Sleep to remember, sleep to protect: increased sleep spindle and theta activity predict fewer intrusive memories after analogue trauma

Recent evidence shows a strong correlative link between sleep disturbances and intrusive memories after traumatic events, presumably due to insufficient (nocturnal) memory integration. However, the underlying mechanisms of this link and the role of specifi c neural activities during sleep are poorly understood so far. Here, we investigated how the intra -individual affective response to an experimental trauma predicts changes in oscillatory activity during subsequent sleep and how these changes predict the pr ocessing of the experimental trauma. In a randomized within -subject comparison, twenty-two female, healthy participants (23.14 ± 2.46 years) watched a well-validated film clip including “traumatic” contents and a neutral film clip before bedtime on two sep arate nights. Heart rate was recorded during the film clips and nocturnal brain activity was recorded using 64-channel high-density EEG during subsequent nights. Intrusive memories were assessed via a seven -days diary and negative affect was assessed using laboratory trauma film reminders one week after the trauma film. An increased intra -individual heart rate during the trauma film predicted higher intra -individual sleep spindle amplitude the following night. Increased theta activity (4.25 - 8 Hz) during rapid eye movement (REM) sleep after the trauma film predicted fewer trauma film related intrusive memories and negative affect. Likewise, an increase in sleep spindles after the trauma film predicted fewer trauma film related intrusive memories. Our findings suggest that an experience-dependent up-regulation of these nocturnal oscillatory activity patterns, which are known to be involved in adaptive memory consolidation processes, serves as a protective factor against trauma-related intrusive memory development. Particularly, increased theta activity during REM sleep and sleep spindle activity seem to be of importance here.

trauma, post-traumatic stress disorder, sleep, emotion memory, heart rate .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 13, 2025. ; https://doi.org/10.1101/2025.03.10.642388doi: bioRxiv preprint 3

Exposure to a stressful or traumatic event, such as interpersonal violence, accidents, or natural disasters, is a risk factor for many psychopathologies, with posttraumatic stress disorder (PTSD) being the most frequent mental disorder that may develop after trauma. Intrusive memories – the involuntary re-experiencing of a traumatic event often triggered by situational reminders – can emerge in the aftermath of a traumatic event. Recurring and distressing intrusions are not only a hallmark symptom of PTSD but are thought to be an early predictor for the development of PTSD or other psychopathological symptoms (Ehlers & Clark, 2000; Haag, Robinaugh, Ehlers, & Kleim, 2017). According to well-recognized etiology models of PTSD (dual representation theory (Brewin & Burgess, 2014); cognitive model of PTSD (Ehlers & Clark, 2000); emotional processing theory (Foa & Kozak, 1986) ), intense arousal during the encoding of a traumatic event primarily stimulates perceptual and associative learning processes, which in turn can lead to an insufficient and fragmented memory trace of that experience. As a consequence, these unstructured m emory representations may entail intrusive memories, which can be considered the result of a disruption in basic memory function (van Marle, 2015). Given this, gaining deeper insights into mechanisms that drive these impaired memory functions after traumatic experiences seems key. One important factor in memory consolidation is sleep. Neural activity during sleep plays a crucial role in integrating new experiences into existing memory systems (Born & Wilhelm, 2012) and in reducing the intensity of distressing memories (Walker & van der Helm, 2009). Thus, it is not surprising that previous studies were able to show that sleep in the early aftermath of an analogue trauma can help reduce intrusive experiences and intrusion-related distress (Azza, Wilhelm, & Kleim, 2020; Kleim, Wysokowsky, Schmid, Seifritz, & Rasch, 2016; Porcheret et al., 2019; Wilhelm et al., 2021) . Furthermore, decades of basic sleep science provide extensive evidence that slow -wave sleep activity, sleep spindle activity (brief .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 13, 2025. ; https://doi.org/10.1101/2025.03.10.642388doi: bioRxiv preprint 4 bursts of neural activity in non-REM sleep (NREM)), and theta activity during rapid eye movement (REM) sleep are pivotal for successful memory consolidation. Additionally, most oscillatory activity during REM sleep can be attributed to the limbic system, a brain region known to be crucial for emotional memory consolidation (Genzel, Spoormaker, Konrad, & Dresler, 2015; Goldstein & Walker, 2014). However, there is growing evidence that neural activities during NREM and REM sleep complement each other in preserving and solidifying declarative aspects of an emotional memory, while at the same time attenuating its affective charge (Cairney, Durrant, Power, & Lewis, 2015; Rawson & Jackson, 2024) . With regard to intrusive memories that emerge after traumatic experiences, these adaptive processes seem to fail. Thus, understanding the underlying mechanisms and roles of specific sleep oscillations that promote adaptive memory consolidation of traumatic experiences, and thereby potentially protect from intrusion development seems paramount in order to bring forth early and late interventions. One way to investigate these underlying mechanisms is the use of an analogue trauma film paradigm. In this paradigm, highly distressing film content (e.g., interpersonal violence) is presented and elicited responses – which have been shown to be analogous to symptoms experienced after actual trauma (e.g., intrusive memories, physiological arousal, negative mood) – can be measured (Holmes & Bourne, 2008; James et al., 2016) . Experimental sleep studies that used such trauma film paradigms have shown that increased slow-wave activity during a nap after film presentation was associated with less intrusion -related distress during the following week (Wilhelm et al., 2021). Furthe rmore, it has been shown that increased sleep spindle activity after film presentation was related to fewer intrusive memories (Kleim et al., 2016) , to adaptive emotional memory processing (Kaestner, Wixted, & Mednick, 2013) , and to improved sleep - dependent anxiety regulation (Natraj et al., 2023). Furthermore, a daytime nap with REM sleep (as opposed to without) after the exposure to an analogue trauma film entailed less subsequent .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 13, 2025. ; https://doi.org/10.1101/2025.03.10.642388doi: bioRxiv preprint 5 intrusions (Wilhelm et al., 2021) and decreased the aversiveness of intrusive memories (Werner, Schabus, Blechert, & Wilhelm, 2021) in healthy individuals. Additionally, in a recent between - subject design study, greater theta activity during REM sleep after an experimental trauma was associated with significantly less intrusive memories (Sopp, Brueckner, Schäfer, Lass-Hennemann, & Michael, 2019). Further supporting the potential protective effects of theta activity, a cross - sectional study showed that individuals who experienced a traumatic event but did not develop PTSD, exhibited higher theta activity compared to individuals who deve loped PTSD (Cowdin, Kobayashi, & Mellman, 2014) . However, these links need replication, have in parts not yet been studied in an overnight sleep study, and – most importantly – need to be investigated in a randomized within -subject design to inform about potential causality and inter -individual protective factors. Given that only 10 -20% of the people who experienced a trauma develop clinical post - traumatic symptomatology (Hidalgo & Davidson, 2000) , this study aimed to determine potential protective factors against intrusive memory formation in a randomized within -subjects design in healthy individuals. For that, we used a validated, highly distressing film clip (i.e., “trauma” film) and a neutral film clip to examine how intra-individual changes in sleep physiology might protect from intrusion development. Firstly, we explored (i) the links between peri -trauma-film heart rate (as an indicator for arousal during encoding) and subsequent sleep oscilla tory activity. Secondly, we expected (ii) increased EEG oscillatory activity in the slow wave, theta, and spindle spectrum after the trauma film exposure. Finally, we expected (iii) increased slow -wave activity, sleep spindle activity, and theta activity t o be predictive of less trauma film related intrusive memories and negative affect.

Participants .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 13, 2025. ; https://doi.org/10.1101/2025.03.10.642388doi: bioRxiv preprint 6 Twenty-two healthy and female participants (mean age: 23.14 ± 2.46) took part in the present study and were recruited by the student mailing list of the University of Lübeck, Germany. Prior to the experiment the Ethics Commission of the University of Lübec k approved all study protocols and participants provided written informed consent. As a compensation, participants could choose to either receive a monetary incentive or course credits for their study program. Exclusion criteria were assessed in advance via a semi-structured telephone interview. These were: (1) experience of traumatic events involving interpersonal violence, (2) frequently watching of violent movies, (3) presence of a neurological or psychiatric disorder, (4) reported habitual consumption of alcohol or cannabis, (5) scheduled intake of medication influencing sleep during the study interval, (6) shift work. All participants were instructed to avoid caffeine on the day of the experiment. Procedure In a randomized within -subject comparison, each participant spent three nights in the sleep laboratory including polysomnographic recordings. The first night (T0) served as an adaptation of sleep to the new environment (i .e., adaptation night). Before going to sleep, participants filled in informed consent and a baseline battery of questionnaires including demographic data, the WHO- Five Well-Being Index (World Health Organization, 1998), the Emotion Regulation Questionnaire (Gross & John, 2003) , the Pittsburgh Sleep Quality Index (Buysse, Reynolds, Monk, Berman, & Kupfer, 1989), and the Edinburgh Handedness Inventory (Oldfield, 1971). On the second night (T1), according to a predefined randomization list, participants either watched a neutral or a highly distressing 12-minute film clip (see section “Trauma film paradigm ”) in a darkened room using headphones. They were further informed that the film material they will see could contain violent and/or distressing scenes and that they are free to withdraw from the experiment at any point in time. The third night (T2) took place with a minimal time gap of seven days. It was conducted identically to T1 except that participants watched the film clip they had not been presented with .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 13, 2025. ; https://doi.org/10.1101/2025.03.10.642388doi: bioRxiv preprint 7 before (neutral or trauma film). After the trauma film night, participants were instructed to document every intrusive memory that came to mind over the following seven days via an online questionnaire linked on their smartphone (i .e., intrusion diary) . Additionally, participants were exposed to trauma film reminders seven days after the trauma film night in our laboratory and were asked to rate their subjective negative affect before and after the presentation of trauma film reminders (i.e., intrusion pr ovocation task). All participants underwent both conditions on two separate nights (see also Figure 1). Figure 1. Study design In a randomized within-subject comparison, participants watched either first a film clip including distressing contents (“Trauma Film”) or a neutral film clip (“Neutral Film”) before bedtime with polysomnography. All participants underwent both conditions on two separate nights (test-nights). For both film clips, heart rate and subjective arousal and mood were assessed. Intrusion diary and intrusion provocation task were only administered after trauma film night. There was a minimal time gap of two days between adaptation night and first test-night and a minimal time gap of seven days between each test -night. Numbers (1) and (2) indicate in what sequence an individual underwent the study procedure depending on randomized assignments. .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 13, 2025. ; https://doi.org/10.1101/2025.03.10.642388doi: bioRxiv preprint 8 Trauma film paradigm1 The analogue traumatic stimulus in this study consisted of a previously used and validated 12 - minute scene from the movie “Irreversible”, directed by Gaspar No é including scenes of sexual violence (e.g., Kleim et al. (2016); Streb, Mecklinger, Anderson, Johanna, & Michael (2016)). Physiological and subjective reactions to the stimulus material were compared to a 12 -minute neutral film depicting neutral social interactions as well as architectural scenes. Heart rate was recorded during both film presentations. Before and after both films, participants rated their mood and levels of arousal on a visual analogue scale, namely the Self -Assessment Manikins (SAM) (Bradley & Lang, 1994). Sleep EEG recordings and analyses Brain activity during sleep was recorded using a Brain Vision system (LiveAmp) with a 64-channel electrode cap (BrainCap) and BrainAmp amplifiers (Brain Products, Munich, Germany). Two vertical electrooculogram (EOG) electrodes were placed above and below the left eye and two additional ones next to the lateral canthi to record horizontal eye movements. Three electromyography (EMG) electrodes (left and right upper chin and beneath chin) and two electrocardiography (ECG; left lower rib cage and right clavicl e) electrodes were additionally applied. Impedances were kept below 10 kΩ. Brain activity was recorded with a sampling rate of 500 Hz. 1 We acknowledge that watching a movie scene with distressing contents does not necessarily reflect real life trauma. We use the term “trauma film” for reasons of better readability only. .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 13, 2025. ; https://doi.org/10.1101/2025.03.10.642388doi: bioRxiv preprint 9 For the analysis, the EEG recordings were re -referenced to mean activity of both mastoids and sleep stages were determined by visual scoring according to the American Academy of Sleep Medicine standard criteria (AASM; Iber, Ancoli -Israel, Chesson, & Quan (2007)) using the software SchlafAus 1.0 (developed by Steffen Gais, unpublished, University of Tuebingen, Germany). Data was further preprocessed by inspecting all channels of each person for possible distortions and marking them. Next, all data was high pass filtered at 0.1 Hz and beforehand defined bad channels were interpolated. Power spectra for slow -wave activity (SWA; 0.5 - 4Hz) were averaged over all of NREM2 and NREM3 episodes, whereas for theta activity (4.25 - 8Hz), spectra were averaged over all REM s leep intervals using Fast Fourier Transformation in Matlab 2016b (The MathWorks Inc., 2016) . To estimate the individual increase of SWA after sleep onset, an SWA slope for each participant was calculated by subtracting the maximum two -minute-mean of SWA during the first sleep cycle from the initial two-minute-mean divided by the number of two- minute intervals. Discrete sleep spindle events (12 - 16Hz) were detected during artifact free intervals of NREM sleep 2 and 3 as described in Ngo, Fell, & Staresina (2020). Intrusion Diary Intrusive memories were assessed via a mobile daily diary for seven days following the traumatic film clip via a questionnaire link installed on the participant's smartphone. The time and date of the entries were recorded. Participants were asked to document the sudden occurrence of any intrusive memories i.e., images, thoughts or auditory stimuli related to the trauma film -clip in open text format. They were further asked to classify the type of memory (image, sound, thought) and to indicate the arousal and the degree of distress they experienced associated with the intrusion on a scale from 0=not at all to 100=very distressing/aroused. (adapted from Kleim, Graham, Bryant, & Ehlers (2013)). Only memories that occurred suddenly during the participants’ day were counted .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 13, 2025. ; https://doi.org/10.1101/2025.03.10.642388doi: bioRxiv preprint 10 as intrusions. To ensure that, we implemented a control item that asked for the suddenness of each memory. Participants received daily email reminders in the evening to remind them of entering possible memories in case they forgot. The main outcome variabl e was defined as the absolute number of intrusive memories reported over the seven days after trauma film exposure. Intrusion provocation task During an intrusion provocation task seven days after the trauma film presentation, five trauma - film-associated pictures (i.e., trauma film reminders) were presented for four seconds each (suit, dark street, stairs, underpass, and a pedestrian tunnel). Aft er each of the pictures, during a one - minute interval, participants were asked to place their fingers on the spacebar, close their eyes, and press the spacebar for every upcoming intrusion. The total number of indicated intrusions for each image was calculated. Before and after the task, self-reported negative affect was evaluated using the Positive and Negative Affect Schedule - German Version (Krohne, Egloff, Kohlmann, & Tausch, 1996). Statistical Analysis Descriptive statistics and comparisons of mean values were analyzed using R (R Core Team, 2017). To validate that the trauma film induced an increase in arousal and negative mood in our sample, paired t -tests (pre/post film presentation) for each of the conditions (trauma/neutral) were calculated. To examine potential differences of the topographical activity patterns between the two conditions more closely, cluster -based permutation testing was conducted for all electrodes and frequency bands of slow-wave activity (0.5 - 4 Hz), theta activity (4.25 - 8 Hz), and spindle activity (12 - 16 Hz). Cluster-based permutation tests were conducted using FieldTrip (paired samples t - test, 1,000 iterations; Oostenveld, Fries, Maris, & Schoffelen (2011)). The cluster test statistic was .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 13, 2025. ; https://doi.org/10.1101/2025.03.10.642388doi: bioRxiv preprint 11 assessed against the Monte -Carlo permutation distribution to test the null hypothesis of no differences between the two conditions. Intra -individual change scores (trauma vs. neutral film) for heart rate and sleep measures were calculated and used for furt her analyses. We examined whether individual changes in heart rate during the trauma film (compared to the neutral film) predicted individual changes in the analyzed sleep measures using Spearman correlation analyses. To control for multiple comparisons, we applied cluster statistics based on the Monte Carlo method. Further, we examined whether individual changes in sleep measures after the trauma film (compared to the neutral film) predicted intrusive memories in the intrusion diary and negative affect in the intrusion provocation task using Spearman correlation (ρ). We set the threshold to control for family-wise error (FWE) to p = 0.05 (one-sided test).

Effects of experimental trauma film on arousal, mood, and sleep architecture In line with previous studies, the participants’ moods significantly deteriorated ( t = 5.78, df = 21, p < .0001) while arousal levels significantly increased ( t = -4.64, df = 21, p < .0001) by the presentation of the trauma film. For the neutral film, mood scores did not change significantly, whereas arousal levels significantly decreased ( t = 2.74, df = 20, p = .031) from pre to post (see Table 1). Table 1. Mood and arousal in response to the film material Pre Film Mean ± SEM Post Film Mean ± SEM t(df) p Trauma Film Mood (SAM) 6.55±0.34 4±0.33 5.78(21) <.0001 Arousal (SAM) 4±0.43 6±0.34 -4.64(21) <.0001 Neutral Film Mood (SAM) 6.9±0.32 6.86±0.28 0.22(20) .825 Arousal (SAM) 3.57±0.51 2.57±0.40 2.74(20) .013 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 13, 2025. ; https://doi.org/10.1101/2025.03.10.642388doi: bioRxiv preprint 12 SEM = standard error of the mean; SAM = Self-Assessment Manikin We further examined whether there were any detectable differences in sleep architecture when comparing the trauma with the neutral film condition. Participants showed significantly increased sleep onset latencies (t = 2.73, df = 21, p = .013) after the trauma film while there was no difference between conditions in any other sleep variable (see Table 2). Interestingly, participants showed a flatter slope of SWA increase during the first sleep cycle across 41 channels primarily over the right hemisphere after the trauma film condition exposure (t = -3.54 – -2.08, df = 21, p ≤ .01). Table 2. Sleep architecture after trauma and neutral condition Trauma film Mean ± SEM Neutral film Mean ± SEM t(df) p Sleep onset latency (minutes) 23.69±2.79 18.15±1.72 2.73(21) .013 NREM 1 (%)1 4.3±0.55 3.93±0.38 0.84(21) .412 NREM 2 (%)1 55.98±1.12 56.88±1.42 -0.61(21) .547 NREM 3 (%)1 13.37±0.8 13.85±0.96 -0.58(21) .570 REM sleep (%)1 21.24±0.87 20.89±0.89 0.56(21) .579 Total sleep time (minutes) 474.59±8.35 473.98±9.13 0.89(21) .891 Wake after sleep onset (%) 5.09±1.14 4.42±0.92 0.56(21) .579 1 Proportion of sleep stages relative to total sleep time; SEM = standard error of the mean; REM = rapid eye movement sleep; NREM = non-REM sleep stages. .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 13, 2025. ; https://doi.org/10.1101/2025.03.10.642388doi: bioRxiv preprint 13 .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 13, 2025. ; https://doi.org/10.1101/2025.03.10.642388doi: bioRxiv preprint 14 Figure 2. Differences in frequency bands comparing trauma and neutral film condition Topographical distribution of the mean power spectra in three parameters of interest (slow -wave activity (A): 0.5 - 4Hz; theta activity (B): 4.25 - 8Hz; spindle count (C), and -amplitude (D): 12 - 16Hz) is illustrated for the trauma film (left), the neutral film condition (middle), and the difference between the two conditions (right). None of the measured frequency bands showed significantly different activities between trauma and neutral film conditions. Effects of experimental trauma film exposure on neural sleep correlates Average slow-wave activity (t = -2.15 – 1.87, p > .05) and sleep spindle activity (count: t = -1.45 – 1.42, p > .05; amplitude: t = -1.12 – 1.01, p > .05) during NREM sleep, as well as average theta activity (t = -1.41 – 1.99, p > .05) during REM sleep after trauma film exposure did not differ significantly compared to that after the neutral film (see Figures 2A - 2D). Correlation between peri-trauma-film heart rate and sleep oscillations Individual peri-trauma-film heart rate during the trauma film did not significantly predict slow - wave activity (see Figure 3A) or theta activity the following night (see Figure 3B). Intra- individually increased peri -trauma-film heart rate significantly predicted intra -individually increased sleep spindle amplitude the following night in 25 channels over frontal, central, parietal, and parieto -occipital areas (ρ = 0.586 – 0.404, p < .05; see also Figure 3D). No significant correlation was found for sleep spindle count (see Figure 3C). .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 13, 2025. ; https://doi.org/10.1101/2025.03.10.642388doi: bioRxiv preprint 15 Figure 3. Correlation between heart rate during film exposure and sleep measures (A-C) No significant correlations were found for slow -wave activity, theta activity, and sleep spindle count. (D) Intra -individual increase in heart rate during the trauma film (as compared to the neutral film) significantly predicted intra -individual increase in sleep spindle amplitude after the trauma film (compared to neutral). Dots in yellow mark the channels contributing to the significant cluster. .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 13, 2025. ; https://doi.org/10.1101/2025.03.10.642388doi: bioRxiv preprint 16 Prediction of intrusions and affective response to trauma film reminders Individual changes in slow -wave activity after the trauma film exposure did not predict intrusive memory formation during the seven days intrusion diary period nor did it predict negative affect during the intrusion provocation task (see Figures 4A and 5A). As hypothesized, an intra-individual increase in theta activity during REM sleep over left frontal, temporal, and parietal as well as right temporal and parietal areas in 30 channels after the trauma film significantly predicted less intrusive memories during the seven days intrusion diary period (ρ = -0.747 – -0.364, p < .05; see also Figure 4B). Similarly, intra-individually increased theta activity during REM sleep over frontal areas in 14 channels after the trauma film predicted less negative affect during the intrusion provocation task (ρ = -0.592 – -0.384, p < .05; see also Figure 5B). Intra -individually increased sleep spindle counts across nearly the entire scalp in 56 channels after the trauma film predicted significantly fewer intrusive memories during the seven days intrusion diary period (ρ = -0.672 – -0.366, p < .01; see also Figure 4C). Other individual changes in sleep spindle activity after the trauma film did not predict intrusive memories nor negative affect during the intrusion provocation task (see Figures 4D, 5C, and 5D). .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 13, 2025. ; https://doi.org/10.1101/2025.03.10.642388doi: bioRxiv preprint 17 Figure 4. Correlations between intra-individual sleep changes and intrusions An intra-individual increase of theta activity (B) and sleep spindle count (C) after the trauma film (compared to neutral) was significantly associated with less intrusive memories reported during the seven days intrusion diary period . No significant correlations were found for slow -wave activity and sleep spindle amplitude (A and D). Dots in yellow mark the channels contributing to the significant cluster. .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 13, 2025. ; https://doi.org/10.1101/2025.03.10.642388doi: bioRxiv preprint 18 Figure 5. Correlations between intra-individual sleep changes and negative affect An intra-individual increase of theta activity (B) after the trauma film (compared to neutral) was significantly associated with less negative affect during the intrusion provocation task one week after the trauma film exposure. No significant correlations were found for slow-wave activity, sleep spindle count and sleep spindle amplitude (A, C, and D). Dots in yellow mark the channels contributing to the significant cluster.

.CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 13, 2025. ; https://doi.org/10.1101/2025.03.10.642388doi: bioRxiv preprint 19 Here, we differentiated for the first time the impact of an experimental trauma film versus a neutral film on intra -individual neural sleep activity using high density EEG recordings in a within-subjects design. This allowed for conducting fine -grained cluster analyses of associations between intra-individual changes in sleep physiology and the processing of aversive experiences. Against our hypothesis, we did not find significant changes in mean slow -wave activity, sleep spindle activity, and theta activity after the trauma film compared to the neutral film. This most likely can be attributed to high inter -individual differences o n group level in response to a comparatively “mild” stressor. Since this study comprised only healthy participants, it is likely that they responded very differently to this “mild” stressor on an individual level. Consequently, the effects of the stressor may be present at the individual level but not at the group level. However, in line with our expectations and with previous studies (Cowdin et al., 2014; Sopp, Brueckner, Schäfer, Lass -Hennemann, & Michael, 2019; Wilhelm et al., 2021) , intra - individually increased theta activity during REM sleep after the trauma film predicted less intrusive memories in the intrusion diary and less negative affect in the intrusion provocation task. This strengthens the notion of theta activity being an affective “depotentiator” (Walker, 2009) and being actively involved in adaptive emotional memory consolidation (Nishida, Pearsall, Buckner, & Walker, 2009) . Congruently, previous studies have shown that increased REM theta power significantly predicted enhanced emotional memory consolidation after a stress exposure (Kim et al., 2020). Thus, our findings support the notion that increased theta activity during REM sleep after a traumatic event is a protective factor against intrusive memories and trauma related negative affect. Moreover, our study is the first to demonstrate that theta activity is regulated in an experience-dependent manner and that this regulation is adaptive in face of a traumatic stressor. Furthermore, in accordance with previous studies (Kleim et al., 2016; Wilhelm et al., 2021), we found an intra -individually increased sleep spindle count after the trauma film to predict less .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 13, 2025. ; https://doi.org/10.1101/2025.03.10.642388doi: bioRxiv preprint 20 intrusive memories in the intrusion diary the following week. This aligns with research that showed increased sleep spindle activity is linked to reduced sleep -dependent anxiety responses in traumatized individuals (Natraj et al., 2023). Additionally, as sleep spindles change in response to experiences and promote synaptic plasticity (Fernandez & Lüthi, 2020) , they play a key role in flexibly adapting to new demands. In sum, our findings suggest that increased sleep spindle activity facilitates adaptive emotional memory consolidation after traumatic events, helping to mitigate the development of intrusive trau matic memories (Natraj & Richards, 2023) . However, individuals with PTSD often show increased sleep spindles, which has been suggested to reflect maladaptive over-consolidation of traumatic events and thus lead to more intrusive memories (van der Heijden et al., 2022) . Considering these somewhat opposing findings, a differentiated assessment seems necessary. In light of the strong correlation between intra -individually increased heart rate during trauma film exposure and subsequent increased sleep spindle amplitude in this study, we propose temporarily increased sleep spindle activity after an acute stressor (accompanied by temporarily increased arousal) to reflect adaptive emotional memory processing. In contrast, increased sleep spindle activity in PTSD patients (acco mpanied by chronic hyperarousal) could reflect some sort of “dysfunctional replay” as a maladaptive attempt of the sleeping brain to integrate traumatic experiences into existing memory systems (see also Natraj & Richards (2023)). Thus, one could speculate that moderately increased emotional reactivity (i.e., heart rate) during a stressful event promotes adaptive nocturnal emotional processing by up -regulating sleep spindle activity, which in turn would mitigate intrusion development. Corroborating this, previous studies have shown that reduced heart rate during a trauma film predicted the development of more frequent intrusive images in healthy individuals (e.g., Chou, La Marca, Steptoe, & Brewin (2014)) and that attenuated skin conductance responses (i.e., physiological arousal) during an emotion regulation task predicted later increased intrusion development in trauma -exposed individuals .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 13, 2025. ; https://doi.org/10.1101/2025.03.10.642388doi: bioRxiv preprint 21 (Shepherd & Wild, 2014). This notion is further supported by a close affect -specific link between heart rate and neural activity in integral parts of the limbic system (Kuniecki, Barry, & Kaiser, 2003; Yang et al., 2007) , a network crucial for memory consolidation (Catani, Dell’acqua, & Thiebaut de Schotten, 2013) and known to be altered in individuals with PTSD (Shin, Rauch, & Pitman, 2006) . Generally speaking, when dealing with traumatic experiences, emotional engagement up to a certain degree as opposed to detachment (i.e., dissociation) has been shown to be beneficial in preventing or reducing PTSD symptoms, presumably by supporting adapt ive (nocturnal) emotional memory processing (Möller, Söndergaard, & Helström, 2017; Rauch & Foa, 2006). However, it remains to be determined whether there are certain levels of arousal at different time points around a traumatic event that are beneficial for or detrimental to adaptive memory encoding and subsequent (nocturnal) consolidation (see e.g., Chou et al. (2014) for an attempt to decipher distinct effects of increased arousal at different time points around an analog traumatic event). Strengths of our study include the conduct of a within-subjects design, which allowed us to assess and examine intra-individual changes in heart rate and sleep measures and thereby emulate real life inter-individual differences in response to traumatic events more accurately. Furthermore, instead of daytime naps, we used whole night high-density EEG recordings, which warrant higher spatial resolution and independence of ultradiane phases. As to limitations, our study comprised a relatively small sample with only female, non-clinical participants, which limits generalizability. Further, although analogue trauma film paradigms are widely used, they do not necessarily reflect real life trauma and our findings therefore ideally would need replication in naturalistic settings. In conclusion, our findings provide further evidence for theta activity during REM sleep and sleep spindle activity during NREM sleep after analogue trauma to be up -regulated in an experience-dependent manner and by this being protective against intrusion development. .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 13, 2025. ; https://doi.org/10.1101/2025.03.10.642388doi: bioRxiv preprint 22 Furthermore, our findings suggest a close interplay between physiological reactivity during a traumatic event and subsequent sleep spindle activity. Given that sleep disturbances are prevalent in individuals with PTSD (Pace-Schott, Germain, & Milad, 2015) and are likely to contribute to PTSD symptom formation in the aftermath of traumatic events (Mellman, Pigeon, Nowell, & Nolan, 2007) , interventions that aim to stabilize post -traumatic sleep quality seem paramount. Additionally, according to our findings, improving theta and sleep spindle activity after a traumatic event could promote adaptive emotional memory consolidation and thereby possibly prevent PTSD symptom formation (for theoretical reviews, see Murkar & De Koninck (2018); Mushtaq, Marshall, Ul Haq, & Martinetz (2024)).

This work was supported by a grant of the Swiss National Science Foundation (SNSF, 10001C_179241). We thank Anna Wick, Fenja Rohrberg, and Lea Strelow for their ambitious help in data collection.

Azza, Y., Wilhelm, I., & Kleim, B. (2020). Sleep early after trauma: A target for prevention and early intervention for posttraumatic stress disorder? European Psychologist, 25(4), 239–251. Born, J., & Wilhelm, I. (2012). System consolidation of memory during sleep. Psychological Research, 76(2), 192–203. Bradley, M. M., & Lang, P. J. (1994). Measuring emotion: the Self-Assessment Manikin and the Semantic Differential. Journal of Behavior Therapy and Experimental Psychiatry, 25(1), 49–59. Brewin, C. R., & Burgess, N. (2014). Contextualisation in the revised dual representation theory of PTSD: .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 13, 2025. ; https://doi.org/10.1101/2025.03.10.642388doi: bioRxiv preprint 23 a response to Pearson and colleagues. Journal of Behavior Therapy and Experimental Psychiatry, 45(1), 217–219. Buysse, D. J., Reynolds, C. F., 3rd, Monk, T. H., Berman, S. R., & Kupfer, D. J. (1989). The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research. Psychiatry Research, 28(2), 193–213. Cairney, S. A., Durrant, S. J., Power, R., & Lewis, P. A. (2015). Complementary roles of slow-wave sleep and rapid eye movement sleep in emotional memory consolidation. Cerebral Cortex (New York, N.Y.: 1991), 25(6), 1565–1575. Catani, M., Dell’acqua, F., & Thiebaut de Schotten, M. (2013). A revised limbic system model for memory, emotion and behaviour. Neuroscience and Biobehavioral Reviews, 37(8), 1724–1737. Chou, C.-Y., La Marca, R., Steptoe, A., & Brewin, C. R. (2014). Heart rate, startle response, and intrusive trauma memories: Startle heart rate moderates trauma response. Psychophysiology, 51(3), 236–246. Cowdin, N., Kobayashi, I., & Mellman, T. A. (2014). Theta frequency activity during rapid eye movement (REM) sleep is greater in people with resilience versus PTSD. Experimental Brain Research. Experimentelle Hirnforschung. Experimentation Cerebrale, 232(5), 1479–1485. Ehlers, A., & Clark, D. M. (2000). A cognitive model of posttraumatic stress disorder. Behaviour Research and Therapy, 38(4), 319–345. Fernandez, L. M. J., & Lüthi, A. (2020). Sleep spindles: Mechanisms and functions. Physiological Reviews, 100(2), 805–868. Foa, E., & Kozak, M. (1986). Emotional processing of fear: exposure to corrective information. Psychological Bulletin, 99(1), 20–35. Genzel, L., Spoormaker, V. I., Konrad, B. N., & Dresler, M. (2015). The role of rapid eye movement sleep for amygdala-related memory processing. Neurobiology of Learning and Memory, 122, 110–121. Goldstein, A. N., & Walker, M. P. (2014). The role of sleep in emotional brain function. Annual Review of Clinical Psychology, 10, 679–708. Gross, J. J., & John, O. P. (2003). Individual differences in two emotion regulation processes: implications .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 13, 2025. ; https://doi.org/10.1101/2025.03.10.642388doi: bioRxiv preprint 24 for affect, relationships, and well-being. Journal of Personality and Social Psychology, 85(2), 348– 362. Haag, C., Robinaugh, D. J., Ehlers, A., & Kleim, B. (2017). Understanding the emergence of chronic posttraumatic stress disorder through acute stress symptom networks. JAMA Psychiatry (Chicago, Ill.), 74(6), 649–650. Hidalgo, R. B., & Davidson, J. R. (2000). Posttraumatic stress disorder: epidemiology and health-related considerations. The Journal of Clinical Psychiatry, 61 Suppl 7, 5–13. Holmes, E. A., & Bourne, C. (2008). Inducing and modulating intrusive emotional memories: a review of the trauma film paradigm. Acta Psychologica, 127(3), 553–566. Iber, C., Ancoli-Israel, S., Chesson, A. L., & Quan, S. F. (2007). The AASM manual for the scoring of sleep and associated events. Westchester: American Academy of Sleep Medicine. James, E. L., Lau-Zhu, A., Clark, I. A., Visser, R. M., Hagenaars, M. A., & Holmes, E. A. (2016). The trauma film paradigm as an experimental psychopathology model of psychological trauma: intrusive memories and beyond. Clinical Psychology Review, 47, 106–142. Kaestner, E. J., Wixted, J. T., & Mednick, S. C. (2013). Pharmacologically increasing sleep spindles enhances recognition for negative and high-arousal memories. Journal of Cognitive Neuroscience, 25(10), 1597–1610. Kim, S. Y., Kark, S. M., Daley, R. T., Alger, S. E., Rebouças, D., Kensinger, E. A., & Payne, J. D. (2020). Interactive effects of stress reactivity and rapid eye movement sleep theta activity on emotional memory formation. Hippocampus, 30(8), 829–841. Kleim, B., Graham, B., Bryant, R. A., & Ehlers, A. (2013). Capturing intrusive re-experiencing in trauma survivors’ daily lives using ecological momentary assessment. Journal of Abnormal Psychology, 122(4), 998–1009. Kleim, B., Wysokowsky, J., Schmid, N., Seifritz, E., & Rasch, B. (2016). Effects of sleep after experimental trauma on intrusive emotional memories. Sleep, 39(12), 2125–2132. Krohne, H. W., Egloff, B., Kohlmann, C.-W., & Tausch, A. (1996). Untersuchungen mit einer deutschen .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 13, 2025. ; https://doi.org/10.1101/2025.03.10.642388doi: bioRxiv preprint 25 Version der “Positive and Negative Affect Schedule” (PANAS). Diagnostica, 42(2), 139–156. Kuniecki, M., Barry, R., & Kaiser, J. (2003). The influence of emotionally relevant context on the evoked cardiac response triggered by an irrelevant stimulus. Journal of Psychophysiology, 17(2), 61–68. Mellman, T. A., Pigeon, W. R., Nowell, P. D., & Nolan, B. (2007). Relationships between REM sleep findings and PTSD symptoms during the early aftermath of trauma. Journal of Traumatic Stress, 20(5), 893–901. Möller, A., Söndergaard, H. P., & Helström, L. (2017). Tonic immobility during sexual assault - a common reaction predicting post-traumatic stress disorder and severe depression. Acta Obstetricia et Gynecologica Scandinavica, 96(8), 932–938. Murkar, A. L. A., & De Koninck, J. (2018). Consolidative mechanisms of emotional processing in REM sleep and PTSD. Sleep Medicine Reviews, 41, 173–184. Mushtaq, M., Marshall, L., Ul Haq, R., & Martinetz, T. (2024). Possible mechanisms to improve sleep spindles via closed loop stimulation during slow wave sleep: A computational study. PloS One, 19(6), e0306218. Natraj, N., Neylan, T. C., Yack, L. M., Metzler, T. J., Woodward, S. H., Hubachek, S. Q., … Richards, A. (2023). Sleep spindles favor emotion regulation over memory consolidation of stressors in posttraumatic stress disorder. Biological Psychiatry: Cognitive Neuroscience and Neuroimaging, 8(9), 899–908. Natraj, N., & Richards, A. (2023). Sleep spindles, stress and PTSD: The state of the science and future directions. Neurobiology of Stress, 23, 100516. Ngo, H.-V., Fell, J., & Staresina, B. (2020). Sleep spindles mediate hippocampal-neocortical coupling during long-duration ripples. eLife, 9. https://doi.org/10.7554/eLife.57011 Nishida, M., Pearsall, J., Buckner, R. L., & Walker, M. P. (2009). REM sleep, prefrontal theta, and the consolidation of human emotional memory. Cerebral Cortex (New York, N.Y.: 1991), 19(5), 1158– 1166. Oldfield, R. C. (1971). The assessment and analysis of handedness: the Edinburgh inventory. .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 13, 2025. ; https://doi.org/10.1101/2025.03.10.642388doi: bioRxiv preprint 26 Neuropsychologia, 9(1), 97–113. Oostenveld, R., Fries, P., Maris, E., & Schoffelen, J.-M. (2011). FieldTrip: Open source software for advanced analysis of MEG, EEG, and invasive electrophysiological data. Computational Intelligence and Neuroscience, 2011, 156869. Pace-Schott, E. F., Germain, A., & Milad, M. R. (2015). Effects of sleep on memory for conditioned fear and fear extinction. Psychological Bulletin, 141(4), 835–857. Porcheret, K., van Heugten-van der Kloet, D., Goodwin, G. M., Foster, R. G., Wulff, K., & Holmes, E. A. (2019). Investigation of the impact of total sleep deprivation at home on the number of intrusive memories to an analogue trauma. Translational Psychiatry, 9(1), 104. Rauch, S., & Foa, E. (2006). Emotional processing theory (EPT) and exposure therapy for PTSD. Journal of Contemporary Psychotherapy, 36(2), 61–65. Rawson, G., & Jackson, M. L. (2024). Sleep and emotional memory: A review of current findings and application to a clinical population. Current Sleep Medicine Reports, 10(3), 378–385. R Core Team. (2017). R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing. Retrieved from Retrieved from https://www.r-project.org/ Shepherd, L., & Wild, J. (2014). Emotion regulation, physiological arousal and PTSD symptoms in trauma-exposed individuals. Journal of Behavior Therapy and Experimental Psychiatry, 45(3), 360– 367. Shin, L. M., Rauch, S. L., & Pitman, R. K. (2006). Amygdala, medial prefrontal cortex, and hippocampal function in PTSD. Annals of the New York Academy of Sciences, 1071(1), 67–79. Sopp, M. R., Brueckner, A. H., Schäfer, S. K., Lass-Hennemann, J., & Michael, T. (2019). REM theta activity predicts re-experiencing symptoms after exposure to a traumatic film. Sleep Medicine, 54, 142–152. Streb, M., Mecklinger, A., Anderson, M. C., Johanna, L.-H., & Michael, T. (2016). Memory control ability modulates intrusive memories after analogue trauma. Journal of Affective Disorders, 192, 134–142. .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 13, 2025. ; https://doi.org/10.1101/2025.03.10.642388doi: bioRxiv preprint 27 The MathWorks Inc. (2016). MATLAB version: R2016b. Natick, Massachusetts: The MathWorks Inc. Retrieved from https://www.mathworks.com van der Heijden, A. C., Hofman, W. F., de Boer, M., Nijdam, M. J., van Marle, H. J. F., Jongedijk, R. A., … Talamini, L. M. (2022). Sleep spindle dynamics suggest over-consolidation in post-traumatic stress disorder. Sleep, 45(9), zsac139. van Marle, H. (2015). PTSD as a memory disorder. European Journal of Psychotraumatology, 6(1), 27633. Walker, M. P. (2009). The role of sleep in cognition and emotion. Annals of the New York Academy of Sciences, 1156(1), 168–197. Walker, M. P., & van der Helm, E. (2009). Overnight therapy? The role of sleep in emotional brain processing. Psychological Bulletin, 135(5), 731–748. Werner, G. G., Schabus, M., Blechert, J., & Wilhelm, F. H. (2021). Differential Effects of REM Sleep on Emotional Processing: Initial Evidence for Increased Short-term Emotional Responses and Reduced Long-term Intrusive Memories. Behavioral Sleep Medicine, 19(1), 83–98. Wilhelm, I., Azza, Y., Brennwald, K., Ehrt-Schäfer, Y., Seifritz, E., & Kleim, B. (2021). Investigating the effect of a nap following experimental trauma on analogue PTSD symptoms. Scientific Reports, 11(1), 4710. World Health Organization. (1998). Use of well-being measures in primary health care-the DepCare project health for all. Target, 12, E60246. Yang, T. T., Simmons, A. N., Matthews, S. C., Tapert, S. F., Bischoff-Grethe, A., Frank, G. K. W., … Paulus, M. P. (2007). Increased amygdala activation is related to heart rate during emotion processing in adolescent subjects. Neuroscience Letters, 428(2-3), 109–114. .CC-BY-NC 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted March 13, 2025. ; https://doi.org/10.1101/2025.03.10.642388doi: bioRxiv preprint