Double blinding techniques are designed to ensure that neither the patient nor the person administering the test knows whether the active treatment or the placebo is being used.

The benefit that comes from double blinding is that it helps to isolate the true value of the treatment from other factors such as bias and the placebo effect.

Double blinding is common place in latter stage clinical trials.

Double blinding is rare in the ad hoc clinical trials run by citizen scientists on themselves. This is to be regretted.

This thread focuses on ways in which citizen scientists experimenting on themselves and working on their own can double blind their white-ratting. In other words, it deals with situations where one person is both the subject and the test administrator.

Soluble substances

(This first appeared on 21st September in the thread http://neurotalk.psychcentral.com/sh...95#post1017095
I repeat it here in an amended form so that everything is in one place.)

1. you need a "sugar pill". For instance, 1 part apple cider vinegar and 9 parts orange juice is tested against 1 part malt vinegar and 9 parts orange juice. All cooled to 5 degrees C.

2. for each day the tester makes each of the two preparations, puts them in opaque, sealed bottles. He/she makes labels which are too complicated to easily notice the difference, e.g.
rr3eerere££RREReddffd6666u:;' and
rr3eerere££RREReddfdd6666u:;'
These are attached to the bottles and the relationship between the labels and the treatments is noted.

3. The bottles are jumbled up until the tester has lost track (except on paper) of which bottle is which.

4. the tester throws a dice to determine, which bottle to use.

5. the tester does not find out the contents of the bottles he/she took until the end of the trial.

Computer controlled potential therapies

This section applies to computer based potential therapies where the subject cannot feel the treatment and where a computer is programmed to:
- randomly make the decision of which therapy or placebo to administer;
- administer the therapy;
- measure the result of the therapy.
An example of an area which this approach could be used is for those types of transcranial stimulation in which the subject cannot feel the treatment.

The interesting point here is that even if the citizen scientist him/her-self writes the program, the results can appear random to them. This is easy to do by using random number generators. For instance, out of our sight a computer could throw a virtual dice then, according to the result, decide whether to use a potential treatment or use a sham treatment.

I'll be grateful for feedback, in particular:
- flaws in my logic breaking the claim that these protocols satisfy the requirements of double blindedness;
- double blind techniques in other circumstances.

John