I have attended scientific experimental meetings organised by the Scottish Society for Psychical Research (SSPR)...
www.sspr.org.uk/Each member of the voluntary audience is seated randomly. This is all decided and carried out by the SSPR folk.
The Medium for the evening's event is chosen by the SSPR and not announced by name, nor are they ever within sight of the audience.
An occupied seat is randomly pre-chosen by the SSPR folk, is not announced and indicated by gesture.
The person in this seat is the object of the reading.
The Medium cannot know who they are reading for, eliminating 'cold-reading'.
The Medium is then asked to link with the higher side and state what evidence is apparently coming forward for the occupant in the chosen seat.
The recipient responds only by nodding - the SSPR folk verbally informing the Medium only if the response is affirmative or negative. The Medium works 'blind' throughout.
The information offered - statements, no questions allowed - is usually about 20-25 pieces of information, is recorded along with the responses.
Next, a copy of the statements is photocopied and each person present given a copy and asked to write their own response to the statements.
At the end of these proceedings, an informal show of hands (if time allows) show that only a few of the Mediums statements apply to only a few folk, but some 80 plus % apply to the recipient.
This type of test has been carried out in various countries among various people at different times.
The number crunching is also carried out by different specialised groups and is coded to preserve the triple-blind element.
Over many years the Mediumistic 'hit' rate is consistently in excess of 80%.
The Society is run under the direction of Professor Archie Roy, Emeritus Professor of Astronomy, Glasgow University.