Hello Dani,
I understand you're having trouble grasping the assumptions behind AMICA and binica. Both of these algorithms assume non-Gaussian independent components (ICs) and do not assume Gaussian ICs.
Here's a simplified explanation:
- AMICA (Adaptive Mixture ICA): AMICA assumes that the ICs in the signal are non-Gaussian. It's designed to handle non-Gaussian ICs and is more computationally intensive due to its adaptive nature. For more details, you can visit the [AMICA GitHub page](https://github.com/sccn/amica/wiki).
- binica (binary ICA): Despite what you might have heard, binica also assumes that the ICs are non-Gaussian, similar to AMICA. It's termed "binary" because it employs a binary method for IC separation. More information can be found on the [binica GitHub page](https://github.com/sccn/binica/wiki).
For additional functions and examples, you might want to check out the EEGLAB documentation. It offers further insights and practical examples for using ICA algorithms in electrophysiological data analysis. You can access it on [MathWorks File Exchange](https://www.mathworks.com/matlabcentral/fileexchange/eeglab).
I hope it helps!
