And your assertion is a classic example of not reading "the strict requirements of the protocol", which stated:
"The Applicant claims to be able to detect which Subject in a group of six Subjects is missing a kidney, to further identify which kidney (left or right) is missing in her selected Subject, and to be able to do this with 100% accuracy in three consecutive trials." See
http://iigwest.org/anitaikonenprotocol.html
Note that the protocol does not specify a one-step process, along the lines of: "The Applicant claims to be able to detect which kidney (left or right) is missing from among a group of six Subjects, five of whom have two kidneys." Rather, it clearly specifies a two-step process: (1) "The Applicant claims to be able to detect which Subject in a group of six Subjects is missing a kidney," and, (2)
"to further identify which kidney (left or right) is missing in her selected Subject."
So, my probability calculation of P=.0567 is not at all a case of "taking experimental results and combing them for patterns outside the strict requirements of the protocol." Rather, it is the proper statistical way to evaluate her performance. The improper way is to pretend, contrary to the protocol, that Anita's test was a one-step process, and that her result produced a P of only .2297.
Now, it is true that Anita failed in her attempt "to do this with 100% accuracy in three consecutive trials", but it is perfectly valid to note that she performed barely outside the range of what is generally accepted to be statistically significant. Perhaps her performance was due to clues that she picked up or perhaps she just got lucky, but that's unknown at the moment.