Comparison of language and somatic experiences between reports of trauma and trauma-related dreams & personality features of trauma-exposed persons reporting trauma-related dreams
OA Version
Citation
Abstract
INTRODUCTION: (Study A) Trauma-Related Nightmares (TRNs) are a core feature of Post- Traumatic Stress Disorder (PTSD). We explored linguistic and somatic-experience differences between self-reports of trauma and those of nightmares related to the trauma.
(Study B) Neurotic personality features are associated with many psychological disorders, including PTSD. Based on this relationship, we explored whether neuroticism predicts the rate of nightmares and bad dreams as well as the number of replicative nightmares (TRNs similar or exactly like their traumatic experience), above and beyond PTSD severity.
METHODS: (Study A) Seventeen participants with varying severity of PTSD symptoms reporting recurring TRNs (mean age 27.47 years, SD = 10.33, 14 females) recalled a traumatic experience and nightmares related to that trauma. Trauma reports were written by participants, while nightmare reports were transcribed from audio recordings made as they were recalled following nightmares. Following both types of reports, participants indicated co-occurring somatic experiences by choosing from a list of 51 selections. Choices were later grouped into cardiovascular, respiratory, interoceptive, and tension categories. Linguistic content was measured using the Linguistic Inquiry and Word Count (LIWC) program and positive emotion, negative emotion, and somatosensory category words were totaled. Since trauma reports had significantly higher word counts than TRNs (p=0.0495), LIWC categories were normalized for total word count. Total and symptom- cluster severities of PTSD were assessed using the PTSD Checklist for DSM-5 (PCL-5). Wilcoxon Signed-Rank Tests and Spearman Correlations were used for statistical analysis, as Shapiro-Wilk tests showed that data were non-normally distributed.
(Study B) 126 participants who had experienced a traumatic event within the past two years were recruited (mean age 24.13 years, SD = 4.994, 69% female) and, for an average of 14.89 nights, completed a dream questionnaire on which occurrence of nightmares (causing awakening) and bad dreams were reported and ranked based on their similarity to their recent traumatic experience. PTSD symptoms were assessed using the PCL-5 and personality features such as neuroticism were measured using the NEO Personality Inventory Revised (NEO PI-R), a questionnaire based on the Five Factor Model of personality. The combined number of nightmares and bad dreams was divided by the total number of nights reported and expressed as a rate, while a replicative nightmare count was generated by summing “similar to traumatic experience” and “exactly like traumatic experience” ratings. Hierarchical regressions were used to determine whether neuroticism predicted the rate of nightmare and bad dreams as well as the number of replicative nightmares above and beyond PTSD severity. Pearson correlations were used to check for relationships between variables and possible collinearity.
RESULTS: (Study A) There were significantly more somatic experiences of interoception (p=0.0084) and tension (p=0.024) in trauma vs nightmare reports. The intrusion cluster of the PCL-5 was associated with cardiovascular (rho=0.592, p=0.0156) and respiratory (rho=0.619, p=0.0109) experiences in trauma reports, and interoception (rho=0.718, p=0.0033) and tension (rho=0.556, p=0.0224) experiences in nightmare reports.
(Study B) In two hierarchical regression models, neuroticism predicted neither nightmare and bad dream rate nor number of replicative nightmares over and above total or PTSD symptom cluster severity (p=0.596; p=0.886). Collinearity checks did demonstrate a moderate positive relationship between these variables (r=0.317, p<0.001).
CONCLUSION: (Study A) More somatic experiences of interoception and tension were recalled from traumas than TRNs. Because the brain is deafferented from sensory input during dreaming, we expected, but did not find, state differences in other somatic experiences. Word categories in narratives also did not show state differences. Only the intrusion symptoms of PTSD predicted bodily sensations in trauma as well as TRN reports.
(Study B) We found that neuroticism did not predict either nightmare and bad dream rate or the number of replicative nightmares above and beyond PTSD severity, when taking demographic factors into account. The positive correlation between PTSD and neuroticism could explain this lack of significance.
SUPPORT: R21MH128619
Description
License
Attribution-NonCommercial-NoDerivatives 4.0 International