A methodological issue in the study of correlation between psychophysiological variables.
ABSTRACT: The authors used previously accumulated skin conductance (SC) and EEG data to examine the effects of their respective autocorrelations on hypothesis testing. It was found that SC data remain autocorrelated for many seconds and that EEG data remain autocorrelated for many fractions of a second depending on filtering parameters. The authors show that the effect of these nonzero autocorrelations on the interpretation of correlation coefficients using normal statistics can lead to substantial and artifactually inflated significance levels. With SC, for example, the high autocorrelation can lead to a Pearson's correlation of 1.0 even under the null hypothesis. The resulting null-hypothesis z-score distribution range is [-20,20], whereas it should be approximately in the range [-3,3]. Alpha EEG, while less auto-correlated than SC, still leads to Pearson's correlations in the range [-0.4,0.4], leading to a null-hypothesis z-score distribution in the range [-15,15]. Beta-band EEG reduces the null-hypothesis z-score range to [-5,5]. The authors demonstrate that standard Monte Carlo techniques can provide valid estimates of the significance levels. The underlying assumptions of conventional statistical tests can be easily ignored, and the resulting error may become embedded into the thinking of a research community. As an example, the authors critically review a 1994 article by Grinberg-Zylberbaum, Delaflor, Attie, and Goswami claiming significant correlation between the EEGs of isolated participants; however, using uncorrelated EEG data from one of our previous studies and Monte Carlo methods to model the true null hypothesis, the authors compute a nonsignificant difference (z=1.22) between their non-"correlated" participants and their "correlated" ones. As a result of their (possibly incorrect) interpretation of these correlations, there is a growing literature proclaiming that these experiments are evidence for Einstein-Podolsky-Rosen (EPR)-like quantum connections in isolated brains. These putative connection s have been used as explanations, or at least plausibility arguments, for a variety of phenomena including distant healing.
|Printer friendly Cite/link Email Feedback|
|Author:||May, E.C.; Spottiswoode, S.J.P.; Faith, L.V.|
|Publication:||The Journal of Parapsychology|
|Date:||Dec 1, 2001|
|Previous Article:||Conjuring up images of India.|
|Next Article:||Mathematical models of the ritual healing theory: Shamanism, anomalous experience, and the origin of religion.|