Recording
It amplifies the electrical signal detected by the electrodes and applies correction/filtering
What does the amplifier do?
How is ICA used to remove blinks and saccades?
It is a linear regression method that assumes pooled variance across experimental items and fits a line of best fit by finding the minimum squared residual variance between actual data points and that line.
What is ANOVA?
At the very least, 20 per experimental condition.
How many trials should each participant see?
EEG is the raw electrical recording, an ERP is the average amplitude of the EEG signal triggered by a particular stimulus, averaged over trials and sometimes participants.
What is the difference between EEG and ERP?
The resistance between electrode and scalp.
What is impedance?
To remove non brain-related frequencies.
What is filtering?
It is a linear regression method that can account for variance between and within (for example) subjects and/or items. It finds a line of best fit by minimising residual variance between that line and the actual data, adjusted for individual variance by item and/or subject.
What is a linear mixed effects model?
It depends: Effect sizes and variability estimates from previous experiments can be used to conduct a power analysis for your specific design.
How many subjects do I need?
Increased accessibility, promotion of diversity, and reproducibility are some examples.
What are the benefits of using open source software for EEG analysis and recording?
It suggests the participant is sleepy.
What does theta activity in the recording mean?
It's often used to distinguish stimulus-related activity from unrelated activity in the EEG.
What is baseline correction?
An effect size estimate and a measure of variability.
What is the minimum set of statistics that should be reported for an ERP analysis for each experimental effect?
Rapid serial verb presentation
How do we stop subjects making eye movements during reading experiments?
50 Hz.
What is the frequency of mains electricity in Germany?
Muscle movement, jaw clenching, blinks, and eye movements are examples.
What are some artifacts in the EEG recording?
Individual artifacts often look quite different in size/frequency and we don't want to reject data unnecessarily, especially if we have a small sample size.
Why might automatic artifaction sometimes not be a good idea?
The probability of observing the test statistic if the null hypothesis were true.
What is a p-value?
Measuring at different words, not controlling for frequency or neighbourhoods, and measuring at different word positions are examples.
What are some common design weaknesses in ERP experiments?
This type of error occurs when we falsely reject the null hypothesis; a "false positive"
What is a type two error?
Because these are bony or cartilegey areas that don't have muscle or blood vessels that might create noise which will make it hard to distinguish brain signal.
Why do we put the reference behind the ear or on the tip of the nose?
Because ICA can identify only a limited number of components and we don't want to waste those on large, obvious artifacts.
Why should we remove large muscle movements and other artifacts before conducting ICA?
Overestimating the effect size, finding the wrong sign on the effect size, and falsely rejecting the null are examples of this design flaw.
What can happen if we have too few trials or subjects?
To get a good estimate of between- and within-participant behaviour: For some ERP components (particularly "later" components), participants may show a lot of variability. Even the same participant may show variability between trials.
Why do we need so many trials per participant?
This type of error occurs when we overestimate the size of an effect due to insufficient data. [500 Bonus points: describe how this happens using the central limit theorem!]
What is a Type-M error?