilation of precapillary arterioles increases blood flow
during hypercapnia; vasoconstriction reduces blood flow during
hypocapnia and can cause brain tissue hypoxia, which may further
alter CNS activity. Associated changes in brain electrical activity
with PaCO
2
changes are less clear ~e.g., Achenbach-Ng, Siao,
Mavroudakis, Chiappa, & Kiers, 1994; Schäfer, 1949; Van der
Worp, Kraaier, Wieneke, & Van Huffelen, 1991!. The effects of
hypercapnia on electroencephalographic ~EEG! activity have been
inconsistent: some studies report a negative correlation between
end-tidal PCO
2
~Pet
CO
2
; a noninvasive measure of PaCO
2
! and
alpha activity ~e.g., Kalkman et al., 1991!; others have shown
either a biphasic or triphasic effect of hypercapnia on cortical
activity, reflecting the stimulating and narcotic properties of var-
ious PCO
2
levels ~e.g., Gibbs, Williams, & Gibbs, 1940; Harter,
1967; Woodbury & Karler, 1960!. Reports on the effects of hypo-
capnia on the EEG are also equivocal. Although hypocapnia is
typically associated with an increase in “slow wave” activity and
Supported by the Spinal Cord Research Foundation, Paralyzed Veter-
ans of America.
We thank Christopher Kovacs and Robert Harrington for extensive help
in programming the event-related potential and cognitive tasks, and for
technical support. We thank the Walter Reed Army Research Institute for
the computerized Performance Assessment Battery, Dr. A. Eberhard for the
MacAster Software, and Dr. C. Stampi for discussions related to the Alpha
Attenuation Test.
Address reprint requests to: Elisabeth Bloch-Salisbury, Ph.D., Physi-
ology Program, Harvard School of Public Health, 665 Huntington Ave.,
Boston, MA 02115, USA. E-mail: ebsalis@hsph.harvard.edu.
Psychophysiology, 37 ~2000!, 418–426. Cambridge University Press. Printed in the USA.
Copyright © 2000 Society for Psychophysiological Research
418
a decrease in alpha activity ~e.g., Van der Worp et al., 1991!, slow
wave activity is rarely defined and, contradictorily, increases in
alpha activity have also been reported ~e.g., Kalkman et al., 1991!.
In addition to increasing brain blood flow and altering brain
electrical activity, acute hypercapnia ~Pet
CO
2
. 48 mmHg! has
been shown to adversely affect performance on some cognitive
and psychomotor tasks ~e.g., Fothergill, Hedges, & Morrison, 1991;
Henning, Sauter, Lanphier, & Reddan, 1990; Hesser, Fagraeus, &
Adolfson, 1978; Sayers, Smith, Holland, & Keatinge, 1987!. De-
spite the variety of tasks employed and the wide range of Pet
CO
2
studied, the general findings suggest that performance deteriorates
at elevated Pet
CO
2
. For example, Sayers et al. ~1987! observed that
at a Pet
CO
2
threshold of 51 mmHg, subjects’ performance on
reasoning tasks slowed ~whereas accuracy remained unaffected!.
Hesser et al. ~1978! suggested this threshold level of Pet
CO
2
to be
much lower, approaching normal Pet
CO
2
~i.e., 40 mmHg!. Fother-
gill et al. ~1991! also showed decrements on some mental perfor-
mance tasks with hypercapnia, but suggested that there was no
threshold Pet
CO
2
at which performance was necessarily affected.
What these studies have in common is that subjects were able to
spontaneously adjust their breathing to accommodate for the dis-
comfort associated with elevated Pet
CO
2
~such as the relief of “air
hunger” associated with increased breathing; see Bloch-Salisbury,
Spengler, Brown, & Banzett, 1998!. However, it is unclear whether
the cognitive deficits observed were a direct effect of PCO
2
on
brain state ~e.g., via vasodilation or cortical slowing!, or whether
the respiratory discomfort and distraction associated with changes
in breathing may have contributed to these deficits. Similarly, sub-
jects’ diminished ability to rehearse and recall information during
hypocapnia ~e.g., Marangoni & Hurford, 1990! may be related to
the distracting physiological effects of hypocapnia ~e.g., tingling
and dizziness! rather than direct effects on CNS function.
Currently, it is difficult to assess the effects of PaCO
2
on cor-
tical electrical activity and cognitive function because of the meth-
odological variability observed in previous studies: ~1! use of
spontaneously breathing subjects, which allowed for uncontrolled
changes in the breathing pattern ~i.e., respiratory frequency and0or
breath size; e.g., Fothergill et al., 1991; Harter, 1967; Sayers et al.,
1987!; ~2! use of different concentrations of PCO
2
, some of which
were measured as only the inspired fraction with no indication of
arterial PCO
2
~e.g., Harter, 1967!; ~3! use of a range of inhalation
~and equilibration! periods with some studies assessing cortical
function during the inhalation period and others reporting mea-
sures during the recovery period ~e.g., Harter, 1967; Henning et al.,
1990!; and ~4! variation among individuals in their sensitivity to
PCO
2
as indicated by irritability ~including anxiety or stress!, dis-
comfort, and other attendant physiological changes ~e.g., flushing,
headache, hyper- or hypotension; e.g., Fothergill et al., 1991; Ma-
rangoni & Hurford, 1990; Sayers et al., 1987!. In awake humans,
any of these observed factors can contribute to variations in cor-
tical function.
Despite inconsistencies in the literature, it is clear that at mod-
erately elevated or reduced PaCO
2
there are some dramatic effects
on the CNS and cortical function. Less clear is whether these
effects occur when there are more modest PaCO
2
changes, span-
ning the normal range. It seems possible that one of the functions
of a tightly regulated arterial blood gas in awake humans is to
optimize cognitive function and alertness. The present study ex-
amined whether modest acute increases and decreases in Pet
CO
2
affect EEG activity, cognitive function, and vigilance in the ab-
sence of changes in minute ventilation, partial pressure of arterial
oxygen ~PaO
2
!, and respiratory discomfort. To accomplish this,
healthy, awake subjects were trained to be passive on a mechanical
ventilator using a mouthpiece; respiratory rate, breath size, and
PaO
2
were held constant while Pet
CO
2
was manipulated to con-
trolled levels at which subjects reported minimal discomfort. A
variety of behavioral and electrophysiological measures were used
to cover a range of important aspects of human cortical activity
and function. Behavioral tests from the Walter Reed Perfor-
mance Battery ~PAB! were used to measure cognitive perfor-
mance ~attention, concentration and memory! and self-assessment
of alertness ~Thorne, Genser, Sing, & Hegge, 1985!. Electro-
physiological measures included an auditory event-related poten-
tial ~ERP! task to assess changes in sensory and attentional
components ~i.e., N1, P2, and P3; e.g., Squires & Ollo, 1986!,
the Alpha Attenuation Test ~AAT! to assess alertness level ~Stam-
pi, Michimori, & Stone, 1995!, and power spectral analyses to
assess discrete changes in the EEG.
Method
This study was approved by the internal review board at the Har-
vard School of Public Health. Subjects provided informed consent.
Subjects
Seven women and two men ~between the ages of 24 and 43 years
old! served as subjects. Subjects had no history of any respiratory,
cardiovascular, or neurological disorders, or head injuries resulting
in loss of consciousness within the 2 years prior to the study. All
subjects were informed that the study concerned the effects of CO
2
on the CNS but they were not familiar with previous work in
this area; three of the women had experience in respiratory
physiology.
Apparatus
Ventilation. Subjects were passively ventilated by mouth using a
pressurized mechanical ventilator ~Puritan Bennett MA-1! while
seated comfortably in a cushioned recliner chair. Inspired gas was
heated and humidified ~Puritan Bennett!. Inspired CO
2
was mixed
from one tank containing 10% CO
2
, 50% O
2
, and 40% N
2
, and
another tank of 50% O
2
and 50% N
2
. ~Hyperoxia was used to
minimize peripheral chemoreflexive drive. This precaution also
served to ensure that brain tissue O
2
would be high in all condi-
tions despite any fluctuations in brain blood flow brought about by
the change in PCO
2
.! Pet
CO
2
was held at the desired level by
manipulating the inspired CO
2
fraction ~Puritan Bennett mixing
valve!. Peak inspiratory flow was set at 50 lpm, and the maximum
warning pressure limit was set at 40 cmH
2
O. Expiration was pas-
sive. To prevent air leaks during the testing periods, each subject
kept their mouth sealed around a rubber mouthpiece that was at-
tached to the ventilatory circuit, and wore a noseclip.
Tidal PCO
2
was sampled at the mouthpiece and recorded with
an infrared gas analyzer ~Beckman LB2!. Gas analyzer responses
were adequate to achieve end-expiratory and end-inspiratory pla-
teaus. The analyzer was calibrated before each session with a
known concentration of gas. Airway pressure was measured at the
mouthpiece with a 656.0 cmH
2
O differential transducer ~Vali-
dyne!, and inspiratory flow was measured with a pneumotachom-
eter transducer ~Fleisch #2! attached to another pressure transducer
~62.0 cmH
2
O; Validyne!. Heart rate, arterial oxygen saturation
and blood pressure were monitored throughout the tests ~Propaq
106EL!. Respiratory signals were recorded and digitized at 50 Hz
during the PAB task, at 250 Hz during the ERP task ~SuperScope
II acquisition software, GW Instruments; Macintosh IICi, Macadios
CO
2
effects on EEG and cognition 419
II A0D Board! and at 128 Hz during the AAT task ~MacADC
V2.7, Macaster acquisition software; Macintosh IICi, Macadios II
A0D Board!.
EEG recording. During the ERP and AAT tasks, EEG and
electrooculographic ~EOG! activity were recorded using Grass gold-
cup scalp electrodes. Electrodes for the ERP task were placed at
Fz, Cz, and Pz, each referenced to the right earlobe. Electrodes for
the AAT task were placed at Pz and referenced to the right earlobe.
For both tasks, an electrode placed below the center of the right
eye monitored eye movements, and an electrode placed on the
forehead served as ground. The impedance of each electrode paired
with the reference was set below 10 kV. Ongoing EEG and EOG
activity were amplified with Grass EEG amplifiers ~model 8-10!
with a bandpass filter of 0.3–70 Hz. During the ERP task, EEG
signals were digitized at a rate of 250 Hz ~SuperScope II acquisi-
tion software, GW Instruments; Macintosh IICi, Macadios II A0D
Board!. Averaged ERPs were constructed offline; epochs contain-
ing EOG artifacts ~650 mV! were rejected from the average ~Su-
perScope II analysis software, GW Instruments!. ERPs were
averaged separately for rare and frequent auditory tones ~seeTasks!
over a 1,000-ms timebase, including a 100-ms prestimulus base-
line. For the AAT task, EEG signals were digitized at 128 Hz
~MacADC V2.7, Macaster acquisition software; Macintosh IICi,
Macadios II A0D Board!.
Procedures
Subjects participated in one ventilator training session lasting
approximately 2 hr, three practice sessions each lasting approxi-
mately 1 hr, and three experimental sessions each lasting approx-
imately 2 hr. For each subject, all sessions were completed within
a 23-day period; the experimental sessions were completed within
3–7 days.
Training session. During the training session, resting Pet
CO
2
was first measured via a thin plastic sampling tube at the nose
while the subject rested comfortably for ;5 min and breathed
spontaneously. After establishing resting Pet
CO
2
, subjects went on
the mouthpiece and were mechanically ventilated. They were
coached to relax to ensure passive ventilation as indicated by mouth-
pressure recordings. Each subject was ventilated at various settings
until a comfortable tidal volume ~i.e., breath size! and respiratory
rate were determined ~Table 1!. For all subjects, this level of
minute ventilation remained constant; as such it was set higher
than normal to enhance passivity on the ventilator and to allow for
comfort across a range of Pet
CO
2
levels ~Manning et al., 1992;
Remmers, Brooks, & Tenney, 1968!. The large ventilation, there-
fore, also generated a Pet
CO
2
that was well below the subject’s
resting level. Inspired CO
2
was adjusted to produce the Pet
CO
2
levels for the experimental conditions ~Table 1!: hypocapnia and
hypercapnia were set 7–10 mmHg below and above the sub-
ject’s resting level, respectively; normocapnia was set on average
within 1 mmHg ~range 0–4 mmHg! of the subject’s resting
Pet
CO
2
. Pet
CO
2
was held for 7–10 min at each level, and subjects
received 5–10-min rest periods off the ventilator between Pet
CO
2
changes.
To ensure that the Pet
CO
2
changes did not produce strong feel-
ings of discomfort, subjects were asked to stop the training trial
any time they felt uncomfortable; adjustments in tidal volume
and0or respiratory frequency were made accordingly. As a final
confirmation that the ventilatory parameters and Pet
CO
2
levels
were comfortable, subjects were asked to rate their level of respi-
ratory discomfort ~i.e., zero, slight, moderate, and extreme! at the
end of each training Pet
CO
2
condition. The Pet
CO
2
levels were set
for zero discomfort for the experimental sessions ~and subsequently,
were never found to exceed slight discomfort!.
Practice sessions. During the practice sessions, subjects prac-
ticed a battery of computerized cognitive tasks from the Walter
Reed PAB ~described below under Tasks!, which they later per-
formed during the experimental conditions. Subjects practiced these
tasks to control for learning effects and to promote training to
optimal performance ~Thorne et al., 1985!. Subjects participated in
three practice sessions: in the first practice session, subjects per-
formed the tasks three or four times while seated comfortably; in
the last two practice sessions, subjects performed the tasks three or
four times while passively ventilated at normocapnia.
Experimental sessions. Subjects participated in three experi-
mental sessions, each performed at approximately the same time of
day to control for diurnal rhythms. Each individual task was com-
pleted within a single session at each of the predetermined levels
of Pet
CO
2
~hypocapnia, normocapnia, and hypercapnia!. This ap-
proach ensured that the Pet
CO
2
conditions to be compared within
each task were close in time, such that any day-to-day effects on
Table 1. Subject Characteristics and Mean Ventilatory Parameters Used in the Experimental Tasks
Subject Gender
Age
~years!
Weight
~kg!
V
T
~ml!
Rate
~bpm!
^V
E
~l0m!
Resting Pet
CO
2
~mmHg!
Low Pet
CO
2
~mmHg!
Normal Pet
CO
2
~mmHg!
High Pet
CO
2
~mmHg!
1 F 31 62 1400 12 16.8 37 29 37 44
2 F 40 59 1000 16 16.0 41 31 41 51
3 F 24 64 900 18 16.2 40 29 39 47
4 F 26 52 900 13 11.7 44 32 40 50
5 M 43 80 1300 14 18.2 42 31 40 49
6 M 24 77 1150 15 17.3 37 28 37 43
7 F 35 67 1000 13 13.0 37 28 37 46
8 F 27 52 900 16 14.4 36 27 36 44
9 F 23 48 900 10 9.0 38 34 39 46
Mean 30 62 1050 14 14.7 39 30 38 47
Note: V
T
5 tidal volume; Rate 5 respiratory frequency; ^V
E
5 minute ventilation.
420 E. Bloch-Salisbury, R. Lansing, and S.A. Shea
Pet
CO
2
~e.g., related to phase of the menstrual cycle! could be
ignored. In one experimental session subjects performed the com-
puterized cognitive tasks; in another session cortical ERPs to au-
ditory stimuli were recorded; and in another session EEG was
monitored during standardized conditions to assess vigilance ~based
on alpha-wave attenuation! and EEG power spectra. These tasks
are described in detail below ~see Tasks!. The order of the three
sessions was counterbalanced among subjects. The order of the
Pet
CO
2
conditions was counterbalanced among subjects. Subjects
were passively ventilated for 5 min at the selected level of Pet
CO
2
before beginning the test, and received a 5–10-min rest period, off
of the ventilator, between conditions.
Tasks
Cognitive performance tasks. The Walter Reed PAB is a comput-
erized battery of psychomotor, perceptual, and cognitive tasks
~Thorne et al., 1985! that has been shown to be sensitive to changes
in cognitive performance induced by such variables as hypoxia
~Crowley et al., 1992! and sleep deprivation ~Newhouse et al.,
1989!. Performance on the tasks is measured by both response
times and error scores.
The battery used in the present study included six tasks: pattern
recognition; matching-to-sample; logical reasoning; two-letter
search; time estimation; and self-assessment of alertness. In half of
the subjects this order of tasks was reversed. In both batteries,
however, subjects always received the alertness scale at the begin-
ning and end of the test. Subjects received visual instructions for
each task from a color video monitor; a set of detailed instructions
was given during the initial practice sessions and a shortened form
of instructions was subsequently presented during the later practice
and all experimental sessions. Subjects were instructed to respond
as quickly and accurately as possible, and made all responses with
their dominant, right hand using a standard computer keyboard that
had been placed on a board across the subject’s lap. Throughout
each condition, subjects remained on the mouthpiece and wore
noseclips and sound-attenuatingheadphones ~Peltor twin cup H10A!
that masked extraneous noise.
Cortical ERPs. ERPs reflect reproducible cortical responses to
both sensory and psychological variables ~Regan, 1989!. To obtain
ERPs, multiple EEG epochs are aligned at the onset of a stimulus
presentation and averaged. “N1” and “P2” are sensory-evoked
potentials that reflect the integrity of the sensory pathway and vary
in morphology with stimulus parameters such as intensity and
frequency. “P3,” on the other hand, is an endogenous potential
influenced primarily by factors extrinsic to the physical stimulus.
It is a positive potential, typically occurring 250–500 ms following
attention to an unexpected stimulus. The speed with which a cog-
nitive decision is processed is reflected in P3 latency; attentional
and motivational changes are reflected in P3 amplitude.
During each of the experimental conditions, 300 tones ~80 dB;
50-ms duration; 1.2-s interstimulus interval! were presented bin-
aurally ~Peltor twin cup H10A headphones!. Subjects kept a silent
count of rare auditory tones ~15% probability; 15 kHz! inter-
spersed among frequent tones ~85% probability; 2.5 kHz!. Prior to
the first Pet
CO
2
condition, subjects received a brief practice period
containing 20 tones to familiarize them with the task. Subjects
were instructed to keep their eyes open and focused on a point
approximately 60 cm in front of them to minimize eye movement,
and to remain on the mouthpiece. They wore the noseclip and
headphones. At the end of each condition, subjects informed the
experimenter of their count.
EEG activity.
1. EEG frequency0power spectra: The EEG was subjected to Fast
Fourier Transform analysis for determination of the centroid
frequency and total power ~MacAster software!.
2. AAT: The AAT is a quantitative, electrophysiological measure
of human alertness ~Stampi et al., 1995!. Alertness level is
defined as the ratio between the mean EEG absolute alpha
power ~8–12 Hz! with eyes closed and eyes open. Accordingly,
alertness is associated with a relatively high AAT ratio and
drowsiness with a relatively low AAT ratio, reflecting the fact
that when a subject is alert, eye closing increases alpha activity
and eye opening decreases it. The opposite is true during drows-
iness; the EEG power in the alpha frequency decreases when
the eyes are closed and increases when the eyes are open.
During the experimental session, subjects alternated between
six, 1-min intervals of eyes-closed and eyes-open periods for each
Pet
CO
2
condition. The experimenter told the subject when to close
and open their eyes. At the beginning of each Pet
CO
2
condition,
subjects were instructed to keep their eyes as still as possible in
order to minimize eye movement and to remain on the mouthpiece.
They wore the noseclip and sound-attenuating headphones. The
same data was used for both AAT and power spectral analyses.
Results
Ventilatory Parameters
Subject characteristics and the mean ventilatory settings used for
the experimental conditions are listed in Table 1. Note that the
mechanical ventilator was set at the same tidal volume and respi-
ratory rate for each condition, and that the Pet
CO
2
levels listed in
each condition reflect the mean across tasks. Separate one-way
analysis of variance ~ANOVA! for each experimental task con-
firmed a highly significant difference among Pet
CO
2
levels @rest-
ing, hypocapnia 5 low Pet
CO
2
, normocapnia 5 normal Pet
CO
2
,
hypercapnia 5 high Pet
CO
2
;
PAB: F~3,32! 5 62.42, p , .0001;
ERP: F~3,32! 5 56.60, p , .0001; AAT: F~3,32! 5 51.09, p ,
.0001#. Post hoc analysis using Scheffe’s test confirmed that for
each task resting Pet
CO
2
was the same as normal Pet
CO
2
, whereas
there was a significant difference between each of the Pet
CO
2
conditions ~ p , .05!. Separate one-way ANOVAs revealed that
there were no Pet
CO
2
level effects ~hypocapnia, normocapnia, hy-
percapnia! on inspiratory peak pressure ~mean 5 13.3 cmH
2
O,
SD 5 4.3!; inspiratory time ~mean 5 1.7 s, SD 5 0.4!, expiratory
time ~mean 5 2.4 s, SD 5 0.5!, inspiratory volume ~mean 5 1.1
liters, SD 5 0.2!, or expiratory volume ~mean 5 1.0 liters, SD 5
0.2! for the ERP and PAB tasks; these ventilatory parameters were
not assessed during the EEG task. Oxygen saturation never fell
below 98%. All of these findings were expected, as dictated by the
experimental protocol, and support the premise that ventilation
was held constant across Pet
CO
2
conditions.
Autonomic Activity
Systolic blood pressure, diastolic blood pressure, and heart rate
measures were subjected to separate repeated-measures ANOVA
testing for effects due to Pet
CO
2
level ~hypocapnia, normocapnia,
hypercapnia!, task ~PAB, ERP, EEG!, and time ~measures taken
immediately following the 5-min equilibration period preceding
the start of a task vs. measures taken immediately following the
completion of the task!. There was a marginally significant effect
CO
2
effects on EEG and cognition 421
of Pet
CO
2
level on systolic blood pressure, F~2,16! 5 3.83,
p 5 .063, Huynh-Feldt
E
5 0.73. Post hoc comparisons revealed
that systolic blood pressure was significantly greater during high
Pet
CO
2
~mean 5 114 mmHg! than either low Pet
CO
2
~mean 5
109 mmHg; paired t test, p , .001! or normal Pet
CO
2
~mean 5
110 mmHg; paired t test, p 5 .001!. Neither heart rate ~mean 5 63
bpm! nor diastolic blood pressure ~mean 5 70 mmHg! were af-
fected by Pet
CO
2
level, task, or time. There was a small but sig-
nificant interaction between Pet
CO
2
and time for heart rate due
to an elevated heart rate in the pretest measurement during low
Pet
CO
2
, F~2,16! 5 4.85, p5 .027; Huynh-Feldt
E
5 0.90; and there
was a marginally significant interaction between task and time due
to an elevated heart rate in the posttest measurement for the PAB
task, F~2,16! 5 3.43, p 5 .064; Huynh-Feldt
E
5 0.91.
Cognitive Performance Tasks
For each PAB task, separate one-wayANOVAs revealed that changes
in Pet
CO
2
did not affect error scores or response times on any of
the four cognitive tasks ~pattern recognition, matching-to-sample,
logical reasoning, two-letter search!. Pet
CO
2
levels also did not
affect time estimation or self-assessment of alertness ~Figure 1!.
Post hoc power analyses, as described by Cohen ~1977!, were
performed on the four cognitive tasks ~separately for both the error
scores and response times! and on the time estimation task. Effect
sizes were calculated from the observed sample data. Using a
significance criterion set at 0.05, and power set at 0.80, the number
of subjects needed to attain significance was calculated for these
and all subsequent power analyses. A small effect size ~ f range 5
0.092–0.176! was found for each of the tasks, with the exception
Figure 1. Mean responses to each of the computerized tasks for each end-tidal PCO
2
~Pet
CO
2
! condition: Pattern5 pattern recognition;
Match 5 match-to-sample; Logic 5 logical reasoning; Search 5 two letter search and recognition; Time 5 time estimation; Alertness
scale 5 self-assessment of alertness with 1 reflecting most awake to 7 reflecting the struggle to remain awake. Bars 5 SEM.
422 E. Bloch-Salisbury, R. Lansing, and S.A. Shea
of the pattern recognition error scores ~ f 5 0.346! and time esti-
mation ~ f5 0.273!. These medium effect sizes indicate that pattern
recognition performance would probably reach statistical signifi-
cance with a sample size of 30 subjects, and time estimation with
45 subjects. By contrast, the other tests would only attain signif-
icant effects of Pet
CO
2
with sample sizes exceeding 140 subjects,
suggesting a negligible magnitude of any effect.
Cortical ERPs
For each subject, ERPs to the rare and frequent tones were sepa-
rately constructed for each Pet
CO
2
condition. N1, P2, and P3 la-
tency and amplitude measures were obtained for each electrode
site. Because of the better signal-to-noise ratio, N1 and P2 mea-
sures were averaged from the ERPs elicited by the frequent tones:
N1 constituted the first most negative trough following stimulus
onset ranging between 70–100 ms; P2 was defined by the peak
immediately following N1, ranging between 150–250 ms. P3 la-
tency and amplitude measures were averaged from the ERPs elic-
ited by the rare tones, and constituted the most positive peak
ranging between 290–420 ms. However, when the waveform con-
sisted of two peaks ~i.e., P3a and P3b; Squires, Squires, & Hill-
yard, 1975! the second peak was selected.
Grand-average ERPs for each Pet
CO
2
condition at each elec-
trode site are shown in Figure 2. Separate repeated-measures
Figure 2. Grand-average event-related potential ~ERP! waveforms to the rare tone for each end-tidal PCO
2
~Pet
CO
2
! condition at the
three midline sites.
CO
2
effects on EEG and cognition 423
ANOVAs were used to analyze differences in latencies and am-
plitudes for N1, P2, and P3, testing for effects due to Pet
CO
2
condition and electrode site. Changes in Pet
CO
2
level did not sig-
nificantly affect either latency or amplitude of any of these cortical
ERPs. Distribution effects were observed only for N1 and P2 am-
plitude: N1 was largest parietally @~PZ mean 523.53 mV! . ~FZ
mean 525.12 mV! 5 ~CZ mean 525.25 mV!; F~2,16! 5 13.58,
p 5 .001; Huynh-Feldt
E
5 0.79; paired t tests, p 5 .001#;P2
was smallest parietally @~PZ mean 5 3.43 mV! , ~FZ mean 5
4.85 mV! , ~CZ mean 5 6.43 mV!; F~2,16! 5 11.64, p 5 .001;
F~2,16! 5 11.64, p 5 .001; Huynh-Feldt
E
5 0.92; paired t test,
p 5 .001#.
Post hoc power analyses ~Cohen, 1977! were performed on the
latencies and amplitudes of N1, P2, and P3 for the Cz electrode
site. A small effect size was found for the ERP amplitudes ~ f 5
0.07, 0.04, and 0.14, respectively! and latencies ~ f 5 0.277, 0.04,
and 0.221, respectively!. This finding suggests that there is little
likelihood that Pet
CO
2
would have affected these cortical ERPs
even with much larger samples.
EEG Activity
EEG frequency0power spectra. Spectral analyses at the PZ site
were conducted over the frequency range of 0–64 Hz, for each one
minute period, and total power and centroid frequency were de-
rived from the entire spectrum. Data were analyzed separately with
repeated-measures ANOVAs using the factors of Pet
CO
2
level and
eyes ~closed vs. open!. Changes in Pet
CO
2
did affect EEG power
spectra ~total power and centroid frequency!. As illustrated in Fig-
ure 3, total power showed a significant decrease from hypocapnia
to hypercapnia, F~2,52! 5 4.28, p 5 .034; Huynh-Feldt
E
5 0.69.
Post hoc analyses revealed that total power was significantly greater
with hypocapnia than either normocapnia ~paired t test, p 5 .006!
or hypercapnia ~paired t test, p 5 .011!. Total power was also
greater during eyes closed ~mean 5 133.61 mV
2
! than eyes open
~mean 5 93.33 mV
2
!, F~1,26! 5 34.42, p , .001.
Centroid frequency was also significantly affected by Pet
CO
2
level ~see Figure 3!, F~2,52! 5 6.57, p 5 .005; Huynh-Feldt
E
5
0.85; centroid frequency was significantly lower during hypercap-
nia than during either normocapnia ~paired t test, p 5 .006! or
hypocapnia ~paired t test, p5 .005!. Centroid frequency was greater
with eyes closed ~mean 5 7.43 Hz! than eyes open ~mean 5
5.14 Hz!, F~1,26! 5 41.23, p , .001, and this was most prominent
during hypercapnia, Condition3 Eyes interaction: F~2,52! 5 3.70,
p 5 .046; Huynh-Feldt
E
5 0.75.
AAT. The Fast Fourier Transforms performed for the spectral
analyses were used to calculate the AAT coefficients. The AAT
ratio was determined for each two consecutive minute periods as
the ratio of mean absolute power in alpha frequency ~8–12 Hz!
during eyes closed to the mean absolute alpha power during
eyes open. Separate ratios were computed for each subject, for
each Pet
CO
2
condition. One-way ANOVA revealed that changes
in Pet
CO
2
did not affect alertness level ~AAT ratio!: low Pet
CO
2
,
mean 5 3.93; normal Pet
CO
2
, mean 5 4.04; high Pet
CO
2
, mean 5
4.53. Post hoc power analyses ~Cohen, 1977! performed on the
AAT ratio revealed a very small effect size ~ f 5 .0769!, suggesting
a great likelihood that Pet
CO
2
affects this experimental assessment
of alertness with only negligible magnitude.
Figure 3. Mean electroencephalogram ~EEG! total power spectra and mean EEG centroid frequency for each end-tidal PCO
2
~Pet
CO
2
!
condition. Bars 5 SEM.
424 E. Bloch-Salisbury, R. Lansing, and S.A. Shea
Discussion
The present investigation demonstrates that modest acute changes
in Pet
CO
2
that span the normal range do not affect cognitive func-
tion or alertness in humans, despite inducing significant changes in
EEG power spectra. Previous studies have shown that with large
acute increases or decreases in Pet
CO
2
, performance declines on
some cognitive tasks. In the present study, healthy, young adults
were passively mechanically ventilated at a constant level of minute
ventilation and oxygen saturation, and with minimal discomfort
~as indicated by no significant change in inspiratory pressure or
any other ventilatory parameter, as well as self-report! for three
different Pet
CO
2
conditions. It is possible, therefore, that the dec-
rements in cognitive function observed in previous studies were
due to alterations in ventilation, discomfort, or other attendant
physiological changes that typically occur with marked increases
or decreases in this blood gas.
Although the present study included a relatively small group of
subjects, the EEG power spectra was nonetheless sensitive to the
modest Pet
CO
2
changes employed. The other electrophysiological
measures and cognitive tests were not as sensitive to these blood
gas shifts. Although modest increases and decreases in Pet
CO
2
each caused small but significant changes in cortical frequency,
these perturbations in the EEG occurred with no apparent distur-
bance in cortical output measures, such as changes in cognition,
attention, and alertness. Post hoc power analyses suggest that there
is little likelihood that changes in Pet
CO
2
would affect many of
these other output measures even if we had conducted this study on
hundreds of subjects. We predominantly found very small effect
sizes among most of our tests. However, we did find medium
effect sizes ~as defined by Cohen, 1977! on pattern recognition
error scores, time estimation, and N1 latency. Further research to
explore the possible effect on these measures, and to elucidate the
possible mechanisms by which Pet
CO
2
may alter them, is war-
ranted. That the overall brain activity patterns changed in response
to modest changes in Pet
CO
2
, whereas ERPs, cognition, and vig-
ilance remained largely unaffected, may be indicative of the brain’s
ability to compensate under these circumstances. However, the
present findings provide no unequivocal evidence that PaCO
2
is
tightly regulated to optimize cognitive function or alertness.
Specific Findings
Autonomic. The Pet
CO
2
levels used in this study were large
enough to marginally induce some of the autonomic changes that
typically are observed ~e.g., Grönroos & Pertovaara, 1994!. For
example, despite no change in heart rate, systolic blood pressure
increased with elevated Pet
CO
2
. Nonetheless, the changes in
Pet
CO
2
were modest, and caused minimal respiratory discomfort
or other noticeable physiological responses.
Cognition. The effects of changes in Pet
CO
2
on mental perfor-
mance studied by other authors have been discussed at length in
the Introduction. Our results demonstrate that acute modest in-
creases or decreases in Pet
CO
2
do not affect performance on a
battery of computerized cognitive tasks. These findings are con-
sistent with those studies in which the manipulated Pet
CO
2
levels
did not induce superfluous clinical symptoms ~e.g., Sheehy, Ka-
mon, & Kiser, 1982!. As such, when mental performance or rea-
soning ability is impaired at moderately high or low levels of
Pet
CO
2
, it is difficult to discern whether it is the psychological
~e.g., dis-comfort, distraction! or physiological effects ~e.g., CNS,
vasoconstriction! that impede cognitive function. With modest
changes in Pet
CO
2
that minimally affect comfort or physiological
responses, cognitive function remains largely unaffected.
ERPs and EEG. Despite the inconsistencies in the literature on
the effects of Pet
CO
2
changes on the EEG ~see Introduction!,as
well as a void in the use of ERPs as a research tool in this field,
there has been much valuable related information advanced through
anesthesia studies. In some respects, anesthesia studies are more
like those of the present study because ventilation is controlled:
that is, during anesthesia, the subject is intubated and ventilated
~albeit, unconscious!. However, because alterations in the EEG are
likely affected by the anesthetic agents, which typically depress
sensitivity of the central chemoreceptors to changes in PaCO
2
,
comparisons between studies are made with caution ~e.g., Kalk-
man et al., 1991; Smith, Greene, Moore, & Keegan, 1994!. Similar
to our findings, Kalkman et al. ~1991! observed distinct changes in
EEG power spectra to hypocapnia and hypercapnia ~range ;23–
50 mmHg! in patients undergoing anesthesia. They, too, found a
significant increase in total power during hypocapnia with no sig-
nificant alteration in a somatosensory cortical evoked response ~of
the posterior tibial nerve!. Other studies involving animals have
also shown little effect of Pet
CO
2
changes on sensory evoked re-
sponses. For example, Freeman, Sohmer, and Silver ~1991! ob-
served that in anesthetized cats, there was no change in the auditory
brainstem-evoked response during moderate hypercapnia ~i.e.,
64 mmHg!, although severe hypercapnia ~i.e., 172 mmHg! did
induce delays in interpeak latencies. Similarly, Gravenstein, Sasse,
and Hogan ~1992! found that hypocapnia ~i.e., 20 mmHg! did
not induce severe deterioration of later subcortical and cortical
somatosensory-evoked potentials in anesthetized dogs. That ERPs
are preserved during modest-to-moderate changes in Pet
CO
2
are
consistent with our negative findings.
Vigilance. Moderate increases and decreases in Pet
CO
2
have
been shown to cause drowsiness in awake subjects ~e.g., Grönroos
& Pertovaara, 1994!. However, anecdotal reports suggest that mod-
est increases might in fact cause hypervigilance; presumably the
discomfort associated with hypercapnia induces an “arousal” phe-
nomenon. That no identifiable changes in alertness occurred with
modest changes in Pet
CO
2
in the present study suggests that the
variability in Pet
CO
2
observed in daily life should not diminish the
level of attentiveness required for routine activities. It is only when
Pet
CO
2
falls outside these limits that behaviors are likely adversely
affected.
Implications
There are practical reasons for clearly identifying the effects of
Pet
CO
2
changes on electrical brain activity, cognition, and vigi-
lance. Even though the modest levels used in this experiment
showed no performance decrements, determining “optimal”
Pet
CO
2
levels might in fact minimize risk and enhance perfor-
mance in some work situations. For example, there is a wide range
of occupations in which workers are exposed to various levels of
CO
2
~e.g., underwater divers, firefighters, astronauts, chemical treat-
ment plant employees!. It is important to identify those levels that
may affect job performance, vigilance, and personal safety. In
addition, there are patients with spinal cord injury or neuromus-
cular disease who are dependent on mechanical ventilation and
typically live at extremely low Pet
CO
2
levels, as well as patients
with lung disease who maintain at extremely elevated Pet
CO
2
.
Such patients may benefit from information on the differential
effects of Pet
CO
2
on cortical and cognitive function.At present, the
CO
2
effects on EEG and cognition 425
effects of long-term exposure to varying degrees of hypocapnia or
hypercapnia on cortical function remains largely unexplored. The
results of this study suggest little effect with modest acute changes
in Pet
CO
2
.
REFERENCES
Achenbach-Ng, J., Siao, T. C. P., Mavroudakis, N., Chiappa, K. H., &
Kiers, L. ~1994!. Effects of routine hyperventilation on PCO
2
and PO
2
in normal subjects: Implications for EEG interpretations. Journal of
Clinical Neurophysiology, 11, 220–225.
Bloch-Salisbury, E., Shea, S. A., Brown, R., Evans, K., & Banzett, R. B.
~1996!. “Air hunger” induced by acute increases in PCO
2
adapts to
chronic elevation of PCO
2
in ventilated humans. Journal of Applied
Physiology, 81, 949–956.
Bloch-Salisbury, E., Spengler, C. M., Brown, R., & Banzett, R. B. ~1998!.
Self-control and external control of mechanical ventilation give equal
air hunger relief American Journal of Respiratory and Critical Care
Medicine, 157, 415–420.
Cohen, J. ~1977!. Statistical power analysis for the behavioral sciences.
New York: Academic Press.
Cunningham, D. J. C., Robbins, P. A., & Wolff, C. B. ~1986!. Integration of
respiratory responses to changes in alveolar partial pressures of CO2
and O2 and arterial pH. In N. S. Cherniack & J. G. Widdicombe ~Eds.!,
Handbook of physiology, section 3. The respiratory system: Vol. II.
Control of breathing ~pp. 475–528!. Bethesda, MD: American Physi-
ological Society.
Crowley, J. S., Wesensten, N., Kamimori, G., Devine, J., Iwanyk, E., &
Balkin, T. ~1992!. Effect of high terrestrial altitude and supplemental
oxygen on human performance and mood. Aviation Space and Envi-
ronmental Medicine, 63, 696–701.
Fothergill, D. M., Hedges, D., & Morrison, J. B. ~1991!. Effects of CO
2
and
N
2
partial pressures on cognitive and psychomotor performance. Under-
sea Biomedical Research, 18, 1–19.
Freeman, S., Sohmer, H., & Silver, S. ~1991!. The effect of stimulus rep-
etition rate on the diagnostic efficacy of the auditory nerve-brain-stem
evoked response. Electroencephalography and Clinical Neurophysiol-
ogy, 78, 284–290.
Gibbs, F. A., Williams, D., & Gibbs, E. L. ~1940!. Identification of the
cortical frequency spectrum by changes in CO
2
, blood sugar, and O
2
.
Journal of Neurophysiology, 3, 49–58.
Gravenstein, M. A., Sasse, F., & Hogan, K. ~1992!. Effects of hypocapnia
on canine spinal, subcortical, and cortical somatosensory-evoked po-
tentials during isoflurane anesthesia. Journal of Clinical Monitoring, 8,
126–130.
Grönroos, M., & Pertovaara, A. ~1994!. A selective suppression of human
pain sensitivity by carbon dioxide: central mechanisms implication.
European Journal of Applied Physiology, 68, 74–79.
Harter, R. A. ~1967!. Effects of carbon dioxide on the alpha frequency and
reaction times in humans. Electroencephalography and Clinical Neuro-
physiology, 23, 561–563.
Henning, R. A., Sauter, S. L., Lanphier, E. H., & Reddan, W. G. ~1990!.
Behavioral effects of increased CO
2
load in divers. Undersea Biomed-
ical Research, 17, 109–120.
Hesser, C. M., Fagraeus, L., & Adolfson, J. ~1978!. Roles of nitrogen,
oxygen, and carbon dioxide in compressed-air narcosis. Undersea Bio-
medical Research, 5, 391–400.
Kalkman, C. J., Boezeman, E., Ribberink, A. A., Oosting, J., Deen, L., &
Bovill, J. G. ~1991!. Influence of changes in arterial carbon dioxide
tension on the electroencephalogram and posterior tibial nerve soma-
tosensory cortical evoked potentials during alfentanil0nitrous oxide
anesthesia. Anesthesiology, 75, 68–74.
Lyons, H. A. ~1976!. Centrally acting hormones and respiration. Pharma-
cology Therapy and Behavioral General Systems Pharmacology, 2,
743–751.
Manning, H. L., Shea, S. A., Schwartzstein, R. M., Lansing, R. W., Brown,
R., & Banzett, R. B. ~1992!. Reduced tidal volume increases “air hun-
ger” at fixed PCO
2
in ventilated quadriplegics. Respiration Physiology,
90, 19–30.
Manzey, D., & Lorenz, B. ~1998!. Effects of chronically elevated CO
2
on
mental performance during 26 days of confinement. Aviation, Space,
and Environmental Medicine, 69, 506–514.
Marangoni, A. H., & Hurford, D. P. ~1990!. The effect of varying alveolar
carbon dioxide levels on free recall. Brain and Cognition, 13, 77–85.
Michenfelder, J. D. ~1988!. Anesthesia and the brain: Clinical, functional,
metabolic and vascular disorders ~pp. 51–61!. New York: Churchill
Livingstone.
Phillipson, E. A., & Bowes, G. ~1986!. Control of breathing during sleep.
In N. S. Cherniack & J. G. Widdicombe ~Eds.!, Handbook of physiol-
ogy, section 3. The respiratory system: Vol. II. Control of breathing
~pp. 649–689!. Bethesda, MD: American Physiological Society.
Newhouse, P. A., Belenky, G., Thomas, M., Thorne, D., Sing, H. C., &
Fertig, J. ~1989!. The effects of amphetamine on arousal, cognition, and
mood after prolonged total sleep deprivation. Neuropharmacology, 2,
153–164
Regan, D. ~1989!. Human brain electrophysiology evoked potentials and
evoked magnetic fields in science and medicine. New York: Elsevier.
Remmers, J. E., Brooks, J. G., & Tenney, S. M. ~1968!. Effect of controlled
ventilation on the tolerable limit of hypercapnia. Respiration Physiol-
ogy, 4, 78–90.
Sayers, J. A., Smith, R. E. A., Holland, R. L., & Keatinge, W. R. ~1987!.
Effects of carbon dioxide on mental performance. Journal of Applied
Physiology, 63, 25–30.
Schäfer, K. E. ~1949!. The influence of a long continuous stay in 3% CO2
on the EEG. Pflugers Archives, 251, 726–740.
Sheehy, J. B., Kamon, E., & Kiser, D. ~1982!. Effects of carbon dioxide
inhalation on psychomotor and mental performance during exercise
and recovery. Human Factors, 24, 581–588.
Shea, S. A. ~1997!. Life without ventilatory chemosensitivity. Respiration
Physiology, 110, 199–210.
Shea, S. A., Walter, J., Murphy, K., & Guz, A. ~1987!. Evidence for indi-
viduality of breathing patterns in resting healthy man. Respiration Phys-
iology, 68, 331–344.
Smith, L. J., Greene, S. A., Moore, M. P., & Keegan, R. D. ~1994!. Effects
of altered arterial carbon dioxide tension on quantitative electroenceph-
alography in halothane-anesthetized dogs. American Journal of Veter-
inary Research, 55, 467–471.
Squires, N. K., & Ollo, C. ~1986!. Human evoked-potential techniques:
possible applications to neuropsychology. In J. Hannay ~Ed.!, Experi-
mental techniques in human neuropsychology ~pp. 386–418!. New
York: Oxford University Press.
Squires, N. K., Squires, K. C., & Hillyard, S. A. ~1975!. Two varieties of
long-latency positive wave evoked by unpredictable auditory stimuli
in man. Electroencephalography and Clinical Neurophysiology, 38,
387–401.
Stampi, C., Michimori, A., & Stone, P. ~1995!. A new quantitative method
for assessing sleepiness in the field and in the lab: The Alpha Attenu-
ation Test. Work and Stress, 92, 368–376.
Thorne, D. R., Genser, S. G., Sing, H. C., & Hegge, F. W. ~1985!. The
Walter Reed Performance Assessment Battery. Neurobehavioral Toxi-
cology and Teratology, 7, 415–418.
Van der Worp, H. B., Kraaier, G., Wieneke, G. H., & Van Huffelen, A. C.
~1991!. Quantitative EEG during progressive hypocarbia and hypoxia.
Hyperventilation-induced EEG changes reconsidered. Electroencepha-
lography and Clinical Neurophysiology, 79, 335–341.
Woodbury, D. M., & Karler, R. ~1960!. The role of carbon dioxide in the
nervous system. Anesthesiology, 21, 686–703.
~Received July 6, 1998; Accepted July 1, 1999!
426 E. Bloch-Salisbury, R. Lansing, and S.A. Shea
- CitationsCitations18
- ReferencesReferences39
- "The initial effects of inhaling CO2 at higher concentrations are increased partial pressure of CO2 in arterial blood (PaCO2) and decreased blood pH. However, PaCO2 is tightly regulated in healthy humans through reflex control of breathing, despite normal variation within and between individuals (Bloch-Salisbury et al. 2000). Inhaled CO2 at concentrations of tens of thousands of parts per million has been associated with changes in respiration, cerebral blood flow, cardiac output, and anxiety (Brian 1998; Kaye et al. 2004; Lipsett et al. 1994; Roberge et al. 2010; Woods et al. 1988). "
[Show abstract] [Hide abstract] ABSTRACT: Background: Associations of higher indoor carbon dioxide (CO2) concentrations with impaired work performance, increased health symptoms, and poorer perceived air quality have been attributed to correlation of indoor CO2 with concentrations of other indoor air pollutants that are also influenced by rates of outdoor-air ventilation. Objectives: We assessed direct effects of increased CO2, within the range of indoor concentrations, on decision making. Methods: Twenty-two participants were exposed to CO2 at 600, 1,000, and 2,500 ppm in an office-like chamber, in six groups. Each group was exposed to these conditions in three 2.5-hr sessions, all on 1 day, with exposure order balanced across groups. At 600 ppm, CO2 came from outdoor air and participants’ respiration. Higher concentrations were achieved by injecting ultrapure CO2. Ventilation rate and temperature were constant. Under each condition, participants completed a computer-based test of decision-making performance as well as questionnaires on health symptoms and perceived air quality. Participants and the person administering the decision-making test were blinded to CO2 level. Data were analyzed with analysis of variance models. Results: Relative to 600 ppm, at 1,000 ppm CO2, moderate and statistically significant decrements occurred in six of nine scales of decision-making performance. At 2,500 ppm, large and statistically significant reductions occurred in seven scales of decision-making performance (raw score ratios, 0.06–0.56), but performance on the focused activity scale increased. Conclusions: Direct adverse effects of CO2 on human performance may be economically important and may limit energy-saving reductions in outdoor air ventilation per person in buildings. Confirmation of these findings is needed.- "Aram and Lodge [1987] showed that in-slice preparations of rat cingulate cortex, which increased the levels of CO 2 , caused a pH-mediated reduction in spontaneous EEG. In awake humans, the total power of spontaneous EEG across frequencies from 1 to 64 Hz has been reported to be reduced at moderate levels of hypercapnia [Bloch-Salisbury et al., 2000]. Similarly, in humans under general anesthesia, Kalkman et al. [1991] found significant increases in the alpha (8–12 Hz) and beta (13–30 Hz) bands of the spontaneous EEG after hyercapnic challenge (PaCO 2 $ 50 mm Hg), but not in the delta (1–4 Hz) or theta (5–7 Hz) frequency ranges. "
[Show abstract] [Hide abstract] ABSTRACT: The effects of neural activity on cerebral hemodynamics underlie human brain imaging with functional magnetic resonance imaging and positron emission tomography. However, the threshold and characteristics of the converse effects, wherein the cerebral hemodynamic and metabolic milieu influence neural activity, remain unclear. We tested whether mild hypercapnia (5% CO2 ) decreases the magnetoencephalogram response to auditory pattern recognition and visual semantic tasks. Hypercapnia induced statistically significant decreases in event-related fields without affecting behavioral performance. Decreases were observed in early sensory components in both auditory and visual modalities as well as later cognitive components related to memory and language. Effects were distributed across cortical regions. Decreases were comparable in evoked versus spontaneous spectral power. Hypercapnia is commonly used with hemodynamic models to calibrate the blood oxygenation level-dependent response. Modifying model assumptions to incorporate the current findings produce a modest but measurable decrease in the estimated cerebral metabolic rate for oxygen change with activation. Because under normal conditions, low cerebral pH would arise when bloodflow is unable to keep pace with neuronal activity, the cortical depression observed here may reflect a homeostatic mechanism by which neuronal activity is adjusted to a level that can be sustained by available bloodflow. Animal studies suggest that these effects may be mediated by pH-modulating presynaptic adenosine receptors. Although the data is not clear, comparable changes in cortical pH to those induced here may occur during sleep apnea, sleep, and exercise. If so, these results suggest that such activities may in turn have generalized depressive effects on cortical activity.- [Show abstract] [Hide abstract] ABSTRACT: A major space-borne radar program is on-going in Europe:the Advanced Synthetic Aperture Radar supported by the European Space Agency. ASAR makes use of an active antenna composed of 20 RF Tile Sub-systems. Each Tile embeds 16 C-Band Transmit/Receive Modules, 1 Control Interface Unit, 4 Power Supply Units and 1 Radiating Panel. It is clear that the overall 320 T/R Module is the key element of the radar performance. Emphasis is put on the T/R Module design which is dual HN polarization, 10 W transmit power, 3 dB receive noise figure and use both discrete FETs (P. HEMTs and Power GaAs FETs) and GaAs MMICs. Commands and DC supply switching are controlled with an integrated ASIC. Overall size is 213x38.4x22 mm(3) and weight is 200 gr. In order to succeed the electrical challenge, new miniaturizing technics and technologies were qualified at space level as integrated MMIC functions, new RF substrate and connections and a lot of assembling processes. Due to the high volume production (more than 320 T/R Modules and 8000 hybrids) the design has been put through a very exhaustive industrialization procedure which is detailed below.
People who read this publication also read
Data provided are for informational purposes only. Although carefully collected, accuracy cannot be guaranteed. Publisher conditions are provided by RoMEO. Differing provisions from the publisher's actual policy or licence agreement may be applicable.
This publication is from a journal that may support self archiving.
Learn more