Results


An overview of the results is given in table 2 (for raw data, click here).

Table 2. Results per condition.
date of callstime

feedback
condition

no. of hits
(each cell represents 10 binary guesses)

devz



R1 R2 R3 R4 R5 R6 R7 R8 R9 R10

23-04 16:45 A: correct 3 8 8 4 7 8 5 8 4 5 +10 1.9
3-05 11:17 B: incorrect 5 4 5 6 3 5 3 4 5 6 - 4 -0.7
4-05 10:32 C: none 5 7 8 3 4 4 6 5 4 5 +1 0.2
3-05 12:13 D: correct 8 4 4 5 4 6 7 6 6 6 + 6 1.1


A+B+C+D 21 23 25 18 18 23 20 23 19 22 +13 1.2


A+D 11 12 12 9 11 14 11 14 10 11 +16 2.2
4-05 12:00 a: direct 7 3 7 7 6 7 6 5 6 6 +10 1.9
4-05 13:02 b: direct 6 6 5 2 8 6 7 5 6 9 +10 1.9
4-05 12:22 c: direct 3 10 6 8 7 8 3 9 8 9 +21 4.1
4-05 11:18 d: direct 6 4 3 6 6 8 7 10 8 8 +16 3.1


a+b+c+d 22 23 21 23 27 29 23 29 28 32 +57 5.6

The total deviation for the conditions (A+D) in which correct delayed feedback is given, is +16 (z =2.19, p < 0.0143 one tailed). This supports the observational model which predicts that correct feedback is essential for psi to occur. The deviation for the set where the subject got incorrect feedback (B) is -4. Then again no positive deviation (-2) shows up when the calls of A.K. were compared with pack B itself, the pack he actually held in his hands during guessing.

The fact that no positive deviation is obtained in the condition where no feedback is given while the target cards were present with the subject (original pack B and pack C) indicates that the subject did not try and succeed to break the Loksure bags, having no prior knowledge of the identity of the set on which he would receive no feedback.

The scoring rates on the sets (a ,b, c, and d) where the subject opened the unprotected bags and got feedback immediately after finishing a set are spectacular (over all z = 5.65, p < 0.00001) and in the same range as those reported by A.K. for his earlier uncontrolled sessions.