Regional differences in cortical electroencephalogram (EEG) slow wave activity and interhemispheric EEG asymmetry in the fur seal

Article (PDF Available)inJournal of Sleep Research 21(6) · June 2012with47 Reads
DOI: 10.1111/j.1365-2869.2012.01023.x · Source: PubMed
Abstract
Slow wave sleep (SWS) in the northern fur seal (Callorhinus ursinus) is characterized by a highly expressed interhemispheric electroencephalogram (EEG) asymmetry, called 'unihemispheric' or 'asymmetrical' SWS. The aim of this study was to examine the regional differences in slow wave activity (SWA; power in the range of 1.2-4.0 Hz) within one hemisphere and differences in the degree of interhemispheric EEG asymmetry within this species. Three seals were implanted with 10 EEG electrodes, positioned bilaterally (five in each hemisphere) over the frontal, occipital and parietal cortex. The expression of interhemispheric SWA asymmetry between symmetrical monopolar recordings was estimated based on the asymmetry index [AI = (L-R)/(L+R), where L and R are the power in the left and right hemispheres, respectively]. Our findings indicate an anterior-posterior gradient in SWA during asymmetrical SWS in fur seals, which is opposite to that described for other mammals, including humans, with a larger SWA recorded in the parietal and occipital cortex. Interhemispheric EEG asymmetry in fur seals was recorded across the entire dorsal cerebral cortex, including sensory (visual and somatosensory), motor and associative (parietal or suprasylvian) cortical areas. The expression of asymmetry was greatest in occipital-lateral and parietal derivations and smallest in frontal-medial derivations. Regardless of regional differences in SWA, the majority (90%) of SWS episodes with interhemispheric EEG asymmetry meet the criteria for 'unihemispheric SWS' (one hemisphere is asleep while the other is awake). The remaining episodes can be described as episodes of bilateral SWS with a local activation in one cerebral hemisphere.
Regional differences in cortical electroencephalogram (EEG)
slow wave activity and interhemispheric EEG asymmetry in the
fur seal
OLEG I. LYAMIN
1,2,3
, IVETTA F. PAVLOVA
2
, PETER O. KOSENKO
2,4
,
LEV M. MUKHAMETOV
2,3
and JEROME M. SIEGEL
1
1
UCLA & VA GLAHS, Sepulveda, CA, USA,
2
Utrish Dolphinarium Ltd, Moscow, Russia,
3
Severtsov Institute of Ecology and Evolution, Moscow,
Russia and
4
South Federal University, Rostov-on-Don, Russia
Keywords
interhemispheric EEG asymmetry, local sleep,
slow wave activity, unihemispheric sleep,
northern fur seal
Correspondence
Dr Oleg Lyamin, PhD, Center for Sleep
Research, VA GLAHS Sepulveda, Plummer
Street 16111, North Hills, CA 91343, USA.
Tel.: +1-818-8917711 ext. 7380;
fax: +1-8-895-9575;
e-mail: olyamin@ucla.edu
Accepted in revised form 22 April 2012; received
15 December 2011
DOI: 10.1111/j.1365-2869.2012.01023.x
SUMMARY
Slow wave sleep (SWS) in the northern fur seal (Callorhinus ursinus)is
characterized by a highly expressed interhemispheric electroencephalo-
gram (EEG) asymmetry, called ÔunihemisphericÕ or ÔasymmetricalÕ SWS.
The aim of this study was to examine the regional differences in slow
wave activity (SWA; power in the range of 1.2–4.0 Hz) within one
hemisphere and differences in the degree of interhemispheric EEG
asymmetry within this species. Thr ee seals were implanted with 10 EEG
electrodes, positioned bilaterally (five in each hemisphere) over the
frontal, occipital and parietal cortex. The expression of interhemispheric
SWA asymmetry between symmetrical monopolar recordings was
estimated based on the asymmetry index [AI = (L)R) (L+R), where L
and R are the power in the left and right hemispheres, respectively]. Our
findings indicate an anterior–posterior gradient in SWA during asymmet-
rical SWS in fur seals, which is opposite to that described for other
mammals, including humans, with a larger SWA recorded in the parietal
and occipital cortex. Interhemispheric EEG asymmetry in fur se als was
recorded across the entire dorsal cerebral cortex, including sensory
(visual and somatosensory), motor and associative (parietal or s upra-
sylvian) cortical areas. The expression of asymmetry was greatest in
occipital–lateral and parietal derivations and smallest in frontal–medial
derivations. Regardless of regional differences in SWA, the majority
(90%) of SWS episodes with interhemispheric EEG asymmetry meet the
criteria for Ôunihemispheric SWSÕ (one hemisphere is asleep while the
other is awake). The remaining episodes can be described as episodes
of bilateral SWS with a local activation in one cerebral hemisphere.
INTRODUCTION
Electroencephalogram (EEG) slow wave activity (SWA) is
considered to be a measure of sleep pressure and intensity
(Borbe
´
ly, 1994). EEG power in the low-frequency range
(<4 Hz) shows a frontal–medial predominance in human
non-rapid eye movement (NREM) sleep [termed slow wave
sleep (SWS) in animals], which is enhanced by sleep
deprivation (Marzano et al., 2010; Werth et al., 1997).
Frequency-specific EEG differences have been also
described in rodents with a larger EEG power in the frontal
derivations compared to the occipital derivations. This
gradient is enhanced further during rebound following sleep
deprivation (mouse: Huber et al., 2000; rat: Schwierin et al.,
1999; Vyazovskiy et al., 2002). The anterior predominance of
EEG slow waves during high sleep demands may reflect the
increased vulnerability of frontal cortex functions to sleep loss
(Horne, 1993).
SWS in dolphins and fur seals is characterized by a striking
EEG asymmetry, called Ôunihemispheric SWSÕ (USWS) or
Ôasymmetrical SWSÕ (Lyamin et al., 2008a,b,c; Mukhametov
et al., 1977). Interhemispheric EEG asymmetry has also
been recorded in some avian species (Fuchs et al., 2009;
Rattenborg et al., 2000), but the expression of the asymmetry
is smaller and the episodes are shorter than in marine
mammals. To date, almost all EEG recordings in dolphins
J. Sleep Res. (2012)
Regular Research Paper
ª 2012 European Sleep Research Society 1
and seals have been performed using a bipolar technique,
with both active electrodes positioned above the cortex
(usually in frontal and occipital regions) and only occasionally
employing a monopolar technique, with a reference electrode
positioned in the nasal bones (Lyamin et al., 2008a,b,c;
Mukhametov et al., 1997). The topographic aspect of EEG
asymmetry has never been examined in detail in any marine
mammal. The fur seal is a unique animal, showing all known
types of sleep (bilateral and asymmetrical SWS as well as
REM sleep) and with the degree of slow wave EEG
asymmetry comparable to that of dolphins. Therefore, the
aim of this study was to examine the regional differences (1)
in SWA (power in the range of 1.2–4.0 Hz) within one
cerebral hemisphere and (2) in the expression of SWA
asymmetry in the fur seal.
MATERIALS AND METHODS
Subjects
Data were collected from three captive male fur seals
(Callorhinus ursinus; 22–25 kg, aged 2–3 years) at the Utrish
Marine Station of the Severtsov Institute of Ecology and
Evolution (Black Sea, Russia). The study was approved by
the Severtsov Institute Committee for Bioethics.
Design and procedure
Surgery and EEG electrode localization
Under general anaesthesia, fur seals were implanted with five
pairs of stainless steel screws (1 mm in diameter) for EEG
recording in symmetrical frontal, parietal and occipital cortical
areas of the right and left hemispheres (
Fig. 1). All EEG
electrodes were epidural. A reference electrode was implanted
into the nasal bone. The electrodes of the frontal group (a total
of 12 electrodes for the three seals combined) were located in
the vicinity of the motor and somatosensory cerebral cortex
(Supin et al., 2001). Relative to S. cruciatus and S. postcru-
ciatus, they could be divided into two subgroups (medial and
lateral; labelled fm and , respectively, in Fig. 1). The occipital
group of electrodes (10 electrodes for the three seals
combined) was located above the visual cortex and could be
divided into two subgroups: medial (four electrodes in seals 1
and 3, located 6–10 mm lateral to the midline) and lateral (six
electrodes in all seals, located along the posterior section of
S. lateralis; labelled om and ol, respectively, in Fig. 1). The
parietal group of electrodes was located in the vicinity of
S. suprasylvian in the area bordering the visual, auditory and
somatosensory cortices (labelled p in Fig. 1). Surgical proce-
dures and post-surgical animal care have been described in
detail in our previous publications (e.g. Lyamin et al., 2008a,b).
Polygraphic recording
Recordings started at 17:00 and finished after 09:00 hours.
The seals were connected to a recording cable and placed
into a 0.8 · 0.8 · 1.0 m experimental chamber positioned in
a sound-attenuated and light-controlled empty indoor pool.
All animals were adapted to the experimental conditions.
They were fed fish at 18:00 hours and sprayed with water for
15 min after feeding and then left undisturbed for the
remainder of the night. Lights were turned off between
20:00 and 08:00 hours. During this period the seals were
monitored via infrared cameras. The room temperature
during recording was maintained at 18–20 C.
EEG from all derivations was recorded relative to a
reference electrode using a conventional cable technique
and a multi-channel polygraph (Medicom 19 26, Taganrog,
Russia; bandwidth 0.3–30.0 Hz) and stored on a hard disk
(sampling rate 250 Hz). Neck electromyography (EMG) was
also recorded simultaneously. The data were analysed offline
using Spike 2 software (Cambridge Medical Design, UK).
Data analysis
For each seal the data collected during the third recording
night were used for the analysis. EEG slow wave power
(SWA) was computed in the frequency range of 1.2–4.0 Hz in
5-s epochs. In two seals (1 and 2) SWA was computed in five
symmetrical derivations of the left and right hemispheres. In
seal 3, due to contamination of the EEG with artefacts in the
left occipital–medial derivation, SWA was computed in four
symmetrical recordings. Epochs with artefacts were excluded
from the analysis. SWA was averaged whenever at least
three of the four consecutive 5-s epochs were artefact-free.
The asymmetry index (AI) of SWA (a measure of the
Figure 1. Localization of EEG electrodes relative to the first-order
sulci (SL, s. lateralis; SSa, s. suprasylvian anterior; SSp, s. supra-
sylvian posterior; SPCr, s. postcruciatus, SC, s. cruciatus) and
bregma (B) in fur seals (s1, s2 and s3) as indicated by color markers.
VC, AC, SSC, MC visual, auditory, somatosensory and motor
cortex, respectively. Circled areas mark the frontal (fm and
frontal-medial and lateral, respectively), parietal (p) and occipital (om
and ol occipital-medial and lateral, respectively) groups of EEG
electrodes (based on Supin et al. 2001). R position of the reference
electrodes relative to the seal brain.
2 O. I. Lyamin et al.
ª 2012 European Sleep Research Society
expression of interhemispheric EEG asymmetry) was eval-
uated in each monopolar derivation for each 20-s epoch, as
follows: AI = (L)R) (L+R), where L and R are standardized
spectral power in the left and right hemispheres, respectively
(described in Lyamin et al., 2008a,b).
During our previous studies we noted that SWA activity in
fur seals in occipital–lateral derivations was usually greater
than in frontal derivations. Therefore, we first scored SWS
based on the EEG recorded from symmetrical occipital–
lateral derivations. To allow between-derivation and between-
animal comparisons, SWA in each 20-s epoch was stan-
dardized to the average SWA in the same derivation during
high-voltage bilateral SWS (BSWS). Such an episode was
selected as a 4–5-min period (or three periods in seal 2;
Table 1) of SWS, with the greatest average SWA in both
hemispheres and minimal SWA asymmetry (based on the
AI). Each of these SWS episodes was followed by REM
sleep. For comparison, in quiet wakefulness SWA varied
between 4 and 14% of that in BSWS (on average 6–12% in
different deviations and seals when calculated for 1-min
episodes of uninterrupted quiet waking).
All episodes of SWS were scored visually and subdivided
into left asymmetrical SWS (LASWS; SWA in the left
hemisphere is greater than in the right hemisphere), right
asymmetrical SWS (RASWS; SWA in the right hemisphere is
greater than in the left hemisphere) or BSWS (SWA in the
right and left hemispheres are comparable) based on the
criteria described in previous publications (Lapierre et al.,
2007; Lyamin et al., 2008a,b,c). Only episodes of RASWS
and LASWS with an average absolute AI > 0.3 were selected
for further analysis. The beginning of episodes of LASWS and
RASWS was determined as the time of the first 20-s epoch in
which SWA exceeded 10% of that seen in BSWS in either
hemisphere. The end of SWS episodes was easily marked by
behavioural awakening or the transition into REM sleep.
Differences in AI between symmetrical derivations and in
SWA within one-hemisphere derivations were tested with
one-way repeated-measures analysis of variance (anova)
followed by a paired TukeyÕs test. All other comparisons were
conducted using either a t-test or Mann–Whitney U-test with
SigmaPlot software.
RESULTS
Characteristics of episodes of asymmetrical SWS
Seals slept while lying on their sides or bellies. During BSWS
both eyes were usually closed. During asymmetrical SWS,
one eye could open briefly and this eye was always
contralateral to the hemisphere with a low-voltage SWA, as
reported in our previous studies (Lyamin et al., 2004).
However, these episodes were rare under conditions of
complete darkness.
Eight episodes of RASWS and four episodes of LASWS
were available in seal 1, five episodes of RASWS and five
episodes of LASWS in seal 2, and six episodes of RASWS
and only one episode of LASWS in seal 3 (Table 1). All
episodes were characterized by a substantial difference in
SWA between the left and right hemispheres in occipital–
lateral derivations (on average greater than a threefold
difference; t-test; P < 0.01 for all comparisons). In two seals
(1 and 2) the average SWA in the hemisphere with a greater
SWA (the right hemisphere during RASWS and left during
LASWS) was smaller than in the same hemisphere during
BSWS. In seal 3, SWA activity in the right hemisphere during
five of six episodes of RASWS was greater than in the same
hemisphere in BSWS. A representative example of an
episode of LASWS, RASWS and BSWS is shown in
Fig.2.
The AI of SWA ranged between )0.07 and +0.02 in
episodes of BSWS, between )0.31 and )0.76 in episodes of
RASWS and between +0.37 and +0.83 in episodes of
LASWS (Table 1). Neither duration of RASWS and LASWS
episodes nor the average SWA in the hemisphere with a
greater power (right hemisphere during RASWS and left
Table 1 Characteristics of episodes of slow wave sleep (SWS) in the fur seal based on the electroencephalogram (EEG) recorded in
occipital–lateral derivations
Seal Episodes SWS episode duration (min)
Slow wave activity (SWA; % in BSWS)
Asymmetry index (AI) of SWASWA-L SWA-R
1 RASWS (n = 8) 11.4 ± 2.6 (4.8, 24.8) 16 ± 2 (11, 23) 85 ± 3 (73, 100) )0.68 ± 0.02 ()0.76, )0.57)
LASWS (n = 4) 9.5 ± 3.0 (5.3, 18.5) 91 ± 10 (64, 109) 21 ± 4 (11, 27) +0.60 ± 0.02 (+0.54, +0.65)
BSWS (n = 1) 4.7 100 100 )0.01
2 RASWS (n = 5) 4.0 ± 0.6 (2.4, 5.4) 22 ± 5 (11, 41) 63 ± 7 (47, 83) )0.49 ± 0.04 ()0.60, )0.41)
LASWS (n = 5) 7.1 ± 1.3 (2.3, 10.7) 82 ± 9 (56, 100) 23 ± 9 (5, 43) +0.54 ± 0.10 (+0.31, +0.83)
BSWS* (n = 3) 4.0 ± 0.3 (3.7, 10.7) 100 ± 1 (100, 101) 100 ± 0 0.00 ± 0.01 ()0.02, +0.02)
3 RASWS (n = 6) 5.4 ± 1.5 (2.0, 12.0) 37 ± 7 (13, 60) 133 ± 15 (75, 175) )0.54 ± 0.07 ()0.74, )0.31)
LASWS (n = 1) 6.3 90 49 +0.32
BSWS (n = 1) 4.0 100 100 )0.07
RASWS, right asymmetrical SWS; LASWS, left asymmetrical SWS; BSWS, high-voltage bilateral SWS.
*Average for three episodes of BSWS. The data are presented as mean ± standard error of the mean with minimal and maximal values given
in parentheses.
Sleep in the fur seal 3
ª 2012 European Sleep Research Society
during LASWS) differed in seals 1 and 2, which displayed
asymmetrical SWS in both hemispheres (t-test, P > 0.05 for
all comparisons). Absolute AI of SWA in the two hemispheres
were within the same range both in RASWS and LASWS
(0.31–0.76 and 0.37–0.80, respectively) and the average
values did not differ (t-test; P > 0.05 for all comparisons).
To summarize, these data indicate that (1) episodes of
asymmetrical SWS in the fur seal selected based, on the
average AI, are characterized by a substantial difference in
the SWA between occipital–lateral derivations of the two
hemispheres; and (2) there is no evidence for predominance
of the right or left hemispheres in the expression of SWA in
these derivations in the fur seal.
Regional differences in SWA during asymmetrical SWS
One way-repeated-measures anova revealed consistent
regional differences in SWA during ASWS in the hemisphere
with greater power in all three fur seals (seal 1: F
(7,28)
= 4.82,
P = 0.004 for RASWS and F
(3,12)
= 5.56, P = 0.018 for
LASWS; seal 2: F
(4,16)
= 5.25, P = 0.007 for RASWS and
F
(4,16)
= 3.75, P = 0.025 for LASWS; seal 3: F
(5,15)
= 6.79,
P = 0.04 for RASWS). In two seals (1 and 2), SWA in the
frontal–medial derivations was the smallest (36–65% of SWA
in BSWS in the corresponding derivations) and SWA in the
parietal derivations was the greatest (67–100% of SWA in
BSWS; post-hoc TukeyÕs test, P < 0.05; in both seals for
RASWS and in seal 1 for LASWS;
Fig. 3, SWA1). SWA in
the occipital–lateral derivations (63–91% of SWA in BSWS)
did not differ from the high level of SWA in the parietal
derivations (67–100%). At the same time, SWA in the
occipital–medial and frontal–lateral derivations were
comparable except for one case (LASWS in seal 1). In seal
3, only RASWS was recorded. As in the other two animals,
SWA in seal 3 was the smallest in the frontal–medial
derivations (on average 61% of SWS in BSWS) and the
greatest in the occipital–lateral derivations (on average
133%; P < 0.05). SWA in the frontal–lateral derivations was
greater than in frontal–medial derivations (113 and 61% of
SWA in BSWS, respectively; P < 0.05). Therefore, these
data indicate clearly an anterior–posterior gradient in SWA in
the hemisphere with an overall greater power during
asymmetrical SWS with a larger SWA recorded in the
parietal and occipital cortex.
We found no consistent regional differences between SWA
in the hemisphere with a low-voltage EEG during asymmet-
rical SWS in fur seals. SWA in the hemisphere with a low-
voltage EEG differed between derivations in two of three
seals (Fig. 3, SWA2). In seal 1, SWA in posterior (parietal
and occipital–medial in LASWS and parietal in RASWS)
derivations was greater than in the frontal derivations (anova,
F
(7,28)
= 20.89, P < 0.001 in RASWS and F
(3,12)
= 35.36,
P < 0.001 in LASWS; post-hoc TukeyÕs test, P < 0.05).
Conversely, in seal 3 the power in the parietal derivations
was the smallest compared to occipital–lateral and frontal–
medial derivations (F
(3,15)
= 5.39, P = 0.01; post-hoc TukeyÕs
test, P < 0.05).
Regional differences in interhemispheric asymmetry of
SWA
One-way repeated-measures anova revealed regional differ-
ences in the degree of interhemispheric asymmetry of SWA
during asymmetrical SWS in all three seals (seal 1:
F
(7,28)
= 42.69, P < 0.001 in RASWS and F
(3,12)
= 9.41,
P = 0.001 in LASWS; seal 2: F
(4,16)
= 11.84, P < 0.001 in
Figure 2. Cortical EEG of the left and right hemispheres and neck electromyogram (EMG) during SWS in a fur seal. LASWS and RASWS
episodes of left (L) and right (R) asymmetrical SWS. BSWS an episode of bilateral SWS. Fl-Ol bipolar recording from frontal-lateral and
occipital-lateral derivations. Ol, Om, P, Fl and Fm mark monopolar recording from occipital-lateral, occipital-medial, parietal, frontal-lateral and
frontal-medial derivations, respectively.
4 O. I. Lyamin et al.
ª 2012 European Sleep Research Society
RASWS and F
(4,16)
= 7.15, P = 0.002 in LASWS; seal 3:
F
(5,15)
= 10.39, P < 0.001 in RASWS). The average absolute
AI in the parietal (which ranged between 0.38 and 0.66 in
different episodes) and occipital (0.49–0.68) derivations
was greater than in frontal–medial derivations (0.06–0.50)
in all seals (Fig. 3, AI). The difference was found to be
significant for all comparisons except for LASWS in seal 1
(post-hoc TukeyÕs test, P < 0.05). In two seals (1 and 2) with
electrodes in the occipital–medial position, the absolute AI in
these derivations was smaller (0.33–0.46) than that in
occipital–lateral (0.49–0.68; P < 0.05 in both seals in RAS-
WS and in seal 1 in LASWS) and parietal (0.38–0.66;
P < 0.05 in both seals in RASWS) derivations. In all three
seals the AI in frontal–lateral derivations (0.33–0.70) was
greater than in frontal–medial derivations (0.06–0.58;
P < 0.05 in seal 2 in LASWS and seal 3 in RASWS). At the
same time, the difference between AI in the frontal–lateral
and other derivations was inconsistent.
Therefore, the expression of interhemispheric SWA asym-
metry in the fur seal was greatest in the occipital–lateral and
parietal derivations and smallest in the frontal–medial deri-
vations, suggesting a posterio–anterior gradient for the
degree of SWA asymmetry in the dorsal cortex of the fur
seal. The degree of lateralization of SWA in medial deriva-
tions was also smaller than in the bordering, more lateral
derivations, suggesting a medial–lateral gradient.
Characteristics of ASWS episodes
In this study we recorded three main types of asymmetrical
SWS episodes in the fur seal based on the regional
differences of SWA and the degree of asymmetry in two
cerebral hemispheres (
Fig. 4). These episodes accounted for
25 of 28 episodes (90% of all episodes) recorded in the three
seals as follows:
1. In nine of 28 episodes (32% of all episodes) the average
SWA in all derivations of the hemisphere with a greater power
exceeded by 70% that in the corresponding derivations in
BSWS, apparently indicating that the dorsal cortex of this
hemisphere was in a state of high-voltage (ÔdeepÕ) SWS. The
average SWA in all areas of the hemisphere with lower power
was, on average, less than 25% of that in the corresponding
derivations in BSWS, except for occipital–medial derivations,
indicating that this portion of the hemisphere is in a state of
transition between waking and SWS. The average absolute
AI in all derivations was greater than 0.40 (i.e. exceeding the
Ô0.3 criteriaÕ set for asymmetrical SWS) except for the
occipital–medial derivations (which, on average, is 0.35). All
these episodes were recorded in seal 1, which was the most
ÔasymmetricalÕ animal. These episodes meet the criteria of
USWS (where one hemisphere is awake and the other is
asleep) with a similar intensity of sleep in all areas of the
ÔsleepingÕ hemisphere.
Figure 3. Power and asymmetry index of slow wave activity (SWA, 1.2-4.0 hz) during episodes of asymmetrical SWS in 3 fur seals. SWA 1 and
SWA 2 average power + SER in the hemisphere with a higher voltage and low voltage EEG activity, respectively, shown as percent of SWA
power in the corresponding derivations in bilateral SWS (described in Methods). L and R left and right hemisphere, respectively. AI
asymmetry index (+ SER). Fm, Fl, P, Om and Ol mark monopolar recording from frontal-lateral, frontal-medial, parietal, occipital-medial and
occipital-lateral derivations, respectively. P < 0.05 - Tukey paired test following one way repeated measures anova.
Sleep in the fur seal 5
ª 2012 European Sleep Research Society
2. In 12 of 28 episodes (43%; these episodes were recorded
in all three seals) the average SWA in the parietal and
occipital areas of the hemisphere with a greater power
exceeded 60% of that in BSWS, indicating that the posterior
area of this hemisphere was in a state of high-voltage (ÔdeepÕ)
SWS. At the same time SWA in the frontal cortex (only in
medial or in both medial and lateral areas) did not exceed on
average 30% of that in BSWS, thus indicating a low-voltage
(ÔlightÕ) SWS in the anterior cortical area. The average SWA
in the opposite hemisphere was smaller than 20% of that in
BSWS, indicating waking or low-voltage SWS. The average
AI in the frontal derivations was a little smaller than that in the
parietal and occipital-lateral derivations, but in all dorsal
cortical derivations met the criteria of asymmetrical SWS.
These episodes also meet the definition of USWS with
regional differences of SWA in the ÔsleepingÕ hemisphere.
3. In four of 28 episodes (14%; all were recorded in seal 3)
the average SWA in one hemisphere exceeded by 60% of
that during BSWS except for the medial–frontal derivations in
which it was slightly smaller (50%), indicating that one
hemisphere was in a state of ÔdeepÕ SWS. The average SWA
in the other hemisphere was low in the parietal and occipital
areas (25–30% of that in BSWS, indicative of a state
transitional between waking and sleep). At the same time,
SWA in the frontal area was very similar to that in the
hemisphere with greater power, indicating that both frontal
areas were in a state of high-voltage (ÔdeepÕ) SWS. The
average AI was greater (meeting the criteria of asymmetrical
SWS) in the parietal and occipital derivations, while in the
frontal areas it was smaller than 0.3 (indicating BSWS).
Therefore, this group of episodes is better described as (1)
episodes with BSWS in the frontal cortex and asymmetrical
SWS in the occipital–parietal area or (2) BSWS with unilateral
activation of the occipital–parietal cortical area.
DISCUSSION
In our previous study we showed that AI can be used to
measure quantitatively the overall expression of EEG
asymmetry in the range of 1.2–16 Hz. The AI defines
substantial differences in the degree and frequency range
of the interhemispheric EEG asymmetry between bilaterally
(a) (b) (c) (d)
Figure 4. EEG spectral power and asymmetry index (AI) in symmetrical monopolar derivations during SWS in a fur seal. The first three row
diagrams are representative examples showing normalized slow wave activity (SWA) power in the right and left hemispheres (R and L) and the
AI of SWA calculated for consecutive 20-sec epochs in episodes of asymmetrical SWS (a,b, and d) and in an episode of BSWS (c). The three
graphs in the bottom row show average SWA power and the average AI SER) during 3 main types (1–3) of episodes of asymmetrical SWS
(described in Results). Black and gray bars SWA in the hemisphere with higher voltage and lower voltage EEG, respectively. The lines
connect circles marking values of the average AI for the corresponding deviations. Fm, Fl, P, Om and Ol - monopolar recording from frontal-
medial, frontal-lateral, parietal, occipital- medial and occipital-lateral derivations, respectively. Episode a, b and d (first three rows of diagrams)
are representative examples of type 1, 2 and 3 episodes, respectively. Note that in episode ÔaÕ SWA in all derivations of the left hemisphere was
substantially greater in the right hemisphere. In episode ÔbÕ and ÔdÕ SWA in the occipital and parietal derivations of one hemisphere (right in ÔbÕ
and left in ÔdÕ) was substantially greater than in the corresponding derivations of the opposite hemisphere. Both in episodes ÔbÕ and ÔdÕ the
difference in SWA between the frontal derivations in two hemispheres was minimal. However, in episode ‘‘b’’ SWA in the frontal area was low
(transitional between quiet waking and SWS) while in episode ‘‘d’’ SWA in the symmetrical frontal areas was large, comparable to that during
high voltage BSWS.
6 O. I. Lyamin et al.
ª 2012 European Sleep Research Society
sleeping rats and unihemispherically sleeping dolphins, as
well as in fur seals and some birds (Lyamin et al., 2008a).
Much smaller state-related interhemispheric EEG asymme-
tries have been reported for humans and some animal
species. Use-dependent local changes and hemispheric
dominance in the expression of different behaviours and
major EEG rhythms may underlie these regional differences
in SWA both within and between hemispheres producing
the EEG asymmetry (e.g. Goldstein et al., 1972; Kattler
et al., 1994; Krueger et al., 1999; Vyazovskiy et al. , 2000,
2002).
The major finding of this study is that interhemispheric
SWA in the fur seal is recorded across the entire dorsal
cerebral cortex. The occipital (medial and lateral) electrodes
in this study were located in the visual cortex. The frontal–
lateral electrodes in all seals were located in the vicinity of the
somatosensory cortex (Supin et al., 2001). The parietal
electrodes were located in the vicinity of S. suprasylvian,
which in mammals is the parietal associative cortex. The
activity of neurones in this area has never been tested in the
fur seal, as was performed for the visual and somatosensory
cortex (Supin et al., 2001). However, the functional charac-
teristics of neurones in this area (cortical fields 5 and 7) have
been examined extensively in the cat and in primates. It is
known that this part of cerebral cortex contributes primarily to
the programming and executions of visually guided move-
ments (e.g. Drew et al., 2008). Finally, the medial–frontal
electrodes were located dorsal to S. postcruciatus, which in
the fur seal is the motor cortex (Supin et al., 2001). Therefore,
the interhemispheric asymmetry of SWA in the fur seal is a
feature of sensory (visual and somatosensory), motor and
associative (parietal or suprasylvian) cortical areas.
It is known that interhemispheric EEG asymmetry corre-
lates with asymmetrical eye state in fur seals and cetaceans
(Lyamin et al., 2004, 2008a,b,c). The duration of episodes of
asymmetrical SWS can be extended and the degree of
asymmetry can be enhanced by auditory stimuli (Lyamin
et al., 2008b), so visual stimulation is not the only type of
sensory stimulus which can cause unilateral cortical activa-
tion during SWS in the fur seal. We hypothesized that
unilateral cortical activation during SWS (manifested in the
EEG asymmetry) in the fur seal allows sensory processing
and continuous vigilance (Lyamin et al., 2004). The data
collected in this study indicate that SWA asymmetry in fur
seals is expressed highly in the visual and parietal cortical
areas and provides additional support for this hypothesis,
specifically for the involvement of cortical areas of one brain
hemisphere in the processing of the visual information during
USWS in marine mammals.
Fur seals sleep in water floating on their sides, while
holding the head above the water and paddling with one front
flipper. When in this posture the vibrissae from only one side
(ipsilateral to the paddling flipper) are directed to the water. It
is reasonable to expect that both ipsilateral vibrissae and
flipper movement supply the brain with ascending afferent
information crucial for controlling the position of the head and
adjusting the movement of the flipper. It is also known that
the projection area of vibrissae to the contralateral somato-
sensory cortex in the fur seal is magnified disproportionally
compared to other parts of the body (Supin et al., 2001).
Previous bipolar recordings from the frontal–occipital cortical
areas showed that the EEG in the fur seal is more
desynchronized (activated) in the hemisphere that is contra-
lateral to the paddling flipper (and the vibrissae directed to the
water) when compared to the EEG in the ipsilateral hemi-
sphere, which displays a higher-voltage SWA. Based on
these data, we hypothesized that the unilateral cortical
activation during SWS in the fur seal also allows motion
and sleep while in water (Lyamin et al., 2008a,b,c). However,
the degree of functional lateralization of the somatosensory
and motor cortex has not been evaluated until this study. The
data collected via monopolar recording revealed that SWA
asymmetry in the fur seal is also expressed in the somato-
sensory cortex. The smallest lateralization of SWA was found
in the motor cortex. This makes sense, because the animals
were lying on the chamber floor without apparent movement.
However, the meaning of functional lateralization of EEG in
the somatosensory cortex (indicating unilateral activation)
while the seal sleeps on the ground remains unclear. It could
be due to generalized unilateral cortical activation spread
from other areas, e.g. from the visual cortex, or whole
hemisphere activation (Nir et al., 2011). It is interesting that
SWA asymmetry in the occipital–medial cortex was smaller
than in the lateral occipital cortex, tentatively suggesting a
smaller functional lateralization of the primary visual cortex
compared to the secondary visual cortex during SWA in the
fur seal.
It remains to be determined whether the EEG asymmetry is
present in other cortical regions (e.g. medial, temporal and
basal) and or subcortical areas. In dolphins EEG was
recorded opportunistically simultaneously from the dorsal
thalamic areas and two cerebral hemispheres (Mukhametov
et al., 1997). In all cases the EEG changes in the ipsilateral
cortex and thalamus were synchronous. Based on these
observations, it was hypothesized that USWS in dolphins is
not only a cortical phenomenon, but appears to involve at a
minimum the entire cortical–thalamic system (Lyamin et al.,
2008c). The asymmetry in the range of 12–16 Hz (spindles)
and lateralized release of acetylcholine in the cortex during
ASWS in the fur seal also suggest a functional asymmetry
between thalamic and basal forebrain (cholinergic) areas,
which promotes wakefulness and sleep and sends ascending
projections to the cortex (Lapierre et al., 2007; Lyamin et al.,
2008a).
Our data show that the majority (90%) of SWS episodes in
the fur seal can be described as USWS. They meet the
definition of USWS because, regardless of regional differ-
ences in the SWA power, the SWA in all derivations of one
(ÔsleepingÕ) hemisphere is greater than that in the other
(ÔwakingÕ) hemisphere. The third group of sleep episodes
concerns episodes with BSWS in the frontal cortex and
asymmetrical SWS in the occipital and parietal areas. These
Sleep in the fur seal 7
ª 2012 European Sleep Research Society
episodes should, rather, be described as episodes of BSWS
with local activation (or lower-voltage sleep) in one cerebral
hemisphere. In all these episodes local activation was
recorded in the occipital–parietal hemisphere, the area
involved in the processing of visual information. Behaviour-
ally, these episodes may be episodes of sleep while
motionless on the floor and briefly opening one eye.
The second finding of this study is the character of regional
differences (topography) of SWA in the fur seal. We found
that SWA in the fur seal in the occipital and parietal
derivations was usually greater than in the frontal deviations.
A frontal predominance of SWA is a characteristic feature of
sleep in humans (Marzano et al., 2010; Werth et al., 1997). In
the rat, EEG spectral power is usually measured in the frontal
derivations, located in the vicinity of the motor and prefrontal
cortex, and in occipital (parietal or posterior) sites, located in the
vicinity of the barrel (somatosensory) cortex. SWA in the frontal
derivations was larger than in the posterior derivations,
particularly during recovery after sleep deprivation (e.g.
Schwierin et al., 1999; Vyazovskiy et al., 2002). We are not
aware of any detailed studies on the regional differences of
EEG activity in the rat beyond the area of the motor and
somatosensory cortex. However, based on visual inspection
the EEG activity in the rat occipital area was reported to be
smaller than that of the frontal area (Timo-Iaria et al., 1970). In
the cat, spindles in the sensorimotor cortex are accompanied
by high amplitude slow waves in the visual cortex (Lucas and
Sterman, 1974), so the power of SWA in the occipital deriva-
tions is obviously greater than that in the frontal derivations.
This would be in agreement with data collected in the fur seal.
There is a possibility that some factors could have an effect
on the inverse gradient in SWA in the fur seal. Due to the
frontal (nasal) position of a reference electrode and differen-
tial amplification of active screw electrodes referenced to it
might be picking up some EEG activity, and this would be
reducing SWA in anterior derivations to a greater degree than
in posterior derivations. However, it seems not to be the key
factor determining the inverse gradient of SWA in the fur seal,
because SWA in the parietal derivation (more anterior) did
not differ from SWA in the occipital–lateral (more posterior)
derivations and SWA in the frontal–lateral derivation (more
anterior) could be greater than in the occipital–medial (more
posterior) derivation (Fig. 3). There is evidence on the
highest synaptic density in the human frontal cortex (e.g.
Ringli and Huber, 2011). No such studies have ever been
conducted in marine mammals, as well as studies on the
thickness of cortex in different cortical regions. The EEG
electrodes were placed at the bottom of holes drilled through
the skull, so skull thickness could not be responsible for the
regional variations we see. Examining the effects of sleep
deprivation on regional differences of SWA in the fur seal
would also help understanding of the difference between
terrestrial mammals and seals.
It has been shown that slow waves may occur locally in
different cortical and subcortical areas in humans, monkeys
and rats, as indicated by EEG slow waves and or the
pattern of neuronal discharge (Krueger et al. , 1999; Nir
et al., 2011; Pigarev et al., 1997; Vyazovskiy et al., 2011).
According to this paradigm, sleep is viewed as a distrib-
uted process in the brain in which parts of the brain can be
asleep while other parts are awake (e.g. Krueger and
Tononi, 2011). Recent human and rodent studies (Nir
et al., 2011; Vyazovskiy et al., 2011) have detailed an
asynchrony in occurrence of the corresponding EEG slow
waves (an indication of local sleep) in the cortex and
subcortical areas. These data indicate that local sleep in
humans and rodents is an issue of synchrony in the
appearance of single slow waves or a characteristic pattern
of neuronal discharge (over a time-frame of a few seconds
or several slow wave events) in the corresponding sleeping
cortical areas, while all these areas are displaying periods
of SWA (over a time-frame of minutes or a SWS episode).
In other words, even if cortical slow waves are generated
locally by multiple generators (Murphy et al., 2009), all
generators in the two brain hemispheres of humans and
rodents are still in the same functional state—going ÔONÕ
and ÔOFFÕ over the time-frame of seconds. If we use the
same terminology during USWS in dolphins and fur seals,
cortical EEG slow wave generators of the two cortical
hemispheres appear to be in different functional states. We
hypothesize that in the extreme case (dolphin USWS),
those generators are going ÔONÕ and ÔOFFÕ in the sleeping
hemisphere and are being largely ÔOFFÕ in the waking
hemisphere over a period of longer than an hour.
ACKNOWLEDGEMENTS
This study was supported by NSF (0919929), NIH (069640)
and Utrish Dolphinarium Ltd. The authors are thankful to J.
Lapierre for comments on this manuscript.
REFERENCES
Borbe
´
ly, A. A. Sleep homeostasis and models of sleep regulation. In:
M. H. Kryger, T. Roth and D. C. Dement WC (Eds) Principles and
Practice of Sleep Medicine, 2nd edn. W.B. Saunders, Philadelphia,
1994: 309–320.
Drew, T., Andujar, J. E., Lajoie, K. and Yakovenko, S. Cortical
mechanisms involved in visuomotor coordination during precision
walking. Brain Res. Rev., 2008, 1: 199–211.
Fuchs, T., Maury, D., Moore, F. R. and Bingman, V. P. Daytime
micro-naps in a nocturnal migrant: an EEG analysis. Biol. Lett.,
2009, 5: 77–80.
Goldstein, L., Stoltzfus, N. W. and Gardocki, J. F. Changes in
interhemispheric amplitude relationships in the EEG during sleep.
Physiol. Behav., 1972, 8: 811–815.
Horne, J. A. Human sleep, sleep loss and behaviour: implications for
the prefrontal cortex and psychiatric disorder. Br. J. Psychiatry,
1993, 162: 413–419.
Huber, R., Deboer, T. and Tobler, I. Topography of EEG dynamics
after sleep deprivation in mice. J. Neurophysiol., 2000, 84: 1888–
1893.
Kattler, H., Dijk, D. J. and Borbe
´
ly, A. A. Effect of unilateral
somatosensory stimulation prior to sleep on the sleep EEG in
humans. J. Sleep Res., 1994, 3: 159–164.
8 O. I. Lyamin et al.
ª 2012 European Sleep Research Society
Krueger, J. M. and Tononi, J. Local-use dependent sleep; synthesis of
the new paradigm. Curr. Top. Med. Chem., 2011, 11: 2490–2492.
Krueger, J. M., Oba
´
l, F. and Fang, J. Why we sleep: a theoretical
view of sleep function. Sleep Med. Rev., 1999, 3: 119–129.
Lapierre, J. L., Kosenko, P. O., Lyamin, O. I., Kodama, T.,
Mukhametov, L. M. and Siegel, J. M. Cortical acetylcholine release
is lateralized during asymmetrical slow-wave sleep in northern fur
seals. J. Neurosci., 2007, 27: 11999–12006.
Lucas, E. A. and Sterman, M. B. The polycyclic sleep–wake cycle in
the cat: effects produced by sensorimotor rhythm conditioning.
Exp. Neurol., 1974, 42: 347–368.
Lyamin, O. I., Mukhametov, L. M. and Siegel, J. M. Relationship
between sleep and eye state in Cetaceans and Pinnipeds. Arch.
Ital. Biol., 2004, 142: 557–568.
Lyamin, O. I., Kosenko, P. O., Lapierre, J. L., Mukhametov, L. M. and
Siegel, J. M. Fur seals display a strong drive for bilateral slow.
J. Neurosci., 2008a, 28: 12614–12621.
Lyamin, O. I., Lapierre, J. L., Kosenko, P. O., Mukhametov, L. M. and
Siegel, J. M. Electroencephalogram asymmetry and spectral
power during sleep in the northern fur seal. J. Sleep Res.,
2008b, 17: 154–165.
Lyamin, O. I., Manger, P. R., Ridgway, S. H., Mukhametov, L. M. and
Siegel, J. M. Cetacean sleep: an unusual form of mammalian
sleep. Neurosci. Biobehav. Rev., 2008c, 32: 1451–1484.
Marzano, C., Ferrara, M., Curcio, G. and De Gennaro, L. The effects
of sleep deprivation in humans: topographical electroencephalo-
gram changes in non-rapid eye movement (NREM) sleep versus
REM sleep. J. Sleep Res., 2010, 19: 260–268.
Mukhametov, L. M., Supin, A. Y. and Polyakova, I. G. Interhemi-
spheric asymmetry of the electroencephalographic sleep patterns
in dolphins. Brain Res., 1977, 134: 581–584.
Mukhametov, L. M., Oleksenko, A. I. and Polyakova, I. G. The Black
Sea bottlenose dolphin: the structure of sleep. In: V. E. Sokolov
and E. V. Romanenko (Eds) The Black Sea Bottlenose Dolphin.
Nauka, Moscow, 1997: 492–512 [in Russian].
Murphy, M., Riedner, B. A., Huber, R., Massimini, M., Ferrarelli, F.
and Tononi, G. Source modeling sleep slow waves. Proc. Natl
Acad. Sci. USA, 2009, 106: 1608–1613.
Nir, Y., Staba, R. J., Andrillon, T. et al. Regional slow waves and
spindles in human sleep. Neuron, 2011, 70: 153–169.
Pigarev, I. N., Nothdurft, H. C. and Kastner, S. Evidence for
asynchronous development of sleep in cortical areas. NeuroRe-
port, 1997, 8: 2557–2560.
Rattenborg, N. C., Amlaner, C. J. and Lima, S. L. Behavioral,
neurophysiological and evolutionary perspectives on unihemi-
spheric sleep. Neurosci. Biobehav. Rev., 2000, 24: 817–842.
Ringli, M. and Huber, R. Developmental aspects of sleep slow
waves: linking sleep, brain maturation and behavior. Prog. Brain
Res., 2011, 193: 63–82.
Schwierin, B., Achermann, P., Deboer, T., Oleksenko, A., Borbe
´
ly, A.
A. and Tobler, I. Regional differences in the dynamics of the
cortical EEG in the rat after sleep deprivation. Clin. Neurophysiol.,
1999, 110: 869–875.
Supin, A. Y., Popov, V. V. and Mass, A. M. The Sensory Physiology
of Aquatic Mammals. Kluwer Academic, Boston, 2001.
Timo-Iaria, C., Negrao, N., Schmidek, W. R., Hoshino, K., De
Menezes, C. E. L. and Da Rocha, T. L. Phases and states of sleep
in the rat. Physiol. Behav., 1970, 5: 1057–1062.
Vyazovskiy, V., Borbe
´
ly, A. A. and Tobler, I. Unilateral vibrissae
stimulation during waking induces interhemispheric EEG asym-
metry during subsequent sleep in the rat. J. Sleep Res., 2000, 9:
367–371.
Vyazovskiy, V. V., Borbely, A. A. and Tobler, I. Interhemispheric
sleep EEG asymmetry in the rat is enhanced by sleep deprivation.
J. Neurophysiol., 2002, 88: 2280–2286.
Vyazovskiy, V. V., Olcese, U., Hanlon, E. C., Nir, Y., Cirelli, C. and
Tononi, G. Local sleep in awake rats. Nature, 2011, 472: 443–447.
Werth, E., Achermann, P. and Borbe
´
ly, A. A. Fronto-occipital EEG
power gradients in human sleep. J. Sleep Res., 1997, 6: 102–112.
Sleep in the fur seal 9
ª 2012 European Sleep Research Society
    • "The local, use-dependent facet of sleep regulation had been suggested about 20 years ago, and since then it has received substantial experimental support (Krueger and Obal 1993; Krueger et al. 2008). The most appealing evidence for the local occurrence of sleep has been obtained from marine mammals, such as dolphins and seals, which exhibit unihemispheric slow wave activity while in water (Lyamin et al. 2012; Mukhametov et al. 1977). Such asymmetries appear to be homeostatically regulated, as manifested in a unilateral increase of SWA after selective suppression confined to one hemisphere (Oleksenko et al. 1992 ). Spontaneous interhemispheric asymmetries of the EEG during sleep have also been found , albeit to a much lesser extent, in birds (Rattenborg et al. 2001Rattenborg et al. , 2012), humans, (Achermann et al. 2001; Nir et al. 2011) and rodents (Vyazovskiy et al. 2002a). "
    [Show abstract] [Hide abstract] ABSTRACT: In the last decades a substantial knowledge about sleep mechanisms has been accumulated. However, the function of sleep still remains elusive. The difficulty with unraveling sleep's function may arise from the lack of understanding of how the multitude of processes associated with waking and sleep-from gene expression and single neuron activity to the whole brain dynamics and behavior-functionally and mechanistically relate to each other. Therefore, novel conceptual frameworks, which integrate and take into account the variety of phenomena occurring during waking and sleep at different levels, will likely lead to advances in our understanding of the function of sleep, above and beyond what merely descriptive or correlative approaches can provide. One such framework, the synaptic homeostasis hypothesis, focuses on wake- and sleep-dependent changes in synaptic strength. The core claim of this hypothesis is that learning and experience during wakefulness are associated with a net increase in synaptic strength. In turn, the proposed function of sleep is to provide synaptic renormalization, which has important implications with respect to energy needs, intracranial space, metabolic supplies, and, importantly, enables further plastic changes. In this article we review the empirical evidence for this hypothesis, which was obtained at several levels-from gene expression and cellular excitability to structural synaptic modifications and behavioral outcomes. We conclude that although the mechanisms behind the proposed role of sleep in synaptic homeostasis are undoubtedly complex, this conceptual framework offers a unique opportunity to provide mechanistic and functional explanation for many previously disparate observations, and define future research strategies.
    Full-text · Article · May 2014
  • [Show abstract] [Hide abstract] ABSTRACT: Unilateral sleep in marine mammals has been considered as a defence against airway obstruction, as a sentinel for pod maintenance, and as a thermoregulatory mechanism. Birds also show asymmetric sleep, probably to avoid predation. The variable function of asymmetric sleep suggests a general capability for independence between brain hemispheres. Patients with obstructive sleep apnea share similar problems with diving mammals, but their eventual sleep asymmetry received little attention. The present report shows that human sleep apnea patients also present temporary interhemispheric variations in dominance during sleep, with significant differences when comparing periods of open and closed airways. The Magnitude of Squared Coherence, an index of interhemispheric EEG interdependence in phase and amplitude rises in the delta EEG range during apneic episodes, while the Phase Lag Index, a measure of linear and non-linear interhemispheric phase synchrony, drops to zero. The L index, which measures generalized nonlinear EEG interhemispheric synchronization, increases during apneic events. Thus, the three indexes show significant and congruent changes in interhemispheric symmetry depending on the state of the airways. In conclusion, when confronted with a respiratory challenge, sleeping humans undergo small but significant breathing-related oscillations in interhemispheric dominance, similar to those observed in marine mammals. The evidence points to a relation between cetacean unihemispheric sleep and their respiratory challenges.
    Article · Nov 2012
  • [Show abstract] [Hide abstract] ABSTRACT: Sleep is universal in animals, but its specific functions remain elusive. We propose that sleep's primary function is to allow individual neurons to perform prophylactic cellular maintenance. Just as muscle cells must rest after strenuous exercise to prevent long-term damage, brain cells must rest after intense synaptic activity. We suggest that periods of reduced synaptic input ('off periods' or 'down states') are necessary for such maintenance. This in turn requires a state of globally synchronized neuronal activity, reduced sensory input and behavioural immobility - the well-known manifestations of sleep.
    Article · May 2013
Show more

Recommended publications

Discover more