Abstract
In a recently published major article in the Journal of Personality and Social Psychology, Daryl Bem (2011) made a strong claim for the existence of a parapsychological phenomenon called retroactive causation. Across nine experiments, aspects of stimuli were shown to correlate with participants’ responses provided before the stimuli were generated by the computer’s random generator. Early critical debates of these provocative findings have been focused on issues of statistical significance testing. Going beyond these issues, we argue that Bem’s research has three crucial shortcomings: (a) a lack of a theoretical explanation, (b) the possibility of selective filtering of empirical results, and (c) the confusion of the explanans and the explanandum. We propose that all three methodological principles should be rigorously applied during the journal review process and the communication of empirical findings in general.
Get full access to this article
View all access options for this article.
