Hi @rsmith,
I have completed a fixel-based analysis pipeline, incorporating both whole-brain analysis and two tracts of interest. At a threshold of 0.95, I identified only one or two significant fixels within my regions of interest. In this context, would such limited findings still be considered sufficiently meaningful to report as significant results? Any guidance would be greatly appreciated.
Hi @H.A.N.K,
Finding a handful of significant fixels is always suspicious. However to instruct on how to interpret / interrogate further / change one’s analysis pipeline requires a fairly large piece of work, and indeed involves a certain amount of criticism of the pipeline as currently established. I was hoping to defer addressing this until I could produce a more exhaustive substitute; but given the long-left-outstanding prevalence of negative results and concerns about statistical power in FBA I’m going to try to smash together a documentation page that might provide some additional instruction.
Regards
Rob
1 Like