Psi in a Skeptic's Lab: A Successful Replication of Ertel's Ball Selection Test

Abstract

In the Ball Selection Test for assessing psi, ping pong balls are drawn blindly from an opaque bag one at a time with replacement. Each ball has an integer from 1-5 and red or green dots marked on it, thereby producing 10 distinct alternatives. On each trial, a participant jumbles the balls, and attempts to guess both the number and the dot color on the ball prior to pulling it out of the bag. Because the 10 ball types are equally represented in the bag, the probability of correctly guessing both the number and the dot color by chance is 10%. In the full protocol, participants first test themselves at home without supervision. Those who score significantly above chance are then retested in the laboratory under an experimenter's supervision. In an experiment by the author with participants of the Georg-Elias-Müller Institute (GEMI), 47 participants achieved a hit rate of 11.6% in the at-home phase of the study, p = 10^-14> by a one-tailed binomial test; nine selected participants retested in the laboratory achieved a hit rate of 17.3% (p = 10^-50). A replication of the laboratory procedure was conducted by two graduating students working under the guidance of a skeptical professor at the Anomalistic Psychology Research Unit (APRU) at Goldsmiths College, University of London. Their 40 unselected APRU participants achieved a hit rate of 10.75, which was very significant by a binomial test (p = .002) and p = .0003 by summed Z^2 values. The lower hit rate of the APRU participants compared with GEMI participants was significant (p = .02) and predicted. It is argued that this low-tech testing procedure is less monotonous and more psi-conducive than conventional multiple choice procedures for testing psi.

Keywords: ESP--Ball Selection Test--skeptics