" I was pointing out that as soon as you involve a person in a study the study becomes flawed."
Yes, flawed, but unusable? The big science papers don't even allow NLP-papers anymore so we'll probably not know if there are newer studies of NLP in therapy settings which perhaps would show different results.
