Psychology dep.; Brock
University, St Cathatarines, Canada
Does yawning have an effect on arousal
level ? Since we could not locate any
studies in the literature addressing this
question, the following analyses were planned as
part of a series of investigations into sleep
onset mechanisms. Twelve university students
spent two consecutive nights each in the sleep
lab where, in addition to standard
polysomnographic recordings, respiration and
behavioral responsiveness (Ogilvie &
Wilkinson, 1984) assessed multiple transitions
into sleep.
(Subjects were awakened after their first
four sets of response failures, used in our lab
to define sleep onset). These data were recorded
on FM tape for subsequent computer analyses.
Split screen CCTV monitoring of the subject and
concurrent electrophysiological activity made it
possible to identify instances of yawning and
relate those episodes precisely to polygraphic
and FM tape recordings.
There were 35 instances of yawning detected
in 10 subjects over the 24 nights. However, the
data were further seriously reduced by the
requirement that there be 30 sec. of
artifact-free EEG activity immediately preceding
and following the yawn. As a control condition,
EEG samples were taken from the same subjects
before and after postural adjustments which
produced movement artifacts of similar
amplitudes and durations to those seen during
yawning.
Thirty sec. samples of EEG (C3-A1) taken
immediately prior to and following yawns and
postural shifts were subjected to FFT analysis
in 5 sec. blocks. Data from only four subjects
were analyzed. Alpha, theta, and spindle
frequencies were examined both in terms of
absolute power and percent power (.4 to 25 Hz
range) for three, 5 sec. blocks (5, 15, and 25
sec.).
These data were entered into 2 x 2 x 3
repeated measures ANOVAs (pre/post;
yawn/postural shift; 3 time blocks).
There were no significant differences in
absolute or percent power in any of the three
frequency bands analyzed for any main effect or
interaction. Unambiguous examples of yawning
behavior recorded in the sleep lab prior to
continuous nocturnal sleep do not appear to be
associated with rapid changes in EEG-based
levels of arousal.
mise à jour
du
27 mars
2005
Sleep
Research
1992;21:14
Electrophysiological
correlates of yawning
KL Regehr, RD Ogilvie, IA
Simons
Psychology dep.;
Brock University, St Cathatarines,
Canada
Although an earlier study from this lab
found no consistent relationship between yawning
behavior and EEG changes, we decided to expand
and replicate that work, for there were
indications in Laing's data that differences
might stabilize in a more comprehensive
comparison of EEG activity 30 sec prior to and
following a yawn. In that work, the clearest
changes appeared 15 sec prior to and following
the yawn in the Alpha and Theta
frequencies.
Fifteen university students were paid to
spend 2 nights in the sleep lab as part of a
larger sleep onset investigation. After
preparations for standard polysomnographic
recording were complete, subjects were asked to
read for 15 min immediately prior to a night's
sleep. During that time, computerized EEG
recordings were obtained, as was a simultaneous
split screen videotaping of the subject reading
and his/her concomitant EEG activity. The
videotapes were viewed for clear instances of
spontaneous yawning. Once yawning episodes were
identified, the appropriate EEG epochs were
located on the computer, and separate FFT
analyses were obtained for each of 3 consecutive
10 sec blocks immediately prior to and following
each yawn. Interest was focused on the middle
period - 10 to 20 sec on either side of the
yawn. 112 artifact free spontaneous yawns were
analyzed. Yawn frequency varied from zero to 26
per subject (X= 13.17; SD= 5.74, N=12). Thus 12
separate FFTs were averaged within subjects, so
that one set of RMS power measures per subject
were treated statistically.
The primary analyses were 2 x 3 (pre/post
yawn x 3, 10 sec blocks) MANOVAs calculated on
RMS power measures for each of the 5 standard
frequency bands. These analyses produced no
significant main effects or interactions.
However, since Laing had found significant
changes when examining 5 sec blocks of EEG 15 to
20 sec prior to and following yawns, we had
planned similar specific comparisons in this
investigation. When correlated t-tests (df=11)
were computed comparing the ore- and post-yawn
10 to 20 sec blocks (all 5 frequencies), the
following significant differences were obtained:
Delta (t = 2.23, p<.05; xpre = 502,xpost =
426); Theta (t = 4.15, p<.002; xpre = 172,
xpost = 219); Sigma (t = 3.09, p<,01; xpre =
30.3, xpost = 37.8); Beta (t = 2.64, p<,02;
xpre = 57.4, xpost = 72.3). These analyses
indicate that there may be transient decreases
in Delta and increases in Theta, Sigma
(spindle), and Beta frequencies associated with
yawning. The observed changes in Theta
frequencies were consistent with those found by
Laing, but no changes in Alpha were
detected.
In summary, there are no relatively
long-lasting changes in EEG activity
attributable to yawning when pre- and postyawn
samples are subjected to FFT analysis. Transient
changes in Delta, Theta, Sigma, and Beta
activity were noted, but similar effects were
not detectable as little as 10 sec nearer or
farther from the yawn. Thus yawning may produce
fleeting changes in EEG indices of arousal, but
presleep yawns do not represent points when the
rate or direction of movement along the arousal
continuum changes dramatically.
