EEG Alpha Oscillations
The inhibition-timing hypothesis
Brain neuroimaging methods
•
•
•
Direct electrode recording
– Recorded by electrodes penetrating the cortex
– Best spatial & temporal resolution
– Highly invasive
Electrocorticography (ECoG)
– Recorded from the surface of Cortex
– Less invasive than direct recording
– High temporal resolution
– Better spatial resolution than EEG
fMRI
– Measures hemodynamic response in the brain
– Millimeter spatial resolution
– Poor temporal resolution
– High imaging costs
•
Near-infrared spectroscopy (NIRS)
– Measures blood oxygen variation
– High temporal and spatial resolution
•
Magnetoencephalography (MEG)
– Measures magnetic field produced by
brain electrical currents
– Non-invasive
– Better spatial resolution than EEG
• Electroencephalography (EEG)
– Measures brain electrical activity
from scalp surface
– High temporal resolution
– Low cost, availability
– Non-invasive
EEG frequency bands
Delta
Up to 4Hz
Sleep
in babies
Has been found during some continuous
attention tasks
Theta
4 – 7 Hz
-young children
-drowsiness or arousal in older children
and adults
-idling
- Associated with inhibition of elicited
responses
Alpha
8 – 12 Hz
-relaxed/reflecting
-closing the eyes
-inhibition/timing hypothesis
Beta
12 – 30 Hz
alert/working
active, busy or anxious thinking,
active concentration
Gamma
30 – 100+ Hz
Displays during cross-modal sensory
processing
Also is shown during short term memory
matching of recognized objects, sounds,
or tactile sensations
Event-Related Potential
• Obtained by averaging many trials, time-locked to the stimuli
• Good temporal resolution
Alpha Oscillations
•
•
•
•
Brain wave first observed by Berger
Has the strongest power in awake EEG
Suppressed by eye opening
Conventional theory
– Mostly observed during idling
– Solely observed as Event-Related Desynchronization (ERD)
The inhibition-timing hypothesis
• α can respond with increase in amplitude (ERS):
– When subject controls the execution of a response
• When subject withholds the execution of a learned motor function
• In tasks in which the subject should block accessing to the past memory
– On sites that probably are under or exert top-down control
• In brain areas that are not task relevant
• Traveling alpha waves
• Oscillatory activity
– is induced by inhibitory cells (not a compulsory assumption)
– reflects phases of maximal and minimal inhibition
• Timing
– Rhythmic oscillations provide limited time window for firing for many
neurons
• Excitatory postsynaptic potential (EPSP): a temporary depolarization of
postsynaptic membrane potential caused by the flow of positively charged
ions into the postsynaptic cell that increases the chance a future action
potential will occur in a postsynaptic neuron
• Inhibitory postsynaptic potential (IPSP): a synaptic potential that
decreases the chance that a future action potential will occur in a
postsynaptic neuron
• Inhibitory Cells: Cells that release GABA (as the main inhibitory
neurotransmitter)
• Excitatory Cells: Cells that release Glutamate (as the main excitatory
neurotransmitter)
- Only 15% of cortical cells are GABAergic (inhibitory)
- Only 10% of synchronously active neurons are capable of generating an
amplitude which is about 10-times of the amplitude of non-synchronized
neurons
Inhibition-timing hypothesis
• Oscillatory activity reflects the changes
between phases of maximal and minimal
inhibition
• Oscillations provide a small time window
for firing of many neurons
Tonic firing
Rhythmic firing
Inhibition
Synchronize alpha activity can be seen In tasks where a learned
response must be withhold
ERS during Encoding
and retention
ERD during
retrieval
Upper alpha
peak during
retention
• ERS reflects inhibitory topdown control during retention
Modified Sternberg task
ERS when a learned motor task must be withhold
• Experiment by Hummel
1. Subjects trained for a finger movement task
2. In ACT condition the subject should look at the cue and perform the
movement
3. In INH condition the subject should look at the cue and withhold performing
the movement
–
–
–
–
–
ERD during ACT condition
ERS observed during INH condition over the same area
ERS in INH was observed only when subjects were trained for the motor task
Blood Oxygenation Level Dependent (BOLD) signal decreases during INH
ERS exhibits top-down control over accessing the learned memory trace
Tonic alpha power, cognitive performance and inhibition
• The extent of ERD also depends on resting alpha power
• High resting alpha power and high ERD during a task is associated with
good performance in memory task
• Perception performance is enhanced if the cortex is already activated (low
resting alpha power)
• Memory performance is enhanced if the cortex is deactivated before a
task (high resting alpha power)
• High level of cortical excitation (low resting alpha power) is helpful to
analyze a visual input
• High level of initial excitation is detrimental when a high selectivity is
needed for accessing a memory trace
Intelligence and the inhibition of task-irrelevant brain areas
Grabner experiment:
Taxi drivers were asked to indicate which place the would pass when driving
from A to B
1. Based on their knowledge of the city roads (Long-Term Memory)
2. Based on a fictional map they had to learn for the task (Working memory)
1.
2.
A very low ERD was observed for subjects with lower IQ whereas a large and
focused ERD for higher IQ
Large and widespread ERD for lower IQ verses a focused ERD and a large
widespread ERS for higher IQ
More intelligent subjects may only use the brain areas that are directly relevant
LTM and WM tasks
Top down and bottom up processing
•
Top down processing can be seen as processing what one is perceiving using past
information. It occurs when someone infers from a generalization, law etc. to
conclude something about a particular example, instance, case etc.
•
Bottom up processing can be seen as starting with no knowledge on a subject. It is
said to occur when one draws generalizations from particular examples, instances,
cases etc. to capture commonalities between them.
Top-down processing
Stroop test
ERS during inhibition of semantic processing
• Read aloud the color of each word below:
Your brain alpha power might increase during reading each color in order
to block the interference from the semantic processing of each word
Inhibition and top-down control
•
Top-down process as an attention control function that keeps focused on highly
selective aspects of task performance by using inhibition to prevent interference
from task-irrelevant brain areas or processing system.
•
The internal control of attention – particularly under situations when sensory
information must not be processed – is associated with an increase in alpha
activity(ERS).
•
with human subjects observing another person doing a complex motor task, ERD
over sensory-motor areas but ERS over the Supplementary Motor Area can be
observed
•
in a WM task, Sauseng et al. (2005b) found upper alpha ERS when subjects had to
manipulate information during a retention interval.
Sauseng Experiment:
• Subjects see a 4x4 matrix with some colored squires
• Two tasks were compared:
1. If the color of squires was red subjects should mirror the image around the
vertical line (manipulation) and after retention period compare it with the
probe item
2. If the color of squires was green subjects were required just to remember the
original image and after retention period compare it with the probe item
– During Manipulation period upper alpha ERS at frontal sites and ERD at
posterior sites was larger than during pure retention
– Absolute alpha power, showed that during manipulation brain activation reflects a state
of “alpha equilibrium”—where alpha power exhibits a topographically flat distribution
– This effect was not only found with alpha power but also with alpha frequency. Alpha
frequency at frontal sites increased to values similar to those at occipital sites during
manipulation. In contrast, during pure retention occipital alpha was clearly faster than
prefrontal alpha
Sauseng Test:
Top-down
processing
- It appears that a state of
alpha equilibrium enhances
or enables cortico-cortical
communication.
Because
during manipulation frontal
sites are leading and
posterior sites trailing (Fig C),
we conclude that anterior
sites control in a top-down
manner mental operations
on the memory trace stored
at posterior sites.
Alpha and metabolic rate
•
large negative correlations between alpha power (at 8–12 Hz, recorded from O1
and O2) and the fMRI– BOLD signal in cortical areas
•
In some areas, however, positive correlations were also obtained.
•
It should be noted that the interpretation of correlations between alpha and
metabolic rate is problematic because:
– Large alpha amplitudes may stem from high synchrony between only a few percent of
neurons with the majority of neurons being relatively inactive and exhibiting only a low
firing rate thereby producing minimal metabolic activity.
– Decreasing synchrony but increasing EPSP amplitudes (leading to increased firing rate) in
a larger number of neurons may produce small EEG amplitudes and a large metabolic
signal.
Timing hypothesis
• Two effects:
1.
In general, decrease in firing rate
•
Reflects inhibition
2. Increase in rhythmic discharges
•
Provides limited time window for firing of neurons
–
–
–
An increase in inhibition (driving an oscillation) is accompanied by an
increase in oscillatory activity that results in a stricter timing of neural
activity.
oscillations are an extremely useful mechanism to control the time window
in which neurons are most likely to fire.
In a complex network, common target cells will tend to receive neural
activity synchronously. This increases the likelihood that these cells will be
activated selectively and that one brain region might influence another.
Alpha phase controls cortical excitability
• Varela’s experiment:
– Subjects were asked to judge whether the two (shortly displayed) stimuli are
simultaneous or moving
– It takes 50ms (due to peripheral transmission time) for a visual stimulus to be
processed at the cortex
– The stimuli which are displayed during positive alpha cycle perceived as
moving. (processed after 50ms when the alpha is in the negative cycle and the
cortex is in the more excitatory state, so facilitating stimulus processing)
– The stimuli which are displayed during negative alpha cycle perceived as
simultaneous (processed after 50ms when the alpha is in the positive cycle
and the cortex is in the more inhibitory state, hence delaying stimulus
processing)
Alpha phase reset
•
•
•
•
•
•
alpha is capable of controlling the timing of cortical activation in the range of
milliseconds
the P1 is the earliest manifestation of a top-down process during early sensory
processing
Under conditions where sensory processing is guided by a specific expectancy, e.g.,
about the spatial location and/or type of stimulus, the P1 amplitude will be larger
than under conditions where specific expectancies are lacking.
the P1 is generated (at least in part) by a phase reset of alpha and that the surface
positivity of this component represents the deactivating/inhibitory phase of alpha
In the time window of the P1 early semantic classification/identification processes
are initiated.
This time window is the earliest where top-down processing can interact with
bottom-up processing.
Hanslmayr’s experiment:
1. Briefly exposed letters “p” and “q” are shown in random followed by a pattern
mask
2. Subject were asked to detect either the letter was ‘q’ or ‘p’
•
•
•
compared to bad performers, good performers showed a larger P1-N1 complex
that was associated with a significantly larger alpha phase locking index (PLI).
good (target categorization/identification) performance is reflected by increased
alpha phase locking.
the amplitude of the P1 most likely is generated (at least in part) by a reset of
alpha
•
The lack of desynchronization
and the large differences in
prestimulus power between
good an bad performers is
clearly
evident
in
the
time/frequency/(wavelet)
power plot.
Some Conclutions:
•
during phase reset cortical excitability is reorganized in a way that those brain
areas that are task-relevant are set in a comparatively less excitatory or inhibitory
state at exactly that time window when bottom-up sensory processing interacts
with top-down processes.
•
Alpha phase reset may be understood as an early inhibitory filter that enables the
emergence of a highly selective and excitatory encoding network.
Schack’s experiment:
1. investigating alpha phase synchronization during the encoding of spoken words
that subjects were instructed to remember.
2. Four lists containing either 25 concrete or 25 abstract nouns were presented
•
•
•
•
•
topographical pattern of alpha phase synchronization showed large differences
between concrete and abstract words already in a very early time window of 100–
200 ms.
differences in alpha phase synchronization between word categories most likely
reflect differences in global categorization processes between abstract and
concrete words.
the speed of
traveling alpha waves was faster for abstract than concrete words. Because
abstract words are less numerous than concrete words, this may indicate that the
search area for abstract words can be narrowed down faster for abstract then
concrete words.
event-related alpha coherence emerges at the time window of the P1 and reflects
the timing of co-activation between brain sites during interactive top-down and
bottom-up processing
Summary
• Inhibition
– α can respond with increase in amplitude (ERS):
• When subject controls the execution of a response
• On sites that probably are under or exert top-down control
– Oscillatory activity
• is induced by inhibitory cells (not a compulsory assumption)
• reflects phases of maximal and minimal inhibition
• Timing
– Rhythmic oscillations provide limited time window for firing for many neurons
– Traveling alpha waves, cyclic changes in cortical excitability in the alpha
frequency range, alpha coherence, phase synchronization and phase locking
are considered manifestations of this timing mechanism.
– upper alpha oscillations are related to top-down processes in a complex
sensory-semantic LTM system that controls the access to and manipulation
with stored information.
– during the early period of stimulus encoding, phase-locked alpha reflects the
top-down influence of semantic categorization processes.
Thank you for your attention
Discussion