Hello everybody.
Following the "simple" addiptive model of the EEG event-related signal (deterministic + zero-mean stocastic process), i am looking for a way to identify just the presence of an ERP inside the epoch of a averaged signal.
I thought that energy would be a good parameter: in the pre-stimulus block, it should be much lower than in the stimulus one. Computing it, such difference is extrimely low and it do not let me to compute a bootstrap analysis (and a hypotesis testing of equal energy in pre-stimulus and stimulus block).
Here is my question: what would be a good parameter to estimate just the presence of an ERP? I am thinking about something like the maximum or minumum value in each averaged block (pre-stimulus or stimulus). Could you also give me some references (papers) wich describes your ideas?
Let me introduce you to some articles. Contributors seriously analyze your question.
Hi,
You may want to consider switching to the time-frequency domain to do this. Using the (variance of) event related spectral perturbation (ERSP) and/or the inter trial coherence (ITC) instead of energy and/or maximal/minimal values of your ERP can probably help you.
Some other measures of the time–frequency transformed ERP such as the event related phase coherence (ERPCOH) can also be helpful.
You can compute all these with ERPWAVELAB, a toolbox for multi-channel analysis of time-frequency transformed event related potentials (Mørup M, Hansen LK, Arnfred SM. 2007).
Regards.
This is a well studied problem in the detection of Auditory Brainstem Responses (ABR, or BERA), lot of papers in the 1980's. The question is to find the best method to minimize the number of sweeps for averaging, to save time. See for example papers (and phd thesis) by De Weerd and colleagues.
- Badges
- Science topic
- Similar topics
- Automotive Engineering
- Machines