Diary of events in a thoroughly unhaunted house

Two subjects were asked to keep a 30-day journal of occurrences in their residence which are traditionally associated with hauntings and poltergeist-like phenomena. It was expected that the instructions would increase the frequency of perception of unusual or rarely occurring events in their residence, resulting in an attentional and "perceptual contagion effect." As expected, several features of the percipients' observations closely resembled those found in previously reported poltergeist episodes. Most importantly, the distribution of perceived occurrences closely resembled the predicted logistic curve, suggesting that hauntings involve a perceptual process where initial observations "infect" later ones.