Epilepsy Information

American Epilepsy Society Poster (Abst. 1.347)

Correlation between psychological non epileptic seizure severity and self reported state/trait anger and anger expression

Authors: M. Lancman, O. Laban, E. Fertig, Y. Taher, J. Lee, K. Perrine and L. Myers

From a psychodynamic perspective, psychological non epileptic seizures (PNES), especially those that have a very violent and extreme motoric output, can be conceptualized as expressions of repressed anger and frustration. As such, an inverse correlation between conscious assertiveness and endorsed trait/state anger and severity of behavioral expressions during PNES seizures could be expected.


Thirty five patients who underwent video-EEG monitoring and were diagnosed with PNES were administered the State Trait Anger Expression Inventory-Second Edition (STAXI-2) as part of their standard neuropsychological battery. The Staxi-2 is a 57-item inventory that measures anger as an emotion state, the disposition to an angry trait, and an index of anger control/expression. All patients were studied in order to determine whether there was a correlation between their state, trait and expression scores and the behavioral severity of their seizure-like events. Subjects with mixed PNES and epileptic seizures, or physiological NES were excluded. A PNES behavioral severity scale was used to rate the intensity of the patients’ episodes on a scale of 0-5. This scale was developed using a previous study which objectively classified PNES by cluster analysis of semiologic features into 3 categories of severity ( Gröppel et al.). Episodes were scored using these criteria: decreased responsiveness (Y=1, N=0), maximal motor activity (no movement=0, trembling=1, clonic/hypermotor=2), head movement (Y=1,N=0), and pelvic thrusting (Y=1,N=0). An epileptologist reviewed at least 3 videos for each subject and selected the most habitual event for scoring.

No significant correlations were noted in any of the STAXI-2 subscales. Spearman's rank correlation coefficient was calculated between the behavioral severity scale and the STAXI-2 variables. The correlation between the overall Index score and severity was not significant (r=-.60, p<.734). Correlations were also non-significant between the behavioral severity score and Anger State (r=.02, p<.928) and Anger Trait (r=-.07, p<.676).

There was no significant correlation between the PNES behavioral severity scale and the overall anger index, state and trait as measured by the STAXI-2. This suggests that conscious endorsement of anger expression in situations of intense fury and angry state/trait characteristics do not correlate with seizure semiology in this population. Possibly, one of the greatest weaknesses in this design is the fact that the study is relying on self-report of an emotion that is so central in this disorder characterized by dissociation and denial. Therefore, another way to obtain anger assessments might be obtained through objective observer rating scales of assertiveness and anger expression.