ERPs and Attention (1) Flashcards
What did Hans Berger do in 1924?
- first to record EEG signals
How did people look at EEG signals in the beginning?
- looked at the wave pattern over time
- see changes in the wave pattern when sleeping for example
What does EEG stand for?
- electroencephalogram
Who conducted the first cognitive ERP study?
- Gray Walter and the CNV 1964
What was the first cognitive ERP study?
- it looked at ERPs for 4 conditions
- A: when you hear a click
- B: light flashes
- C: click followed by flashes
- D: click, followed by flashes and press a button
What was found in the first cognitive ERP study?
- when the participant needed to press a button after the ERP waveform differed as they anticipated the flashes
- there was no difference in stimuli, just the mental state/process
What is an ERP?
- looking at a specific point in time in response to stimulus or event
When looking at an ERP, which way is up? What is CNV?
- negative is up
- contingent negative variation
- contingent: dependent, negative: plotted up
What is the most important difference between EEG/ERPs and TMS?
- manipulation vs measurement
What is the instrument used for recording EEG?
- EEG cap with electrodes in specific places
- international 10-20 system
- place cap by measuring head and using bony bumps to mark certain places
What is the EEG lab on campus called?
- Brain and Cognition Event-related Potential Laboratory
What are strong signals caused by?
- eye blinks
- can also be from eye, neck or mouth movements
Where do ERPs come from?
- many orthogonal cortical pyramidal cells
- excitatory transmitter released on apical dendrites causes positive charges to flow into dendrites (net negative on outside)
- polarity reverses with inhibitory transmitter
What does the polarity at the scalp depend on?
- orientation of the cortical surface and the position of the reference electrode
In what conditions are scalp-related potentials possible?
- only for open field or layered structures with consistent orientations
- primarily cerebral cortex
- if a closed field, the neurons are all facing in different directions and the consequent electromagnetic fields will cancel out instead of summing
How are voltages spread out?
- voltages are spread through the brain by “volume conduction”
- skull causes lateral spread
- measured as positive in direction of flow and negative in opposite direction of flow
- not much recorded where positive and negative meet
How do we get ERPs from EEG?
- stimulus is presented repeatedly and EEG averaged over all trials
- consistent peaks and troughs come through that are associated with the stimulus
What are the ERPs “locked” to?
- stimulus locked: line up on stimulus and then average
- response locked: line up on responses and then average
What is the temporal resolution of ERPs?
- about 1 ms (because recording 250 to 1000 per second)
What are some typical ERPs?
- brainstem responses
- mid-latency response
- long-latency responses
What is N170?
- voltage reflects face-related activity plus everything else that is active at 170 ms
- difference reflects only brain activity that differentiates between faces and cars
What is the difference wave?
- difference wave shows what is changing in brain activity between two different types of trials
- can sometimes reveal the time course of the underlying component
What is the superposition problem?
- the voltage at an electrode at time t is a weighted sum of all components that are active at time t
- this makes it difficult to determine underlying components from observed waveforms
- the recorded wave will look different depending on where we record it on the scalp
What is rule #1 about peaks a d components?
- peaks and components are not the same thing
- there is nothing special about the point at which the observed waveform reaches a local maximum
- peaks are maximums and minimums in observed waveform
- components are possible waves that made up the waveform
What is rule #2 about peaks and components?
- it is impossible to estimate the time course or peak latency of an underlying ERP component by looking at a single ERP waveform
- there may be no obvious relationship between the shape of a local part of the waveform and the underlying components
What is rule #3 of peaks and components?
- an effect during the time period of a particular peak may not reflect a modulation of the underlying component
- as in it may look like one component is being affected, when it’s really a different one
What might difference waves look like for different electrodes/areas of the brain?
- the waveforms will look very different bu the difference waves will look the same with different magnitudes
- this is because they are recording the same component from different locations
What is rule #4 of peaks and components?
- differences in peak amplitude do not necessarily correspond to differences in component size and differences in peak latency do not necessarily correspond to changes in component timing
Why is there an inverse problem (source localization)?
- we are using the resulting waveform to try to figure out the components that caused it
- there is no unique solution and an infinite amount of combinations are possible
- given noise, the correct solution may differ substantially
What does a single superficial dipole look like?
- creates relatively focused scalp distribution
- distribution changes if dipole is rotated or shifted
- easy to localize
What does a single deeper dipole look like?
- creates broader scalp distribution
- changes in dipole location have smaller impact on scalp distribution
- harder to localize
- hard to distinguish from broadly distributed superficial activity
What does two superficial dipoles look like?
- can still localize reasonably well with 2 dipoles if they are far apart and superficial
What does two collinear dipoles look like?
- looks very similar to single superficial dipole
- makes localization very difficult
How is ERP spatial resolution?
- as number of dipoles increases, it becomes more difficult to localize
- spatial resolution is poor and it is complex and hard to define
- it’s easier to say where it doesn’t come from
- computer programs will give best guess
What are some example ERP components?
- visual sensory
- auditory sensory
- P3 family
- mismatch negativity (MMN): flash red change to green
- language related (N400): hear ungrammatical sentence
- response-related (Lateralized readiness potential and error-related negativity: brain processing it made mistake)
What are two visual sensory responses?
- P1
- N1
What is P1?
- Extrastriate cortex
- sensitive to visual features, arousal and attention
What is N1?
- Extrastriate, parietal and frontal subcomponents
- sensitive to attention
- N170 is a subcomponent
What are later components associated with?
- higher levels of processing
What are some auditory sensory responses?
- brainstem auditory evoked responses: brief clicks, cochlea, auditory nerve and brainstem nuclei
- midlatency responses: medial geniculate nucleus and primary auditory cortex, modulated by highly focused attention
- auditory “long-latency” sensory responses: at least 3 N1 subcomponents, sensitive to attention
What is P3?
- P3 is a component that is much larger when a rare name is said (amplitude depends on probability of a task-defined stimulus category)
- processing that the category is different which occurs after earlier auditory processing
- latency is tied to amount of time required to perceive and categorize
What is mismatch negativity (MMN)?
- elicited by an auditory stimulus that physically mismatches preceding stimuli
- generated in auditory cortex
- appears to reflect comparison of incoming stimulus with echoic memory
- largely automatic but can be attentuated with strong attention
What is N400?
- negativity happening at about 400 ms after stimulus
- responds to semantic violations: word does not make sense in context
What is lateralized readiness potential?
- lateralized: only found in one side of the brain
- ex. if responding with right hand, see LRP in left hemisphere
- difference starts when you figure out what hand you’re going to use
- response occurs at peak
What is error-related negativity?
- flanker task: respond to central letter (XXOXX) or (XXXXX)
- more likely to make mistake when flanked by wrong letter
- the wave shows one initiating the response and realizing they’ve made a mistake as they’re responding
What are the risks to ERPs?
- almost non-existent
- discomfort from cap and gel
- need to remain still
What are ERPs good for?
- determine if an experimental manipulation influences process A or process B
- identifying multiple neurocognitive processes
- covert monitoring of processing (person may not be consciously aware of change but something still occured in brain)
When are ERPs not useful?
- want certainty of neuroanatomical location
- interested in slow activity (greater than 2 s) that is not time-locked to something
- cannot collect large numbers of trials
- process of interest does not have an ERP component
- subject frequently moves or needs to move for experiment (talking)
