Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83; site oakhill.UUCP
Path: utzoo!linus!philabs!cmcl2!seismo!ut-sally!oakhill!davet
From: davet@oakhill.UUCP (Dave Trissel)
Newsgroups: net.philosophy
Subject: Re: I would like to know.
Message-ID: <578@oakhill.UUCP>
Date: Fri, 1-Nov-85 21:46:57 EST
Article-I.D.: oakhill.578
Posted: Fri Nov  1 21:46:57 1985
Date-Received: Tue, 5-Nov-85 04:53:52 EST
References: <1138@decwrl.UUCP>
Reply-To: davet@oakhill.UUCP (Dave Trissel)
Organization: Motorola Inc. Austin, Tx
Lines: 16

In article <1138@decwrl.UUCP> williams@kirk.DEC (John Williams 223-3402) writes:

>About these psi tests: How is the input stimuli randomized?

One method is to use a random event generator (REG).  Some run off of a noise
diode and some from a radioactive source (Strontium 90 I think is one.) The
electronic noise types allow a very high rate of generation - one I was
tested on generated 10,000 events a second.  Experiments have even been done
which tie an atomic type into the electronic type with no difference in
results.

As far as simple decisions such as choosing randomly which condition will be
the target (i.e. on versus off light) a coin toss can be done or random table
consulted (in some random way, of course :-) .)

  --  Dave Trissel   {seismo,ihnp4}!ut-sally!oakhill!davet