The Times Higher Education
Scientists fail to see eye to eye over girl's 'X-ray vision'
Published: 10 December 2004
see also published comment
, concluding remarks
A Nobel prizewinning scientist has clashed with one of Britain's leading
experts on the paranormal in a row over the purported talents of a Russian
schoolgirl who claims she uses X-ray vision to diagnose medical problems.
Brian Josephson, a Cambridge University professor who won the Nobel Prize
for Physics in 1973, has given his backing to claims by Natasha Demkina
that she can see inside people's bodies.
The professor, who has been scorned by colleagues for his enthusiasm
for the paranormal, has claimed that an experiment for a forthcoming terrestrial
TV documentary that apparently disproves Ms Demkina's claims was "a fix"
designed to ensure she failed.
Richard Wiseman, professor of psychology at Hertfordshire University
and a key member of the respected Committee for the Scientific Investigation
of Claims of the Paranormal, who helped design the experiment, hit back
He said that the attack from Professor Josephson
who is a physicist with no known record of publishing on parapsychology
[but see below], "does not carry much weight", as it was posted on his personal
website without any refereeing process.
But he admitted that the trial could have been improved.
The experiment to test Ms Demkina's claims was filmed for a Discovery
Channel documentary, The Girl with the X-ray Eyes, which is also due to appear
on Channel 4 next year.
In the programme, Ms Demkina correctly identifies the medical conditions
of four out of seven patients, and misdiagnoses three.
Professor Josephson says, on his Cambridge University-hosted website,
that "many viewers ended up with a strong impression that the test... had
been deliberately set up with a view to ensuring that she would fail it".
He says that, in difficult circumstances, Ms Demkina overcame odds of
more than 50 to one to correctly diagnose four patients.
"Surely a case for celebrating Natasha's success?" he says.
But instead, it was declared that Ms Demkina had failed the test as the
experimenters had agreed with her ahead of the test that anything fewer
than five matches did not "support any belief in her claimed abilities".
Professor Josephson said: "A statistically very significant result was
obtained... but the experimenters concealed the fact with their talk of
her failing the test."
Professor Wiseman said he agreed that the number of matches Ms Demkina
had to achieve to be deemed a success was set higher than the standard
probability of one in 20 normally used in psychology tests.
"We were asking her to jump high, but that was because her claim would
present a huge challenge to science if it were true.
"I don't see how you could argue there's something wrong with having
to get five out of seven when she agrees with the target in advance."
He added: "I'm not saying that this experiment was perfect or that all
Professor Josephson's comments are wrong - like any first-time study conducted
under the pressure of time and with limited resources it could be improved
- but overall I think the results give us an additional insight into Natasha's
The following three letters
in response to the above article appeared in the December 17th. issue of
the Times Higher Education Supplement, in some cases in an edited form:
Your account of the controversy about the girl with "X-ray vision"
(December 10) was so obviously biased that I wondered why your reporter felt
the need to display his prejudices.
He said Brian Josephson is "scorned by colleagues for his enthusiasm
for the paranormal", whereas he called Richard Wiseman a key member of the
"respected" Committee for the Scientific Investigation of Claims of the
Paranormal. He could have said with at least as much accuracy that CSICOP
is an ideologically motivated debunking organisation and that Josephson
is much respected for his courage and independence.
It would have been more respectful of readers' diverse views simply to
report the controversy without taking sides
Given his propensity for talking to the press about his own unrefereed
work, it is ironic to find Richard Wiseman dismissing my web commentary
on his research on account of it being unrefereed (Scientists fail to see
eye to eye over girl's 'X-ray vision', December 10th.). It is precisely
because of his tendency to ask subjects to 'jump high' and then to announce
that they have failed, as well as his carelessness in experimental design
(in this case, some of the medical conditions of the subjects were ones
that Natasha Demkina had in the past indicated were ones she had difficulties
with), that parapsychologists are doubtful about Wiseman's approach.
Simply to have stated that the investigation had not definitively confirmed
Natasha's claims would have been unproblematic, but the Discovery programme
showed the investigators also drawing the further, scientifically unwarranted
conclusion that they had actually refuted her claims.
I have reviewed Professor Josephson’s
arguments, published on his web page, and find them to be scientifically
and statistically correct. In contrast, the statement of Professor
Wiseman, of CSICOP, “I don’t see how you could argue there’s anything wrong
with having to get five out of seven when she agrees with the target in advance”,
demonstrates a complete lack of understanding of how experimental data should
be interpreted statistically, as pointed out by Professor Josephson in his
The experiment is woefully inadequate
in many ways.
The chance of the observed 4 successes
7 subjects by pure guessing, is 1 in 78, an indication of a significantly
non-random result, as claimed by Professor Josephson. But, suppose
Natasha had a diagnosis rate of 1 in 2, compared with the chance rate of 1
in 7: then there is equal chance of getting 4 or more from 7, or 3 or less
from 7. That is, the probability (power) of detecting a true 50% diagnosis
rate on 7 subjects using a 0.01 significance level is only 50% (calculations
use a Binomial approximation). There should have been at least 21 subjects
to ensure a 90% probability/power of detecting a true diagnosis rate of
50% (using a 0.01 significance level test). Only if Natasha had a
true diagnosis rate as high as 72% would there have been a 90% chance of
detecting the effect using a 0.01 test (i.e. 4 more as the criterion) on
7 subjects. The experiment, as designed, had high chances of failing to detect
important effects. This may have been due to the lack of involvement
of a statistician in the design of the CSICOP experiment, rather than its
Finally, the lack of blinding of Natasha
to the sex, age, and state of the subjects means that no inferences are possible
about the cause of any increased diagnosis rate that might have been detected
- from Keith
Rennolls, Professor of Applied Statistics, University of Greenwich:
And now some concluding comments. The fact that Wiseman's main
reaction to the critique of his work on my web pages was to suggest that,
since it has not been refereed, it "does not carry much weight", may suggest
to some (a) that there is little he can find to say in defence of his position,
and (b) the name of the game is indeed propaganda (action directed primarily
at putting the other side in a bad light), as opposed to proper science.
Admittedly, these web pages here are also propaganda, but I hope that the
science and the logic are sounder.
In an email, Wiseman explained that he thought it important that the
public should understand about peer-review. I equally think (as noted)
that it is important that people understand CSICOP better. Seeing
how Wiseman chooses to respond to my critique may help them do this.
Notice the comment "I don't see how you could argue there's something
wrong with having to get five out of seven when she agrees with the target
in advance." I have already indicated on the web page what is wrong
with this, thus: 'real science does not work on a basis of getting someone
to sign their agreement to a long list of conditions, then later coming back
saying "this is what you signed; the challenge goes to us!" '. How pleasant
just to be able to ignore such arguments, on the basis of an assertion to
the effect that the web page they were brought up on had not been subjected
By the way, with regard to factual matters, the number of colleagues
who "scorn" me for my "enthusiasm for the paranormal" can be counted on
the fingers of one hand. And I have published papers on parapsychology,
some of which can, with a little effort, be found on, or via, my home page