Abstract

Comments on 2 recent articles by H. Kreitler and S. Kreitler which reported 4 experiments on ESP showing a relatively small but constant change in Ss' performance in line with the monitored ESP messages. It is argued that the main problem with these experiments is the confounding of possible real events with statistical regression. A hypothetical example of coin tosses is presented to illustrate this problem, and several specific points of disagreement are examined.