I just thought I should explain to SaskMick the rationale for the different parts of the tests that we have proposed to make it clear that we aren't just being resistant to his "proofs":
1. You need completely opaque containers that always look identical on the outside and give no external indication of what is inside and are not permanently linked to what is inside (this is one problem with using a painted interior in your test). Even subtle clues as to the contents need to be eliminated and nothing on the outside, even small imperfections, can tell you what is inside.
2. You need some objects that can be placed inside the containers without altering the container's outside properties and appearance in any way. Again, this is so even subtle clues on the outside are eliminated (smells, looks, stability, etc.). Ideally, the inside objects should not even touch the containers themselves.
3. To be extra safe, you need control objects about the same shape, color, and weight as the target so every container has a target or control object. I like coins, because they are very low heights, do not smell, can be of similar weights and colors, and similar sized coins look much alike so that there is no clue communicated to the outside. If American and Canadian pennies are hard to come by, try British currency instead. Again, the shape, weight, color, and thickness of the target and control coins should be much the same. One problem: currency is usually designed to feel and look different to avoid confusion, so you may need to use a foreign coin to approximate a British one. One established "magic" trick is to use film cans as the container and only fill one with water. Notably, just rocking the containers a little bit, by leaning on the table, will result in the one with water rocking much differently than those without. That is say all of the containers need controls inside that are very close to the properties of the target. You can even use paper currency (a 5 vs 10 pound note) and envelopes instead of buckets if the envelopes are completely opaque. Or use paper currency in envelopes in the buckets to be extra safe.
4. The person who sets up the target vs. controls should use a random means of selecting their locations (a dice toss, a random number, etc.). This is because people are known to choose numbers non-randomly if they are allowed to.
5. The person who sets up the target vs. the controls must leave the room during the test, and no one in the test room should know which is the target vs, controls location. This is to remove subconscious clues from the "people in the know" to the testee (catches of breadth, stares, subtle body language). In fact, even horses can be shown to pick up on these clues. Spouses are particularly good at picking up on these non-verbal clues.
6. Obviously you should be well out of visual and auditory range during the set up of the test. You may not know that you are using subtle information from the setup of the tests, but people do indeed.
How shall I phrase this in order not to get yellow carded?
