New vistas for Alpha-frequency band oscillations

Department of Biological and Environmental Sciences, University of Helsinki, Helsinki, Uusimaa, Finland
Trends in Neurosciences (Impact Factor: 13.56). 05/2007; 30(4):150-8. DOI: 10.1016/j.tins.2007.02.001
Source: PubMed


The amplitude of alpha-frequency band (8-14 Hz) activity in the human electroencephalogram is suppressed by eye opening, visual stimuli and visual scanning, whereas it is enhanced during internal tasks, such as mental calculation and working memory. alpha-Frequency band oscillations have hence been thought to reflect idling or inhibition of task-irrelevant cortical areas. However, recent data on alpha-amplitude and, in particular, alpha-phase dynamics posit a direct and active role for alpha-frequency band rhythmicity in the mechanisms of attention and consciousness. We propose that simultaneous alpha-, beta- (14-30 Hz) and gamma- (30-70 Hz) frequency band oscillations are required for unified cognitive operations, and hypothesize that cross-frequency phase synchrony between alpha, beta and gamma oscillations coordinates the selection and maintenance of neuronal object representations during working memory, perception and consciousness.


Available from: Satu Palva
New vistas for a-frequency band
Satu Palva
and J. Matias Palva
Neuroscience Center, University of Helsinki P.O. Box 56, FI-00014 University of Helsinki, Finland
BioMag Laboratory, Helsinki University Central Hospital P.O. Box 340, FI-00029 HUS, Finland
Cognitive Brain Research Unit, Department of Psychology, University of Helsinki P.O. Box 9, FI-00014 University of Helsinki,
Department of Biological and Environmental Sciences, University of Helsinki P.O. Box 65, FI-00014 University of Helsinki, Finland
The amplitude of a-frequency band (8–14 Hz) activity in
the human electroencephalogram is suppressed by eye
opening, visual stimuli and visual scanning, whereas it is
enhanced during internal tasks, such as mental calcu-
lation and working memory. a-Frequency band oscil-
lations have hence been thought to reflect idling or
inhibition of task-irrelevant cortical areas. However,
recent data on a-amplitude and, in particular, a-phase
dynamics posit a direct and active role for a-frequency
band rhythmicity in the mechanisms of attention and
consciousness. We propose that simultaneous a-, b- (14–
30 Hz) and g- (30–70 Hz) frequency band oscillations are
required for unified cognitive operations, and hypothes-
ize that cross-frequency phase synchrony between a, b
and g oscillations coordinates the selection and main-
tenance of neuronal object representations during work-
ing memory, perception and consciousness.
Hans Berger was among the first to witness
electroencephalogram (EEG) rhythms in the a- (8–
14 Hz) and b- (14–30 Hz) frequency bands [1]. During
the following four decades, the parieto-occipital a rhythm
was found to be attenuated by eye opening, visual stimuli
and by increased attentiveness. These findings inspired
the idea of a oscillations functioning as an ‘idling’ rhythm
that characterizes an alert-but-still brain state [2]. Today,
the idling hypothesis has been largely overtaken by a
framework where the amplitude of a oscillations reflects
a level of cortical inhibition [3–7]. Accumulating data on a
phase dynamics [8–12], however, add a twist to the story;
endogenous as well as stimulus-locked a-band phase
correlations seem to have a direct role in the neuronal
machinery underlying behavioral-level phenomena, such
as attention, STM and sensory awareness. These findings
challenge the inhibition hypothesis, as well as the prevail-
ing interpretation of a amplitude dynamics.
Do large-amplitude a oscillations reflect cortical
Early on, the a-band oscillations were observed to be
strengthened during internal tasks, such as mental
arithmetic and visual imagery, which was interpreted by
Ray and Cole [3] to reflect rejection of sensory information
intake. This idea was advanced into an a-inhibition hy-
pothesis by Klimesch [4] and Pfurtscheller [5,6], who pro-
posed that small a amplitudes are a signature of regions of
active neuronal processing, whereas large-amplitude a
oscillations reflect the inhibition and disengagement of
task-irrelevant cortical areas. The core phenomenon sup-
porting this hypothesis is the a-amplitude suppression,
also known as event-related desynchronization (ERD),
that follows sensory stimuli in the corresponding sensory
areas [13] (Figure 1a). In addition, in some experimental
paradigms, a amplitude is enhanced in EEG electrodes
surrounding those where the ERD is observed [5–7].
Although changes in the amplitude of a oscillations indeed
seem to be involved in the continuation of sensory stimulus
processing [14], we argue here that it is unfeasible to
deduce in a one-track fashion that large a amplitudes
correspond to inhibited or disengaged cortical states.
A large body of recent dataconfirms thata oscillations are
strengthened by internal tasks, such as mental calculation
[11] and mental imagery [15–17]. In addition, a-band ampli-
tude is also enhanced during the short-term- (STM) and
working memory (WM) retention period, and is suppressed
thereafter [18–20] (Figure 1b). Klimesch et al. [7] suggest
that these large-amplitude oscillations during memory
retention inhibit the retrieval of memorized items, and that
this is then reflected in the subsequent amplitude suppres-
sion. We propose that the STM- and WM-related a oscil-
lations in the frontoparietal network [10,11,18] during the
memory retention period are an essential constituent of the
network activity that sustains the neuronal representations
of memorized items. In this light, the retrieval-associated a
suppression [18] (Figure 1b) reflects, in part, the termin-
ation of the memory process itself. This view is supported by
the positive correlation of the a amplitude with the STM and
WM load [18,19] (Figure 1d) and task difficulty [20].
Besides being modulated by sensory stimuli and
movements, the a-frequency band amplitude is also modu-
lated by attention. In visual attention tasks, visual [21] and
auditory [22] cue stimuli are followed by an a suppression
that is larger in the occipital cortex contralateral than
ipsilateral to the attended visual hemifield (Figure 1c).
According to the inhibition framework, this post-cue
and pretarget stimulus a suppression, as well as the
TRENDS in Neurosciences No.x
Corresponding authors: Palva, S. (;
Palva, J.M. (
Available online xxxxxx.
TINS-503; No of Pages 9 0166-2236/$ see front matter ß 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.tins.2007.02.001
Please cite this article in press as: Palva, S. and Palva, J.M., New vistas for a-frequency band oscillations, Trends Neurosci. (2007), doi:10.1016/j.tins.2007.02.001
Page 1
post-target-stimulus ERD, reflects the ‘spotlight of
attention’ that is physiologically implemented indirectly
by releasing the task-relevant areas from inhibition,
through a ERD, and by suppressing task-irrelevant
regions with large-amplitude a oscillations [7]. However,
the effect of attention on mean a amplitude in the visual
cortex is relatively minor (10% of the baseline level
[21,22]) in comparison with the trial-by-trial variation in
a-frequency band amplitude, which is, in general, consider-
able [13,23] (Figure 1a). Interestingly, in a somatosensory
detection task, small and large prestimulus a-frequency
band amplitudes in the sensorimotor region led to equal
psychophysical performance [23]. In fact, in this study,
intermediate prestimulus a amplitudes were associated
with facilitated detection and the fastest reaction times
(Figure 1d). These observations are not in line with the
concept of the a amplitude having a linear relationship
with cortical inhibition, and nor would these results be
seen in amplitude estimates averaged across trials. Unfor-
tunately, this type of a single-trial analysis has not
been published for the attention-task data [21,22]. More-
over, in the study by Linkenkaer-Hansen et al. [23], the
Figure 1. Event-related a-band amplitude dynamics in the human brain. (a) In the sensorimotor cortex, the amplitude of ongoing a-frequency band oscillations, as recorded
by MEG, is consistently suppressed by suprathreshold electrical stimuli to the median nerve. Amplitude of single trials is color coded (blue, small; red, large). Amplitude
(‘ampl.’) is averaged across trials and expressed relative to the baseline amplitude. Modified, with permission, from Ref. [13]. (b) Subjects were presented visually with six
items (‘memory set’), of which two, four or six had to be memorized until the probe stimulus was presented and the subject indicated whether the ‘probe’ item was one of
memorized items. Parietal EEG a-frequency band amplitude is enhanced throughout the STM retention period (yellow and red colors correspond to large amplitudes).
During the retention period, a-frequency band amplitude dependent on the number of memorized items. Modified, with permission, from Ref. [18]. Abbreviation: norm.,
normalized (c) A symbolic ‘cue’ (an arrow) pointing to the to-be-attended hemifield was followed by a ‘stimulus’ in the left or in the right hemifield. The subject was
instructed to react to specific target stimuli in the attended hemifield. After the cue, EEG a-frequency band amplitude, averaged over trials, is smaller in the occipital cortex
contralateral to the attended visual hemifield. Modified, with permission, from Ref. [21]. (d) The subjects’ finger tips were stimulated electrically at the threshold of
sensation. The subjects were instructed to react to perceived stimuli. In MEG recordings, the intermediate (but not the small or large) prestimulus a-frequency band
oscillations in the sensorimotor cortex facilitate stimulus detection. In the parietal cortex, by contrast, large amplitude oscillations are associated with best behavioural
performance. Modified, with permission, from Ref. [23].
TRENDS in Neurosciences No.x
TINS-503; No of Pages 9
Please cite this article in press as: Palva, S. and Palva, J.M., New vistas for a-frequency band oscillations, Trends Neurosci. (2007), doi:10.1016/j.tins.2007.02.001
Page 2
best behavioral performance was preceded by the largest
parietal a oscillations (Figure 1d). This can be seen to
reflect the inhibition of (task-irrelevant, visual) cortical
areas but, equally well, the finding might show task-
relevant a-frequency band network activity in parietal
regions that contribute to the attentional control of the
detection task.
All things considered, investigations of amplitude
dynamics, as such, are inconclusive on the functional
role of a-frequency band oscillations. In the majority of
the amplitude data, a oscillations might be seen to reflect
active processing in task-relevant networks or active inhi-
bition of task-irrelevant regions, or both. Notably, many of
the conclusions have been based on EEG recordings that
involve considerable spatial smearing and are therefore
difficult to exploit in the identification of the task-relevant
or -irrelevant cortical regions. More importantly, it is pro-
blematic, if not impossible, to interpret the neurophysiolo-
gical underpinnings and consequences of changes in field
oscillation amplitude (Box 1). Therefore, the functional
significance of a-amplitude dynamics in task execution
remains unclear.
Alpha-frequency band phase correlations
Compared with the large body of data on a-amplitude
dynamics, investigations on a-phase dynamics have
remained infrequent, although this line of research is
now rapidly gaining popularity. Phase synchrony is essen-
tial in the formation of transient neuronal assemblies [24],
in communication therein [25] and, consequently, in large-
scale integration [26]. Phase interactions thus define
functional networks in the cerebral cortex. Therefore, we
propose here that it is the phase (Box 1), not amplitude,
dynamics that reveals the functional significance of a-
frequency band rhythmicity.
Using magnetoencephalography (MEG), we observed
that ongoing a-frequency band oscillations in human sen-
sorimotor, as well as frontoparietal regions known to be
relevant for attention [27] and consciousness [28,29], phase
lock selectively to weak somatosensory stimuli that become
consciously perceived [12] (Box 1, Figure Ia, b). Also,
g-frequency band (here, 30–50 Hz) phase locking was
selective for perceived stimuli but was very local, perhaps
present only in the primary somatosensory area (Box 1,
Figure Ib). However, although the phase locking of large-
scale u- (4–7 Hz) and more local b- (14–30 Hz) frequency
band oscillations was clearly correlated with conscious
perception, this phase locking was highly significant also
for the unperceived stimuli. Thus, perception-selective
a-frequency band phase locking in sensorifrontoparietal
networks indicates a direct role for a-frequency band phase
correlations in neuronal processes supporting conscious-
ness. In these data, the phase locking in not only a-, but
also in b- and g-bands was followed by an amplitude
suppression that was greater for the perceived than for
the unperceived stimuli. Nevertheless, this amplitude sup-
pression (or ERD), especially in the a-frequency band, is
unlikely to reflect active stimulus processing because
the most pronounced suppression takes place after the
behavioral responses of the subjects (Box 1, Figure Ic).
The relevance of a-frequency band phase dynamics has
also been recognized in several other studies. Gail et al. [30]
Box 1. Event-related phase and amplitude dynamics
We describe here the data-analysis approaches that have been most
widely used in the investigations on a oscillations presented in here.
For several decades, averaging across several stimulus presenta-
tions has been used to isolate information on event-related neuronal
processing from recordings of ongoing neuronal activity. Tradition-
ally, peristimulus epochs of recorded data are ‘cut’ and averaged to
obtain an ‘evoked’ response [Figure I(a)(i)]. Evoked responses contain
time domain signal components that are locked to the stimulus,
whereas activities uncorrelated with the stimuli are lost in the
averaging procedure.
It is possible to identify comp onents that are not stimulus-locked by
averaging estimates of the si gnal amplitude instead of the signal itself
[Figure I (a)(iii)]. This approach reveals ‘mean amplitude dynamics’ or
‘induced’ oscillations [51]. In many studies on a -frequency band
oscillations, the amplitude is estimated by squaring or rectifying a
conventionally filtered signal before averaging, as in the ‘ERD’ [6] or
‘temporal spectral evaluation’ [15] methods, respectively. Alterna-
tively, the continuous ‘analytical’ amplitude [as in Figure I (a)(iii)] can
be obtained with complex wavelets [12,51], as well as with conven-
tional filtering utilized with the Hilbert transform [12].
Classical time-domain averaging is an acceptable approach in a
scenario in which activities of interest are evoked by stimuli but
buried in ongoing, uncorrelated background ‘noise’. However, it has
been realized that the evoked responses in EEG and MEG might also
arise from a phase reset of ongoing oscillations. Because conven-
tionally filtered signals contain both phase and amplitude informa-
tion, the magnitude of evoked responses does not index the extent of
such phase locking. The phase locking of ongoing activity to stimuli
can be quantified by first obtaining the continuous phase [Figure I
(a)(ii)] and averaging as earlier, but by utilizing a measure of phase
dispersion, such as the PLF [12], across the epochs for each point in
time [Figure I (a)(ii)]. The continuous phase can be obtained with both
wavelet and Hilbert-transform approaches [12,51]. Finally, it should
be noted that, in addition to phase reset, evoked components are also
visible in phase-locking analyses.
Interpretation of field amplitude dynamics
The classical assumption underlying the interpretation of EEG and
MEG amplitude changes is that the overall neuronal activity level is
approximately constant, and the field amplitude changes are caused
by changes in neuronal synchrony. Hence, an amplitude decrease is
called an ERD and an amplitude increase an ‘event-related synchro-
nization’ (ERS).
However, in the absence of changes in actual synchrony, the field
amplitude might change if the number of neurons entrained to the
oscillation is changed. In addition, phase relationships of subpopula-
tions strongly affect the fi eld amplitude. For instance, without
changes in the number of active neurons or in the total degree of
synchrony, a thalamic a oscillation waxes and wanes depending on
the size of an antiphase subpopulation in the thalamic network [39].
Moreover, considerable phase ordering might take place in the
absence of amplitude changes [12] or even during simultaneous
amplitude suppression [9,10]. Thus, caution must be exerted when
making physiological or functional inferences based on oscillation
However, regardless of the recording method and level of inspec-
tion, the phase of an oscillation is always, at least approximately,
related to spike timing. The presence of field-signal phase synchrony
thus indicates the presence of neuronal-level spike synchrony, which
is relevant in the light of the crucial role of spike timing in neuronal
communication [25,79]. Similarly to the case of amplitude, the
estimation of phase dynamics also contains several pitfalls but it is
our view that the phase information remains crucial in the identifica-
tion of functional neuronal networks.
TRENDS in Neurosciences No.x 3
TINS-503; No of Pages 9
Please cite this article in press as: Palva, S. and Palva, J.M., New vistas for a-frequency band oscillations, Trends Neurosci. (2007), doi:10.1016/j.tins.2007.02.001
Page 3
recorded field potentials and multiunit activity from the
monkey primary visual cortex and found that the field
potential coherence in 4–12 Hz and 12–28 Hz bands was
positively correlated with perception in binocular rivalry.
Mima et al. [9] presented human subjects with meaningful
and meaningless visual objects and found that,
when attended, meaningful objects strengthened EEG
a-frequency band coherence in the occipitotemporal region.
Meaningless or unattended objects were followed by
decreased coherence. Hanslmayr et al. [31] used a percep-
tual discrimination task and found stronger phase locking
in the EEG a-frequency band for the good than bad per-
formers. Using depth electrodes in human intracranial
recordings, Halgren et al. [10] observed phase synchrony
in a- and other frequency bands in occipital, parietal,
frontal and Rolandic regions during periods of mental
calculation and WM maintenance. In MEG recordings of
human cortical ongoing activity, we found that mental
calculation is associated with enhanced frontoparietal
a- and b-frequency band phase synchrony [11]. Taken
together, these data show that cognitive tasks involve
pronounced large-scale a-frequency band phase syn-
In their seminal work, von Stein and colleagues [8],
recorded field potentials from the primary visual cortex
(area 17) and from a higher-level association area (area 7)
Figure I. (a) Simulated 8–12 Hz ongoing activity (i) with phase (ii) and amplitude (iii) dynamics reminiscent of those observed in MEG data [12]. Note that for
visualization purposes, the strength of phase locking (ii), as well as the magnitude of event-related amplitude suppression (iii) have been exaggerated here. (b) (i) Phase
locking of ongoing activity over the primary somatosensory cortex to weak somatosensory stimuli. Phase locking in the a-frequency band is significant exclusively for
the stimuli that were consciously perceived. (ii) Unlike the phase locking in the b- and g-frequency bands, the a-frequency band phase locking to the perceived stimuli
involves a large-scale frontoparietal network. (c) (i,ii) Interestingly, early phase locking in the a-, b- and g-frequency bands was not associated with significant changes in
mean amplitude, whereas phase locking in the u-frequency band was closely paralleled by an enhanced amplitude. This suggests that u-frequency band phase locking
reflects a classical evoked component, whereas a- and b-frequency band phase locking is produced by a phase reset of ongoing activity [44]. Phase locking (b) and
amplitude (c) are color scaled. The transparency scale indicates the statistical significance of the results: Bc, Bonferroni correction with the number of samples (b, 240) or
channels (c, 102); cSI and cSII, putative locations of contralateral primary and secondary cortices, respectively; NS, not significant; P
and P
, P value of Binomial and
Wilcoxon signed-rank statistics, respectively. Modified, with permission, from Ref. [12].
TRENDS in Neurosciences No.x
TINS-503; No of Pages 9
Please cite this article in press as: Palva, S. and Palva, J.M., New vistas for a-frequency band oscillations, Trends Neurosci. (2007), doi:10.1016/j.tins.2007.02.001
Page 4
in cats performing a ‘Go–No-Go’ task. Their data showed
that interareal a-frequency band synchrony was prominent
during ‘Go’ trials and negligible during ‘No-Go’ trials. More-
over, this interareal a-frequency band synchrony was
prominent in the responses to the expected stimuli
but absent in responses to occasional novel stimuli
(Figure 2a). These data show that a-frequency band syn-
chrony is involved in top-down modulation of behaviorally
Figure 2. Evidence supporting a direct role for a-frequency band oscillations in the mechanisms of the global neuronal workspace. (a) (i) Cats were trained to perform a Go–No-
Go task. The well-learned target and non-target stimuli were then interspersed with novel stimuli. Cross-correlation analyses of local field potentials recorded from the primary
visual cortex (area 17) and distant visual association cortex (area 7) show that a-frequency band synchrony (upper row) is prominent for ‘expected’ target stimuli (left-hand
column) and insignificant for both the ‘novel stimuli (right-hand column) and the No-Go stimuli (not shown). g-Frequency band synchrony (lower row), by contrast, was
stronger for the novel than for the expected stimuli. (ii) Patterns of strengthened (unbroken lines) and attenuated (dashed line) synchrony during Go’ stimuli. g-Frequency-band
synchrony (red) is predominantly local and is found only in the granular and supragranular layers of area 17. a-Frequency band synchrony (blue), by contrast, is robust
both intra- and interareally, and especially between the infragranular layers of area 7 and supragranular layers of area 17. The data in (i) and (ii) indicate a role for large-scale
a-frequency band synchrony in attentional top-down modulation. Modified, with permission, from Ref. [8]. (b) (i) MEG recordings of ongoing neuronal activity in the human
brain reveal that g- (red) and a- (blue) frequency band oscillations can become transiently 1:3- and 1:4-phase synchronized. The phase-locking factor (‘PLF
and ‘PLF
’) traces
above the gray horizontal rectangle indicate periods of statistically significant cross-frequency phase synchrony. 1:3 ga synchrony is also easily seen in the continuous ‘phase
traces of g-anda-frequency band oscillations (see red-blue striped arrows). (ii) Compared with active rest, continuous mental calculation tasks enhance both within-frequency
(1:1) and cross-frequency (1:2 and 1:3) phase synchrony among a-, b-andg-frequency band oscillations. Modified, with permission, from Ref. [11]. (c) (i) Continuous wagon-
wheel illusion experiments with human subjects show that illusory reversals are most frequent at motion frequencies in the a-frequency band. The graph shows reported actual
(red squares) and opposite (blue squares) motion perception averaged across subjects. These data suggest that human visual perception takes place in discrete frames
occurring at a-frequencies. Modified, with permission, from Ref. [69]. (ii) The difference in EEG amplitude between a perception of real and illusory motion is significant (orange
square) only in the upper a-frequency band and is independent of the temporal frequency of the stimulus. The findings suggest that a-frequency band neuronal oscillations are
involved in perceptual sampling. Modified, with permission, from Ref. [72]. (d) (i) Smooth self-paced finger movements are characterized by discontinuities at 6–9 Hz, which
indicates that human action, similarly to perception [see (c)], takes place in discrete frames. Finger velocity (upper trace) and electromyography (EMG) of finger agonist and
antagonist muscles (middle and lower histograms) were averaged around velocity discontinuities. These discontinuities were associated with clear out-of-phase agonist
antagonist bursting in the EMG. (ii) EMG signals in the 6–9 Hz band were coherent with the ongoing neuronal oscillations in the contralateral sensorimotor cortex so that
coupling in the postcentral gyrus (somatosensory cortex) was predominantly afferent (red), whereas the coupling in the precentral gyrus (motor cortex) was pr edominantly
efferent (blue). Thus, similarly to perceptual sampling, discrete motion is also coordinated by cortical a-frequency band oscillations. Modified, with permission, from Ref. [73].
TRENDS in Neurosciences No.x 5
TINS-503; No of Pages 9
Please cite this article in press as: Palva, S. and Palva, J.M., New vistas for a-frequency band oscillations, Trends Neurosci. (2007), doi:10.1016/j.tins.2007.02.001
Page 5
significant neuronal processing. Two additional lines of
evidence strongly support this conclusion. First, synchro-
nized a-frequency band oscillations in the primary visual
cortex lagged behind those in the association cortex, imply-
ing that the association area drove oscillations in the
primary visual cortex through feedback connections
(Figure 2a). Second, in accordance with the known cortical
source and target layers of feedback projections, a coherence
was prominent between the deep layers of area 7 and
superficial layers of area 17 [8,32] (Figure 2a).
In conclusion, phase interactions seem to be informative
in understanding the functional significance [24,26] of the
a-frequency band oscillations. Observations of a-frequency
band phase correlations in task-relevant networks can all
be seen to support a direct role for a-frequency band
rhythmicity in the neuronal mechanisms of top-down
modulation and attention. However, in studies reporting
strengthened a-phase synchrony, some show a simul-
taneous amplitude increase [11,30,31], whereas others
show an associated amplitude suppression [9,10,12].
Thalamic burst discharges during a-frequency band
At the cellular level, the concept of a-frequency band
inhibition has been largely based on an association between
sleep-state a spindles and burst discharges of thalamocor-
tical neurons [6]. Thalamocortical relay neurons have long
been known to operate in two distinct modes: in a depolar-
ized state they tonically fire single spikes, whereas following
a period of hyperpolarization (70 mV), long enough to de-
inactivate T-type calcium channels, they discharge spike
bursts (interspike intervals 2–5 ms) riding on low-threshold
calcium potentials (LTCPs) at rates in the a-frequency band
or below [33,34]. Thalamic burst firing characterizes some
sleep stages and many pathological conditions, whereas
normal waking behavior was long thought to be exclusively
associated with single-spike firing. Moreover, single-spike
discharges relay sensory information with high fidelity,
whereas burst firing was historically thought to lead to
unreliable and imprecise information transmission [33–
35]. Hence, single-spike and burst-discharge modes had
been concluded to reflect ‘open and ‘closed’ thalamic gates,
respectively [35,36]. Among others, Pfurtscheller [6]
suggested that not just a spindles in sleep, but also large
a oscillations in the awake brain reflect a ‘closed thalamic
gate’, where no information is relayed to the cortex.
a-Frequency band oscillations in an awake brain are,
indeed, often associated with thalamic burst firing [37–39].
However, two recent discoveries overturn the notion of a
‘closed gate’. First, occurrence of LTCP bursts in an awake
brain is phenomenologically distinct from that during a
spindles or pathological rhythmic activities [34], and, even
in the bursting mode, thalamocortical neurons relay sen-
sory information efficiently. Whereas the firing rate in
tonic mode is linearly related to the strength of sensory
signals and is thereby well suited for detailed signal
analysis, the transmission in bursting mode takes place
in an all-or-none fashion and might be better suited for
signal detection [33,34]. Second, Hughes et al. [39] have
recently discovered a novel form of thalamic burst firing
producing a-frequency band rhythmicity, which could be
important for neocortical a
-frequency band oscillations. A
subset of thalamocortical neurons synchronized by gap
junctions discharge high-threshold (>55 mV) bursts of
spikes (interspike intervals >10 ms) following activation
by metabotropic glutamate receptors, and hence also fol-
lowing sufficient corticothalamic modulatory excitation.
Thalamic high-threshold bursting is synchronized with
EEG a oscillations [39,40], and it might be more prevalent
than LTCP bursting during in vivo a oscillations.
In addition to the reciprocal thalamocortical circuitry,
intracortical mechanisms [41] might also be crucial in the
generation of in vivo a oscillations (see also the study by
Nicolelis and Fanselow [34]). These neocortical circuits
might also be important for the interaction [11] of
a-frequency band oscillations with b- and g-frequency band
oscillations [42].
At the behavioral level, it has become well established
that large-amplitude awake-state a oscillations in the
primary sensory cortex are not associated with either a
‘closed thalamic gate’ or with inhibited or disengaged
cortical networks. The detection of somatosensory stimuli
by humans [23] and rats [43] is equally probable for both
small and large prestimulus a-frequency band oscillations
in the primary somatosensory cortex (Figure 1d). Taken
together, it seems that the available cellular-level data do
not suggest that the thalamocortical a-frequency band
oscillations would necessarily be associated with disen-
gaged cortical states or blocked thalamocortical signal
Phase reset of ongoing a oscillations
The classical event-related potentials (ERPs) are
influenced by both peristimulus amplitude and phase
dynamics (Box 1). In contrast with the classical view of
ERPs revealing stimulus-evoked components from ongoing
‘noise’, several human EEG studies suggest that some
early ERP components, N1 in particular, at least partly
emerge from a stimulus-caused phase reset of ongoing u
[44,45] and a oscillations [44–46] (Box 1, Figure Ib, c).
These findings place a large body of data on the N1
component into an interesting light. It has long been
known that the N1 component of human auditory [47]
and visual [48] ERPs, as well as the corresponding com-
ponent in rhesus monkey somatosensory [49] ERPs, is
robustly correlated with conscious detection. Also, the
enhancement of N1 by attention is historically well known.
Moreover, the principal sources of monkey somatosensory
N1 are in the superficial layers of the primary somatosen-
sory cortex, where it is produced by top-down inputs from
higher-order areas, such as the secondary somatosensory
cortex [50], which is in good agreement with the data of von
Stein et al. [8] and the concept of a having a top down role.
Although the contribution of phase reset in the generation
of N1 still warrants further consolidation, these data sup-
port the role of a-frequency band phase dynamics in atten-
tion and consciousness (see also [8,12]).
Phase synchrony among a, b and g oscillations
In recent years, b- and g-frequency band oscillations have
attracted widespread interest. Transient synchronization
of neuronal activity seems to be a key mechanism in the
TRENDS in Neurosciences No.x
TINS-503; No of Pages 9
Please cite this article in press as: Palva, S. and Palva, J.M., New vistas for a-frequency band oscillations, Trends Neurosci. (2007), doi:10.1016/j.tins.2007.02.001
Page 6
binding of anatomically distributed feature processing into
coherent perceptual objects, where it is often associated
with b or g oscillations [24]. Also, the phenomenology of g
oscillations in the human EEG is in line with a role in the
formation of neuronal object representations [51], and,
accordingly, g oscillations are present during the WM
retention period [52], modulated by attention [53–55]
and correlated with conscious perception [56,57]. von Stein
[8] proposed that, in contrast with a top-down role for
a-frequency band synchrony, g-frequency band synchrony
would be significant in bottom-up processing.
Thus, it seems that a-, b- and g-frequency band
oscillations all contribute to the neuronal underpinnings
of attention, WM and consciousness. The mechanisms that
integrate such spectrally distributed processing are likely
to involve cross-frequency interactions [26,58]; however,
the latter are a largely uncharted field. In a recent article,
we examined the possibility that cross-frequency phase
synchrony between distinct frequency bands could be used
to integrate spectrally distributed processing. Indeed, we
found that 1:2- and 1:3-phase synchrony (Figure 2b) was
present among cortical oscillations throughout the fre-
quency range from the d- (1–4 Hz) to the g-frequency band
[11]. Moreover, we found that a WM-intensive mental
calculation task enhanced cross-frequency phase syn-
chrony between globally synchronous networks in a- and
b-frequency bands, as well as between a and more local g
oscillations (Figure 2b). Importantly, in these data, the
strength of 1:3 ga-phase synchrony was positively corre-
lated with WM load. In other investigations, 1:2-phase
synchrony between b and a oscillations [59], as well as
verbal memory load-dependent 1:2 synchrony between a
and u (4–7 Hz) oscillations [60], has been observed in the
human EEG.
These findings thus support an idea that cross-frequency
phase synchrony might coordinate the integration of spec-
trally distributed neuronal processing [11]. However, the
findings of WM load-dependent cross-frequency phase
synchronization of global-scale a with u, b and g oscillations
support the notion that these oscillations cooperate in
Spectral integration and coordination might also involve
nested oscillations [61–63], which might, in addition to
cross-frequency phase synchrony, be important in many
cognitive functions. In nested oscillations, the amplitude,
but not the phase, of the faster oscillation is modulated by
the phase of the slower oscillation. Lisman, Idiart and
Jensen [64,65] have suggested that nested u and g oscil-
lations underlie the retention and capacity limits of WM.
VanRullen and Koch [66] postulated that nested a and g
oscillations produce discrete perception, so that g waves
constitute the contents of each ‘snapshot’, with the entire
percept being mediated by the a waves. Future research is
likely to reveal a fundamental role for hierarchical multi-
band oscillations and cross-frequency interactions in the
system-level mechanisms coordinating scattered neuronal
activity into perception, cognition and action.
Discrete perception and action
When visual objects are presented serially at fixation,
humans can recognize and categorize them at rates up
to 8–12 Hz [67,68]. In a continuous wagon wheel illusion
experiment, illusory reversals are most probable at wheel-
motion frequencies of 10 Hz, which is suggestive of dis-
crete perceptual sampling [69] (Figure 2c). Similarly, rats
sample and discriminate odors at a rate of 8 Hz [70] and
use active whisking at 7–14 Hz for tactile perception [34].
Hence, perception seems to operate in discrete ‘snapshots’
of 100 ms, which might correspond to consecutive cycles
of a oscillation [66,71]. Indeed, the perception of illusory
motion reversals is correlated with a-frequency band
amplitude in the ongoing EEG [72] (Figure 2c).
In parallel with perceptual sampling, smooth movements
are also realized in discrete steps (Figure 2d). Movement
discontinuities are caused by phasic muscular activity that
is 1:1 phase locked to cortical a-frequency band oscillations
[73]. The coordination of such discrete motion is achieved in
a large-scale cerebello-thalamo-cortical network that is
coupled through coherent a-frequency band oscillations
[73,74] (Figure 2d). Several lines of evidence thus link the
temporal segmentation of perception and action with
neuronal a-frequency band rhythmicity.
Discrete cognition: a oscillations in the ‘global neuronal
According to the current dogma of neuronal network
dynamics, synchronous g-frequency band assemblies
account for ‘active’ neuronal processing, whereas the roles
of a oscillations are in the inhibition and ‘inactivation’ of
task-irrelevant cortical regions. However, as discussed
here, an accumulating body of evidence emphasizes a
direct involvement of a oscillations in the mechanisms of
top-down modulation, attention and consciousness.
Neural correlates of consciousness (NCC) are widely
recognized to involve neuronal coalitions [71] and synchro-
nous assemblies [58], recurrent processing [75] and the
frontoparietal network [28]
. One conceptual model of NCC
is the ‘global neuronal workspace’ (GNW) [76,77], in which
sensory information enters awareness following an inter-
action between the sensory and frontoparietal network
[29]. It is important to see that the GNW model is
applicable not only in perception and attention, but also
in STM and movement execution.
We propose here a framework in which a subset of
neurons engaged in a oscillations belongs to the NCC
through an a-frequency band synchronized sensori-
fronto-parietal network that defines the GNW. This frame-
work is based on the following six lines of evidence: (i) top-
down modulation can be mediated by a-frequency band
phase interactions [8], and thus a rhythmicity could con-
tribute to recurrent processing [75] and top-down amplifi-
cation [29] ; (ii) a-frequency band oscillations can phase lock
between widely separated cortical regions [8,10,11,74] and
thus form functional large-scale networks [26]; (iii)
enhanced a-frequency band synchrony in the frontoparie-
tal network is associated with the execution of cognitive
tasks [10,11,74]; (iv) a-, b- and g-frequency band oscil-
lations coexist and are colocalized during stimulus proces-
sing and task execution [8,10–12,30,74,78]; (v) a
oscillations can be synchronized with u, b and g oscillations
in response to cognitive demands [11,60], which is likely to
be essential for the coordination and communication [25] in
TRENDS in Neurosciences No.x 7
TINS-503; No of Pages 9
Please cite this article in press as: Palva, S. and Palva, J.M., New vistas for a-frequency band oscillations, Trends Neurosci. (2007), doi:10.1016/j.tins.2007.02.001
Page 7
multiband networks; (vi) perception and action can proceed
in discrete ‘snapshots’ involving a-frequency band period-
icity [30,66,71,73,74]. This is natural or even mandatory if
a oscillations are involved in the GNW. Some predictions
associated with this framework are presented in Box 2.
In conclusion, an elucidation of the functional roles of a
oscillations seems to be mandatory to the understanding of
large-scale integration in the brain. Whether a oscillations
mediate idling, inhibition, attention, binding within the
GNW or any combination of these, remains a topic for years
to come.
This work was supported by the Academy of Finland and by the Ella and
Georg Ehrnrooth Foundation.
1 Berger, H. (1929) Uber das elektroenkephalogramm des Menschen.
Arch. Psychiatr. Nervenkr. 87, 527–570
2 Adrian, E.D. and Matthews, B.H. (1934) The interpretation of potential
waves in the cortex. J. Physiol. 81, 440–471
3 Ray, W.J. and Cole, H.W. (1985) EEG alpha activity reflects attentional
demands, and beta activity reflects emotional and cognitive processes.
Science 228, 750–752
4 Klimesch, W. (1996) Memory processes, brain oscillations and EEG
synchronization. Int. J. Psychophysiol. 24, 61–100
5 Pfurtscheller, G. (2001) Functional brain imaging based on ERD/ERS.
Vision Res. 41, 1257–1260
6 Pfurtscheller, G. (2003) Induced oscillations in the alpha band:
functional meaning. Epilepsia 44 (Suppl. 12), 2–8
7 Klimesch, W. et al. (2006) EEG alpha oscillations: the inhibition-timing
hypothesis. Brain Res. Brain Res. Rev. 53, 63–88
8 von Stein, A. et al. (2000) Top-down processing mediated by interareal
synchronization. Proc. Natl. Acad. Sci. U. S. A. 97, 14748–14753
9 Mima, T. et al. (2001) Transient interhemispheric neuronal synchrony
correlates with object recognition. J. Neurosci. 21, 3942–3948
10 Halgren, E. et al. (2002) Rapid distributed fronto-parieto-occipital
processing stages during working memory in humans. Cereb. Cortex
12, 710–728
11 Palva, J.M. et al. (2005) Phase synchrony among neuronal oscillations
in the human cortex. J. Neurosci. 25, 3962–3972
12 Palva, S. et al. (2005) Early neural correlates of conscious
somatosensory perception. J. Neurosci. 25, 5248–5258
13 Nikouline, V.V. et al. (2000) Dynamics of m-rhythm suppression caused
by median nerve stimulation: a magnetoencephalographic study in
human subjects. Neurosci. Lett. 294, 163–166
14 Makeig, S. et al. (2004) Mining event-related brain dynamics. Trends
Cogn. Sci. 8, 204–210
15 Hari, R. et al. (1997) Magnetoencephalographic cortical rhythms. Int. J.
Psychophysiol. 26, 51–62
16 Cooper, N.R. et al. (2003) Paradox lost? Exploring the role of alpha
oscillations during externally vs. internally directed attention and the
implications for idling and inhibition hypotheses. Int. J. Psychophysiol.
47, 65–74
17 Cooper, N.R. et al. (2006) Investigating evoked and induced
electroencephalogram activity in task-related alpha power increases
during an internally directed attention task. Neuroreport 17, 205–208
18 Jensen, O. et al. (2002) Oscillations in the alpha band (9–12 Hz)
increase with memory load during retention in a short-term
memory task. Cereb. Cortex 12, 877–882
19 Busch, N.A. and Herrmann, C.S. (2003) Object-load and feature-load
modulate EEG in a short-term memory task. Neuroreport 14, 1721–
20 Sauseng, P. et al. (2005) EEG alpha synchronization and functional
coupling during top-down processing in a working memory task. Hum.
Brain Mapp. 26, 148–155
21 Worden, M.S. et al. (2000) Anticipatory biasing of visuospatial attention
indexed by retinotopically specific alpha-band electroencephalography
increases over occipital cortex. J. Neurosci. 20, RC63
22 Thut, G. et al. (2006) Alpha-band electroencephalographi c activity over
occipital cortex indexes visuospatial attention bias and predicts visual
target detection. J. Neurosci. 26, 9494–9502
23 Linkenkaer-Hansen, K. et al. (2004) Prestimulus oscillations enhance
psychophysical performance in humans. J. Neurosci. 24, 10186–10190
24 Singer, W. (1999) Neuronal synchrony: a versatile code for the
definition of relations? Neuron 24, 49–65 111–125
25 Fries, P. (2005) A mechanism for cognitive dynamics: neuronal
communication through neuronal coherence. Trends Cogn. Sci. 9,
26 Varela, F. et al. (2001) The brainweb: phase synchronization and large-
scale integration. Nat. Rev. Neurosci. 2, 229–239
27 Kastner, S. and Ungerleider, L.G. (2000) Mechanisms of visual
attention in the human cortex. Annu. Rev. Neurosci. 23, 315–341
28 Rees, G. et al. (2002) Neural correlates of consciousness in humans.
Nat. Rev. Neurosci. 3, 261–270
29 Dehaene, S. et al. (2006) Conscious, preconscious, and subliminal
processing: a testable taxonomy. Trends Cogn. Sci. 10, 204–211
30 Gail, A. et al. (2004) Perception-related modulations of local field
potential power and coherence in primary visual cortex of awake
monkey during binocular rivalry. Cereb. Cortex 14, 300–313
31 Hanslmayr, S. et al. (2005) Visual discrimination performance is
related to decreased alpha amplitude but increased phase locking.
Neurosci. Lett. 375, 64–68
32 Siegel, M. et al. (2000) Integratin g top-down and bottom-up sensory
processing by somato-dendritic interactions. J. Comput. Neurosci. 8,
33 Sherman, S.M. (2001) Tonic and burst firing: dual modes of
thalamocortical relay. Trends Neurosci. 24, 122–126
34 Nicolelis, M.A. and Fanselow, E.E. (2002) Thalamocortical
optimization of tactile processing according to behavioral state. Nat.
Neurosci. 5, 517–523
35 Lopes da Silva, F. (1991) Neural mechanisms underlying brain waves:
from neural membranes to networks. Electroencephalogr. Clin.
Neurophysiol. 79, 81–93
36 Steriade, M. and Llinas, R.R. (1988) The functional states of the
thalamus and the associated neuronal interplay. Physiol. Rev. 68,
37 Guido, W. and Weyand, T. (1995) Burst responses in thalamic relay
cells of the awake behaving cat. J. Neurophysiol. 74, 1782–1786
38 Fanselow, E.E. et al. (2001) Thalamic bursting in rats during different
awake behavioral states. Proc. Natl. Acad. Sci. U. S. A. 98, 15330–
39 Hughes, S.W. et al. (2004) Synchronized oscillations at alpha and theta
frequencies in the lateral geniculate nucleus. Neuron 42, 253–268
40 Hughes, S.W. and Crunelli, V. (2006) Just a phase they’re going
through: the complex interaction of intrinsic high-threshold bursting
Box 2. Hypotheses derived from the framework advanced in
this article
Frontoparietal a synchrony should be found during focused
attention, WM and STM, as well as during conscious perception,
cognition and action, but it sho uld not phase lock to the stimulus
processing when the stimu li are subliminal, unattended or are not
consciously perceived for other reasons.
Those a-frequency band oscillations that are not phase locked to
the synchronous frontoparietal network do not belong to the
GNW and thereby might have different functional roles, such as
idling or inhibition.
The interaction between the GNW and the neuronal object
representations should influence the dynamic structure of these
representations according to variations in the cognitive context
(e.g. anaesthetized or awake; attended or ignored; one or multiple
perceptual objects).
a Oscillations and ag interactions should be related to the
attentional enhancement of g-frequency band synchrony.
Cross-frequency phase synchrony between the frontoparietal a
network and locally synchronous g-frequency band assemblies
underlies the selection and inclusion of neuronal object repre-
sentations into the focus of attention and sensory awareness.
TRENDS in Neurosciences No.x
TINS-503; No of Pages 9
Please cite this article in press as: Palva, S. and Palva, J.M., New vistas for a-frequency band oscillations, Trends Neurosci. (2007), doi:10.1016/j.tins.2007.02.001
Page 8
and gap junctions in the generation of thalamic alpha and theta
rhythms. Int. J. Psychophysiol., DOI: 10.1016/j.ijpsycho.2006.08.004
41 Silva, L.R. et al. (1991) Intrinsic oscillations of neocortex generated by
layer 5 pyramidal neurons. Science 251, 432–435
42 Roopun, A.K. et al. (2006) A beta2-frequency (20–30 Hz) oscillation in
nonsynaptic networks of somatosensory cortex. Proc. Natl. Acad. Sci.
U. S. A. 103, 15646–15650
43 Wiest, M.C. and Nicolelis, M.A. (2003) Behavioral detection of tactile
stimuli during 7–12 Hz cortical oscillations in awake rats. Nat.
Neurosci. 6, 913–914
44 Makeig, S. et al. (2002) Dynamic brain sources of visual evoked
responses. Science 295, 690–694
45 Gruber, W.R. et al. (200 5) Alpha phase synchronization predicts P1 and
N1 latency and amplitude size. Cereb. Cortex 15, 371–377
46 Naruse, Y. et al. (2006) Influence of seamlessnes s between pre- and
poststimulus alpha rhythms on visual evoked potential. Neuroimage
32, 1221–1225
47 Parasuraman, R. and Beatty, J. (1980) Brain events underlying
detection and recognition of weak sensory signals. Science 210, 80–83
48 Pins, D. and Ffytc he, D. (2003) The neural correlates of conscious
vision. Cereb. Cortex 13, 461–474
49 Kulics, A.T. (1982) Cortical neural evoked correlates of somatosensory
stimulus detection in the rhesus monkey. Electroencephalogr. Clin.
Neurophysiol. 53, 78–93
50 Cauller, L. (1995) Layer I of primary sensory neocortex: where top-
down converges upon bottom-up. Behav. Brain Res. 71, 163–170
51 Tallon-Baudry, C. and Bertrand, O. (1999) Oscillatory gamma activity
in humans and its role in object representation. Trends Cogn. Sci. 3,
52 Tallon-Baudry, C. et al. (1998) Induced gamma-band activity during
the delay of a visual short-term memory task in humans. J. Neurosci.
18, 4244–4254
53 Steinmetz, P.N. et al. (2000) Attention modulates synchronized
neuronal firing in primate somatosensory cortex. Nature 404, 187–190
54 Fries, P. et al. (2001) Modulation of oscillatory neuronal
synchronization by selective visual attention. Science 291, 1560–1563
55 Bauer, M. et al. (2006) Tactile spatial attention enhances gamma-band
activity in somatosensory cortex and reduces low-frequency activity in
parieto-occipital areas. J. Neurosci. 26, 490–501
56 Lutz, A. et al. (2002) Guiding the study of brain dynamics by using first-
person data: synchrony patterns correlate with ongoing conscious
states during a simple visual task. Proc. Natl. Acad. Sci. U. S. A.
99, 1586–1591
57 Meador, K.J. et al. (2002) Gamma coherence and conscious perception.
Neurology 59, 847–854
58 Engel, A.K. and Singer, W. (2001) Temporal binding and the neural
correlates of sensory awareness. Trends Cogn. Sci.
5, 16–25
59 Nikulin, V.V. and Brismar, T. (2006) Phase synchronization between
alpha and beta oscillations in the human electroencephalogram.
Neuroscience 137, 647–657
60 Schack, B. et al. (2005) Phase synchronization between theta and upper
alpha oscillations in a working memory task. Int. J. Psychophysiol. 57,
61 Chrobak, J.J. and Buzsaki, G. (1998) Gamma oscillations in the
entorhinal cortex of the freely behaving rat. J. Neurosci. 18, 388–
62 Canolty, R.T. et al. (2006) High gamma power is phase-locked to theta
oscillations in human neocortex. Science 313, 1626–1628
63 Vanhatalo, S. et al. (2004) Infraslow oscillations modulate excitability
and interictal epileptic activity in the human cortex during sleep. Proc.
Natl. Acad. Sci. U. S. A. 101, 5053–5057
64 Lisman, J.E. and Idiart, M.A. (1995) Storage of 7 +/ 2 short-term
memories in oscillatory subcycles. Science 267, 1512–1515
65 Jensen, O. and Lisman, J.E. (2005) Hippocampal sequence-encoding
driven by a cortical multi-item working memory buffer. Trends
Neurosci. 28, 67–72
66 VanRullen, R. and Koch, C. (2003) Is perception discrete or continuous?
Trends Cogn. Sci. 7, 207–213
67 Broadbent, D.E. and Broadbent, M.H. (1987) From detection to
identification: response to multiple targets in rapid serial visual
presentation. Percept. Psychophys. 42, 105–113
68 Chun, M.M. and Potter, M.C. (1995) A two-stage model for multiple
target detection in rapid serial visual presentation. J. Exp. Psychol.
Hum. Percept. Perform. 21, 109–127
69 VanRullen, R. et al. (2005) Attention-driven discrete sampling of
motion perception. Proc. Natl. Acad. Sci. U. S. A. 102, 5291–5296
70 Uchida, N. and Mainen, Z.F. (2003) Speed and accuracy of olfactory
discrimination in the rat. Nat. Neurosci. 6, 1224–1229
71 Crick, F. and Koch, C. (2003) A framework for consciousness. Nat.
Neurosci. 6, 119–126
72 VanRullen, R. et al. (2006) The continuous wagon wheel illusion is
associated with changes in electroencephalogram power at
approximately 13 Hz. J. Neurosci. 26, 502–507
73 Gross, J. et al. (2002) The neural basis of intermittent motor control in
humans. Proc. Natl. Acad. Sci. U. S. A. 99, 2299–2302
74 Pollok, B. et al. (2005) The cerebral oscillatory network associated with
auditorily paced finger movements. Neuroimage 24, 646–655
75 Lamme, V.A. (2006) Towards a true neural stance on consciousness.
Trends Cogn. Sci. 10, 494–501
76 Baars, B.J. (1988) A Cognitive Theory of Consciousness, Cambridge
University Press
77 Dehaene, S. et al. (1998) A neuronal model of a global workspace in
effortful cognitive tasks. Proc. Natl. Acad. Sci. U. S. A. 95, 14529–14534
78 Siegel, M. and Konig, P. (2003) A functional gamma-band defined by
stimulus-dependent synchronization in area 18 of awake behaving
cats. J. Neurosci.
23, 4251–4260
79 Traub, R.D. et al. (1998) Gamma-frequency oscillations: a neuronal
population phenomenon, regulated by synaptic and intrinsic cellular
processes, and inducing synaptic plasticity. Prog. Neurobiol. 55, 563–
TRENDS in Neurosciences No.x 9
TINS-503; No of Pages 9
Please cite this article in press as: Palva, S. and Palva, J.M., New vistas for a-frequency band oscillations, Trends Neurosci. (2007), doi:10.1016/j.tins.2007.02.001
Page 9
  • Source
    • "Increase in alpha power has been proposed to reflect either active processing related to the short term retention of information in working memory (Palva & Palva, 2007) or functional inhibition (Klimesh et al.,2007). The first view holds that increased alpha power represents an integral part of the distributed network activity related to the active processing of information in perceptual and cognitive tasks (Palva & Palva, 2007). The second view holds that such increases reflect the functional inhibition of cortical areas representing potentially disruptive task-irrelevant information (Klimesh et al., 2007, Jensen & Mazaheri, 2010 ). "
    Full-text · Article · Jul 2016 · International Journal of Cognitive Informatics and Natural Intelligence
    • "However, an alternative view is that our results reflect fluctuations in the weighting of priors on decision, rather than the prior probability itself. On this alternative view, alpha phase reflects the attentional state of the system, consistent with previous theoretical work ( Jensen, Bonnefond, & VanRullen, 2012; Palva & Palva, 2007), so that priors are assigned a greater weight on perceptual decision when sensory signals are expected to be unreliable. Here, perceptual expectations would increase or decrease the excitability of relevant neural populations or gain according to whether a target is expected to appear or not. "
    [Show abstract] [Hide abstract] ABSTRACT: Prior expectations have a powerful influence on perception, biasing both decision and confidence. However, how this occurs at the neural level remains unclear. It has been suggested that spontaneous alpha-band neural oscillations represent rhythms of the perceptual system that periodically modulate perceptual judgments. We hypothesized that these oscillations instantiate the effects of expectations. While collecting scalp EEG, participants performed a detection task that orthogonally manipulated perceptual expectations and attention. Trial-by-trial retrospective confidence judgments were also collected. Results showed that, independent of attention, prestimulus occipital alpha phase predicted the weighting of expectations on yes/no decisions. Moreover, phase predicted the influence of expectations on confidence. Thus, expectations periodically bias objective and subjective perceptual decision-making together before stimulus onset. Our results suggest that alpha-band neural oscillations periodically transmit prior evidence to visual cortex, changing the baseline from which evidence accumulation begins. In turn, our results inform accounts of how expectations shape early visual processing.
    No preview · Article · Apr 2016 · Journal of Cognitive Neuroscience
  • Source
    • "One of the most prominent coupling relationships we identified is delta–alpha. It reflects how delta activity, associated with deep dreamless sleep [55], influences the alpha oscillations which are said to reduce the information processing [55,56] and play a key role in consciousness [57,58]. During the maintenance of general anaesthesia, the alpha and delta activities were increased [2,59]. "
    [Show abstract] [Hide abstract] ABSTRACT: The precise mechanisms underlying general anaesthesia pose important and still open questions. To address them, we have studied anaesthesia induced by the widely used (intravenous) propofol and (inhalational) sevoflurane anaesthetics, computing cross-frequency coupling functions between neuronal, cardiac and respiratory oscillations in order to determine their mutual interactions. The phase domain coupling function reveals the form of the function defining the mechanism of an interaction, as well as its coupling strength. Using a method based on dynamical Bayesian inference, we have thus identified and analysed the coupling functions for six relationships. By quantitative assessment of the forms and strengths of the couplings, we have revealed how these relationships are altered by anaesthesia, also showing that some of them are differently affected by propofol and sevoflurane. These findings, together with the novel coupling function analysis, offer a new direction in the assessment of general anaesthesia and neurophysiological interactions, in general.
    Full-text · Article · Apr 2016 · Philosophical Transactions of The Royal Society A Mathematical Physical and Engineering Sciences
Show more