My experiment includes two groups and two conditions, and I would like to compare between groups and conditions. To do so, I created a study design, uploaded the data sets and could make a comparison between groups and conditions at the amplitude level. My question is, can EEGLAB calculates the significant differences with ERP latency?
I don't think that EEGLAB is really set up to do this sort of analysis without some custom scripting on the user end. I would recommend downloading Steve Luck's ERPLAB plugin for EEGLAB which allows you to do this sort of analysis relatively easily within the GUI (using the jackknife technique or with the raw latencies). This will however require some reprocessing of the data on your part to conform to the way that ERPLAB groups events in its binlister to create files in its .erp format. Also, you will have to perform the statistical tests on these extracted latencies either in MATLAB or another statistical program on your own.
Thanks Brian Kraus for your answer. I can understand you; however, I have another related question. Can ERPLAB manage group level analysis (in my case, I have two groups, two tasks, and two conditions) and plot as EEGLAB-study design is doing? .. As far as i know, ERPLAB can be only used to plot one group and to compare between conditions (Bins), Did I miss something?
Yes, you are correct that ERPLAB is not set up to do plots of between group averages (without some user workaround such as adding events so you could create unique bins for each group). You would have to process the data with bins for each level of task/condition within each group and then plot the waveforms separately for each group. However, the time series of these averaged waveforms are pretty easily accessible with some basic scripting and you could either plot them in MATLAB with both groups in the same plot or export the data to a program like Excel if you felt more comfortable plotting them there.
You can give a try a Fieldtrip. As far as I remember it has the statistical analysis that you are looking for.
Also you can give a look to BESA Statistics (not free, but you can have fully working trial version for 1 month)
- Badges
- Science topic