EEG IN ACTIVATED STATES: WAKEFULNESS AND REM SLEEP

Roger Godbout, Élyse Limoges, Anne-Marie Daoust, Christianne Bolduc, Laurent Mottron
Département de Psychiatrie, Université de Montréal, and
Centre de recherche Fernand-Seguin and Neurodevelopmental Disorders Program
Hôpital Rivière-des-Prairies
7070 boul. Perras, Montréal, Québec, Canada, H1E 1A4

Abstract: Spectral analysis was performed on EEG samples recorded from right-handed participants during two endogenously-activated states, i.e., resting wakefulness with eyes closed and REM sleep. Two questions were asked: 1) Can Beta‑1 EEG activity (14-20 Hz), thought to reflect the activity of "REM sleep-ON" neurons, segregate different brain states? 2) Do measures of Beta-1 EEG lateralization segregates between different brain states? We found that wakefulness and REM sleep generate comparable levels of Beta‑1 activity in control participants but not in persons with Autistic Spectrum Disorders (ASD), a neurologically-based Pervasive Developmental Disorder. Temporo-occipital Beta‑1 activity was lower in ASD compared to controls during REM sleep only. EEG lateralization analysis showed a REM sleep right frontal and left posterior temporal Beta‑1 asymmetry in control participants only; wakefulness did not show any lateralization effect. These results suggest that spontaneous Beta‑1 EEG activity generated during wakefulness and REM sleep are at least partly supported by different, topographically specific thalamo-cortical influences.

INTRODUCTION

Quantified analysis of the EEG can be used to characterize specific cortical areas that are activated under experimentally controlled conditions. REM sleep is a state typically showing an endogenous activation of the central nervous system; it thus constitutes an appropriate model to characterize spontaneous organization of neural networks such as the thalamo-cortical loop, on which EEG genesis is based [1,2]. REM sleep onset and maintenance mechanisms involve selective and specific ponto-mesencephalic and prosencephalic neural elements designated by the term "REM-ON" [3]. It has been proposed that EEG activity in the Beta range (14-30 Hz) is the one that best reflects the activity of REM-ON executive and permissive mechanisms [4]. Indeed it has been reported that Beta is the only EEG activity band that does not decrease relative to wakefulness and slow-wave sleep (SWS) [4]. It entertains a negative relationship with Delta activity during SWS. In fact, Delta, Theta, Alpha and Sigma activity bands are highly correlated amongst each other while Beta activity varies independently. Beta is also reported to be the only EEG activity window not showing a decrease in REM sleep; only Beta activity shows greater values in REM than in SWS. The host of these observations constitutes the basic arguments suggesting that Beta activity reflects the activity of REM-ON elements [4].

Measures of left-right EEG asymmetry have been found to reflect cognitive task hemisphericity [5]. Few studies have investigated EEG lateralization during sleep, some of which being performed during REM sleep. An early report has suggested that the right hemisphere is more activated than the left during REM sleep, as indexed by higher EEG power in healthy participants [6]. The aim of the various studies summarized below was to investigate the following two questions: 1) Can Beta‑1 EEG activity, thought to reflect the activity of REM sleep "ON" neurons, segregate different brain states? 2) Do measures of Beta-1 EEG lateralization segregates between different brain states?

METHODS

Subjects included 18 healthy control participants (10M, 8F; mean age 21.3 ± 2.1 years) and nine persons diagnosed with Autistic Spectrum Disorders (8 M, 1 F; mean age 22.2 ± 4.1 years). All participants were right-handed and had a full-scale IQ of at least 80. The electrode montage included bilateral frontal (Fp1, Fp2, F7, F8), central (C3, C4), temporal (T3, T4, T5, T6), parietal (P3, P4) and occipital (O1, O2) electrodes [7]. EEG was recorded for two consecutive nights using monopolar montage referred to linked ears with serial 10 KOhms resistance for signal equilibrium purposes. Filter settings and amplification factors were as follows: EEG: 1/2 amplitude low frequency filter = 0.3 Hz, 1/2 amplitude high frequency filter = 100 Hz, gain = 20,000. Records were digitized at a sampling rate of 128 Hz and stored for off-line visual inspection on a computer screen. Each participant was recorded for two consecutive nights. The first night was meant to adapt to experimental conditions as well as to formally screen sleep disorders, including sleep apneas and periodic leg movements. Sleep stages were scored in 20‑sec. epochs according to standard methods [8]. Wakefulness EEG activity was recorded just before lights out (between 22h00 and 23h00) and the morning after (between 07h00 and 08h00). Recordings lasted five minutes with the participants having their eyes closed. EEG samples were made of four-second segments totaling 60 to 96 seconds of artifact-free EEG tracing. REM sleep EEG samples were made of 15 to 24 four-second segments, taken in equal proportions from the first three REM sleep periods of night two, during artifact-free, quiescent sleep. All EEG samples were Fast Fourier-transformed using cosine window smoothing, with a frequency resolution of 0.25 Hz. Spectral analysis was performed on Beta-1 activity (14.0-19.75 Hz). EEG lateralization was determined by computing a coefficient using the formula: ((Right - Left) / (Right + Left) x 100), applied on spectral amplitude values (mV) from homologous right and left electrodes. Persons with ASD were statistically compared to controls as a group, on an age and gender basis.

Figure 1. Absolute spectral amplitude of Beta-1 activity in Control and ASD participants upon wakefulness and REM sleep.

Table 1. Lateralization coefficients of Beta-1 activity during REM sleep
in Control participants (Mean ± s.e.m.)

Fp	1.28±0.1
F	3.02±1.0*
C	0.14±0.8
T3-T4	1.36±2.2
T5-T6	-5.35±1.0*
P	1.39±0.6
O	0.78±0.8
*: p<.05, within-group paired t-tests.

RESULTS

The 2 (Group) X 3 (moment) ANOVA showed that Beta-1 activity was comparable during wakefulness and REM sleep despite an apparent, non significant tendency toward decreased levels in the latter state. Figure 1 shows typical results, for the C3 electrode. Control participants and persons with ASD showed comparable Beta-1 activity during wakefulness.  On the other hand, during REM sleep, persons with ASD showed a significantly decreased Beta-1 activity compared to controls on temporo-occipital electrodes (T5, T6, O1, and O2).

Analysis of EEG lateralization showed a right frontal (F8) and a left temporal (T5) Beta‑1 asymmetry in control participants, not in persons with ASD, while wakefulness did not show any lateralization effect (see Table 1).

DISCUSSION

Like others [4] we found that Beta‑1 voltage is comparable for wakefulness and REM sleep in control participants. On the other hand we found decreased Beta-1 activity during REM sleep compared to wakefulness in persons with ASD, a neurologically-based Pervasive Developmental Disorder [15] shown to display abnormalities of REM sleep [9]. Together with the fact that Temporo-occipital Beta‑1 EEG activity was selectively decreased during REM sleep in ASD, these results supports the notion that REM control mechanisms may be atypical in ASD.

EEG lateralization analysis in control participants showed a right frontal and a left posterior temporal Beta‑1 asymmetry to be present during REM sleep only. REM sleep dreaming is thought to be associated with the right hemisphere in right handers [10,11, but see 12]. On the other hand, the left temporal dominance could reflect controlateralized visuo-spatial and/or limbic activities related to REM sleep cognition. The fact that no such lateralization effect was found in participants with ASD, together with observations of decreased temporo-occipital activity, could be related to the atypical dream content in this clinical entity [13,14].

We conclude that spontaneous Beta-1 EEG activity generated during wakefulness and REM sleep are at least partly supported by different, topographically specific thalamo-cortical influences.

Supported by the Canadian Institutes for Health Research and the Fonds de la recherche en santé du Québec.

REFERENCES

[1] McCormick D.A., Bal T. “Sleep and arousal: Thalamocortical mechanisms.” Annual Review of Neuroscience, vol. 20, pp. 185-215, 1997.

[2] Steriade M., McCormick D.A., Sejnowski T.J. “Thalamocortical oscillations in the sleeping and aroused brain” Science, vol. 262, pp. 679-685, 1993.

[3] Hobson J.A., Pace-Schott E.F., Stickgold R. “Dreaming and the brain: toward a cognitive neuroscience of conscious states,” Behavioral & Brain Sciences, vol. 23, pp. 793-842; discussion 904-1121, 2000.

[4] Merica, H., Blois, R. “Relationship between time courses of power in the frequency bands of human sleep EEG,” Clinical Neurophysiology, vol. 27, pp. 116-128, 1997

 [5] De Toffol B., Autret, A., Degiovanni, E., Roux, S.  “Spectral analysis of the EEG (alpha rythm) and activity in the left hemisphere: the effects of lateral gaze” Neuropsychologia, vol. 28 (8), pp. 879-882,1990.

[6] Goldstein L., Stoltzfus N.W., Gardocki J.F. “Changes in interhemispheric amplitude relationships in the EEG during sleep” Physiology & Behavior, vol. 8(5), pp. 811-815, 1972.

[7] Jasper, H.H. “The ten-twenty electrodes system of the International Federation”. Electroencephalography and Clinical Neurophysiology, vol. 10, pp. 371-373, 1958.

 [8] Rechtschaffen, A., Kales, A. A manual of standardized terminology, techniques and scoring system for sleep stages of human subjects. Los Angeles: BIS/BRI, University of California at Los Angeles, 1968.

[9] Godbout R., Bergeron, C., Limoges, E., Stip, E., Mottron, L. “A laboratory study of sleep in Asperger’s syndrome,” NeuroReport, vol. 11, pp. 127-130, 2000.

[10] Serafetinides E.A.. “Cerebral dominance, sleep and dream phenomena” International Journal of Neuroscience , vol. 71, pp. 63-70, 1993.

[11] Gabel S. “The right hemisphere in imagery, hypnosis, rapid eye movement sleep and dreaming. Empirical studies and tentative conclusions” Journal of Nervous & Mental Disease, vol. 176, pp. 323-331, 1988.

 [12] Antrobus J. Sleep and cognition. “The neurocognition of sleep mentation: rapid eye movements, visual imagery and dreaming”. Bootzin RR, Kihlstrom JF, Schacter DL, editors. Washington DC: American Psychological Association, pp.3-24. 1990.

[13] Craig J., Baron-Cohen, S. “L’hypothèse de la théorie de l’esprit: les enfants autistes parlent-ils de leurs rêves?,” Psychologie Française, vol. 43, pp. 169-176, 1998.

[14] Daoust, A.M., Mottron, L., Godbout, R. “Dream content analysis in Asperger’s Syndrome,” Sleep, vol. 24 (Suppl.), pp. A184-A185, 2001.

[15] Minshew, N.J., Sweeney, J.A., Bauman M.L. “Neurological aspects of autism”, Handbook of Autism and Pervasive Developmental Disorders, D. Cohen, F.Volkmar (eds.). New York: Wiley, pp. 344-369, 1997.