I agree, but this is also what makes it difficult to do a placebo trial, as stated in the article.

They still need to find a method of creating a state where the subject doesn't know if they are getting ketamine or not.