A study of the relationship between narcissism, magical thinking, and body boundaries

Five indices were used in the study: two magical thinking scales (a 70 item, 7-point Likert Scale and an Essay Scale); the Holtzman Inkblot Test for body boundaries (scored for Barrier and Penetration based on the work of Fisher and Cleveland, 1970); the Self-Focus Questionnaire (based on Exner's work, 1973), to measure one aspect of narcissism; and, the Hanley Defensiveness Scale (1957) for the purpose of eliminating highly defensive subjects from the experimental population.

The results were compiled on a population of 258 university students who were administered the questionnaires in 1 of 12 standardized group presentations. Thirty-four (34) of these original subjects fell in the highest standard deviation on the Hanley Scale and were thus eliminated from the final results which were then statistically corrected for restriction of range.

The inter-rater reliability of the scales were all within acceptable limits. In addition, the two magical thinking scales were adequately cross-validated.

The results regarding the hypotheses of the study indicate: 1) there was no systematic relationship found in the predicted direction between Narcissism and Magical Thinking, nor between Narcissism and Body Boundaries; 2) a significant relationship was found between Magical Thinking and Body Boundaries in the predicted direction, with small to moderate correlations.

A factor analysis of Magical Thinking items clearly distinguished 5 factors and further highlighted the above results: Narcissism did not correlate with any of the factors in the predicted direction; Body Boundary subscales correlated significantly with all of the factors with correlations ranging in absolute value between .19 and .36. The highest and most significant correlations were between Body Boundary subscales and Factor 4 ("Superstitions, Objects of Good and Bad Luck"). This result was related to the use of "transitional objects" by adults during moments of anxiety and separation.

The results concerning Body Boundaries and Magical Thinking corroborate the theoretical position that magical thinking occurs to some extent within the framework of "syncretism" and "participation." That is, there is a partial confusion and diminished demarcation between subjective-internal and objective-external agency implied in magical thinking. The ramifications of this outcome regarding creativity was discussed

Unexpected results regarding the Narcissism Scale indicate a number of correlations between External-Focus of the Narcissism Scale and 3 Magical Thinking indices, specifically the Magical Thinking- Likert Scale, Factor 3 (“Death and Spirits of the Dead”), and Factor 5 (Magical Thinking-Power of Thoughts'). Although these results were not immediately consistent with the hypotheses, they made sense when understood in terms of the dynamics of depression, seen as the loss of significant others and the wish for reconstitution or reunion. An attempt was also made to understand these results in terms of the paranoid process, a stable externally-focused orientation based on the mechanism of projection; that is, the placing of unsatisfying internal impulses into the environment. Factor 3 (Death and Spirits of the Dead) was observed to be well-suited to the mechanism. In addition, Factor 5 (Magical Thinking-Power of Thoughts') was seen to be appropriate to the paranoid cognitive style of exquisitely focused perception with the aim of finding verification of one's thoughts and suspicions. Both the paranoid process and paranoid cognitive style were seen as "narcissistic defenses," that is, as ways of protecting tenuous self-esteem. This may have been partly responsible for the inconclusive results regarding the Narcissism index and the other 2 constructs.

An additional finding involved the Self-Focus Questionnaire concerned the "Neutral" Subscale's correlation to a number of subscales of both Body Boundaries and Magical Thinking. The direction of these correlations led to 2 possible purposes served by this scale: 1) as an index of "egocentric balance" as described by Exner (1973), seemingly replacing his proposed "Self-Total" Subscale (equaling the difference between "Self" and "External" subscales); or 2) that NEuT served a defensive function which could not be detected by the Hanley Scale, perhaps because it was not designed for a sentence-completion type of questionnaire.

Finally, the major suggestions for future research were to examine: 1) psychologically more impaired groups to determine if clearer relationships among the 3 constructs could be found; and 2) to examine other non-psychiatric groups to determine to what extent the relationships among the constructs are group-bound or culture-bound, especially in light of the suggested connection between cultural and religious positions and beliefs, on the one hand, and the tendency to endorse magical modes of thinking, on the other.