[color=#000000]Just for reference, the latest CONN release which will be out before the end of this week also includes a template with a set of commonly-found ICA networks (these were computed from HCP 497-subjects ICA analyses, and they include default mode, salience, fronto-parietal, dorsal attention, language, visual, sensorimotor, and cerebellar networks) as well as a few streamlined procedures to more easily identify/label the data-driven networks resulting from your specific ICA analyses. [/color]
[color=#000000]Best[/color]
[color=#000000]Alfonso[/color]
[i]Originally posted by Shady El Damaty:[/i][quote]There are various variables that are generated -- ICA can extract spatial or temporal sources and both are available for analysis in conn. Spatial sources represent clusters of voxels that reliably activate together and temporal sources provide time-domain information for those clusters. It is difficult to say what networks they represent without looking at the data because every dataset will look a little different, however there are specific networks that you can expect to reliably extract across datasets.[/quote]
[color=#000000]Best[/color]
[color=#000000]Alfonso[/color]
[i]Originally posted by Shady El Damaty:[/i][quote]There are various variables that are generated -- ICA can extract spatial or temporal sources and both are available for analysis in conn. Spatial sources represent clusters of voxels that reliably activate together and temporal sources provide time-domain information for those clusters. It is difficult to say what networks they represent without looking at the data because every dataset will look a little different, however there are specific networks that you can expect to reliably extract across datasets.[/quote]