In a few sentences please describe what the P300 wave and oddball paradigm is.
The P300 is a pronounced positive wave in neural activity arising around 300ms (or later according to Professor Luck) after a rare or unexpected stimulus is presented. An example of this could be a letter you were looking for finally arriving in a P300 speller. The oddball paradigm is the experimental structure that induces the P300, like the P300 speller where one stimulus is an 'oddball'.
Given this figure of a EEG recording, what can we say (with the most confidence) what we are looking at within the waveform and why?
Figure taken from Steve Luck’s “A Broad Overview of the Event-Related Potential Technique”, Figure 1.5.
The repeated peaks at 100ms intervals suggest the presence of a 10Hz alpha wave signal in our signal. This could imply that the participant has a calm and unfocused mental state.
What is an ERP, or an event related potential?
An ERP is a time specific neural response to a given stimulus. It is different from EEG in that EEG is a form of neural recording while an ERP is a systematic variation in electrical response to an event. An example of an ERP is the famous P300.
Does a given frequency band always reflect a specific process?
T / F
What are some artifacts that can appear in your EEG recording?
Outlet noise, blinking, eye movements, head movement, chewing, and unrelated mental states in the participant (drowsiness, inattentiveness).
What are some of the limitations of EEG and ERPs? (talk abt both separately)
ERPs are derived from EEG data, so weakness of EEG will be weaknesses of ERP. A weakness of both is that EEG has poor spatial resolution and is especially prone to noise. On top of this, it's difficult/impossible to record accurate data occurring deep within the brain as EEG only records the scalp. A weakness of ERP specifically is that is that not all mental processes have distinguishable ERP signatures, and ERP experiments won't be able to find a clear signal.
What is filtering and what does it do?
Filters are used to remove frequencies that likely aren't related to neural data. Examples of this are high frequencies, low frequencies, and specific sources of noise (60 Hz outlets). This can be done through high-pass, low-pass, and notch filters respectively.