Modelling childhood antecedents of anomalous experiences and beliefs : fantasy proneness, hypnagogic/hypnopompic and sleep experiences

• Main Author: Sherwood, Simon John
• Published: University of Edinburgh 2001
• Subjects: 155
• Online Access: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.661856

Two world-wide web (WWW) surveys (involving 108 and 130 adults, respectively) found that the level of childhood fantasy proneness and the incidences of a number of childhood HG/HP and sleep experiences had significant positive correlations with measures of childhood and adulthood anomalous experiences. The incidences of childhood and adulthood anomalous experiences were also significantly positively correlated with the current level of anomalous beliefs. These results provided some preliminary support for the paths in the proposed models. A further non-WWW version of the survey, with more traditional pencil and paper completion of the same questionnaire by a sample consisting mainly of university college staff and students (N-139), was also conducted. Intercorrelations within and between the childhood incidences of the hypnagogic/hypnopompic and sleep-related experiences suggested possible underlying factors. Subsequent factors analyses of the WWW and non-WWW survey data identified and confirmed three factors (HG/HP experiences, dream experiences, sleep disorder symptoms) within the sleep-related items and two factors (non-UFO-related, UFO-related) within both the anomalous experiences and beliefs measures. These findings were incorporated into revised models that proposed that these childhood factors, plus fantasy proneness, would be significantly intercorrelated and that each would also be a significant positive predictor of global measures of anomalous experiences. The models also proposed that the anomalous experience measures would be significant predictors of the levels of their respective beliefs. SEM analyses of data from both the WWW and the non-WWW surveys found that revised versions of the ES, CS and RC models provided a good fit to the data, although the RC model was considered to be the best of those tested for both empirical and theoretical reasons. These findings support Irwin’s rather than Lawrence’s model of the childhood antecedents of anomalous experiences and beliefs. The SEM analyses also suggested that, as Lawrence has argued, fantasy proneness is best modelled as only having an indirect relationship with anomalous beliefs via its relationship with measures of anomalous experiences.