Violence in Sleep

Francesca Siclari; Ramin Khatami; Frank Urbaniok; Lino Nobili; Mark W. Mahowald; Carlos H. Schenck; Michel A. Cramer Bornemann; Claudio L. Bassetti


Brain. 2010;133(12):3494-3509. 

In This Article

Disorders Underlying Sleep-related Violence

True sleep-related violence occurs in the course of parasomnias and nocturnal seizures. In nocturnal dissociative disorders (Schenck et al., 1989b), factitious disorder, malingering (Griffith and Slovik, 1989; Mahowald et al., 1992) and dementia, violence may be related exclusively to the nocturnal period, but occurs in a state of wakefulness (Table 1).


Disorders of Arousal Disorders of arousal consist of incomplete awakenings from non-rapid eye movement (non-REM) sleep characterized by reduced vigilance, impaired cognition, retrograde amnesia for the event and variable motor activity, ranging from repetitive and purposeless movements to more complex behaviours such as eating, drinking, driving, sexual intercourse and aggression (American Academy of Sleep Medicine, 2005). Confusional arousals consist of confusional behaviour upon awaking from sleep, most often from slow-wave sleep in the first part of the night. If an individual leaves the bed and starts walking, the disorder is referred to as sleepwalking. In sleep terrors, the arousal is characterized by intense autonomic activation and typical behavioural features such as sitting up in bed and screaming, while the person remains unresponsive to external stimuli. Different arousal disorders may coexist in the same individual and an episode may start as one arousal disorder and evolve into another (i.e. sleep terror evolving into agitated sleepwalking). Arousal disorders preferentially occur in the first half of the night and tend not to recur during the same night or to arise from daytime naps. Dreamlike mentation associated with episodes of sleepwalking or sleep terrors are mostly unpleasant, involving aggression on the part of the sleeper (24%), misfortune (54%) and apprehension (84%) (Oudiette et al., 2009).

Disorders of arousal are common during childhood but may persist (18–25%) or arise de novo during adulthood (0.6%) (Hublin et al., 1997). Genetic predisposition in the case of sleepwalking is suggested by a 10-fold increased prevalence among first-degree relatives of sleepwalkers (Kales et al., 1980). The genes that confer the risk of sleepwalking remain essentially unknown. The only established marker is the presence of the HLA DQB1 allele found in 35% of sleepwalkers, compared with only 13% of normal subjects (Lecendreux et al., 2003).

It is generally assumed that episodes of arousal disorders are more likely to occur in genetically predisposed individuals in the presence of an increased pressure for slow-wave sleep and factors favouring arousals or fragmenting sleep. Increased slow-wave sleep pressure occurs after sleep deprivation and with a large variety of psychotropic medications, as reviewed elsewhere (Pressman, 2007b). Among these, sedative–hypnotic medications, particularly zolpidem, are associated with complex and aggressive sleep-related behaviours (Dolder and Nelson, 2008; Lam et al., 2008; Hwang et al., 2010). Based on the fact that a few early studies showed an increase in slow-wave sleep after alcohol ingestion (Prinz et al., 1980; MacLean and Cairns, 1982; Dijk et al., 1992; Feige et al., 2006; Van Reen et al., 2006), alcohol has been claimed to induce sleepwalking episodes by defendants in criminal cases. A recent review of the literature on the subject, however, revealed that low levels of alcohol in social drinkers resulted in an increase in slow-wave sleep in only 6 of 19 published studies. Studies in which alcohol was administered to a clinically diagnosed sleepwalker, or providing evidence for increases in slow-wave sleep in severely intoxicated individuals were not identified (Pressman et al., 2007). Conditions known for increasing the likelihood of arousals from sleep include sleep disordered breathing, periodic leg movements in sleep, fever (Karacan et al., 1968), stress and provocation by noise and touch (Pressman, 2007a). Among these, obstructive sleep apnoea is increasingly recognized as a trigger for violent disorders of arousal (Baron and Auckley, 2004; Lateef et al., 2005; Schenck and Mahowald, 2008). Experimental sleep deprivation (Joncas et al., 2002; Zadra et al., 2008) and forced arousals (Pilon et al., 2008) have been shown to increase the number and complexity of somnambulistic episodes in sleepwalkers.

Violence in Arousal Disorders Disorders of arousal have a potential for sleep-related violence. Homicide and attempted homicide (Yellowless, 1878; Podolsky, 1959, 1961; Bonkalo, 1974; Chuaqui, 1975; Hartmann, 1983; Oswald and Evans, 1985; Tarsh, 1986; Howard and D'Orban, 1987; Scott, 1988; Ovuga, 1992; Broughton et al., 1994; Nofzinger and Wettstein, 1995), filicide (Luchins et al., 1978), along with completed and attempted 'pseudo-suicide' (Chuaqui, 1975; Mahowald et al., 2003) have been reported in the setting of arousal disorders in the medical literature (Table 2). Inappropriate sexual behaviours (Wong, 1986; Hurwitz et al., 1989; Buchanan, 1991; Fenwick, 1996; Shapiro et al., 1996; Rosenfeld and Elhajjar, 1998; Guilleminault et al., 2002; Ebrahim, 2006; Bejot et al., 2010) occurring during arousal disorders are frequently associated with violence. Among the 31 patients with parasomnia-related sleepsex reviewed by Schenck and coworkers, 45% displayed assaultive behaviour, 29% had sex with minors and 36% sustained legal consequences from their sexual sleep-related behaviour.

Pressman (2007a) reviewed 32 legal and medical case reports of violence associated with disorders of arousal and found that aggressive behaviour occurred in different ways across confusional arousals, sleepwalking and sleep terrors. In confusional arousals, violence was preferentially elicited when individuals were awakened from sleep by someone else. Violent behaviour during sleepwalking, in contrast, tended to occur when the sleepwalking episode was already underway and the individual was approached by another person or incidentally encountered someone else. Violence related to sleep terrors appears to be a reaction to a (dreamed or hallucinated) concrete, frightening image that the individual can subsequently describe. In the same series of cases, Pressman (2007a) examined the role of physical contact and proximity as triggering factors of sleep-related violence and found that 100% of confusional arousals, 81% of sleep terrors and 40–90% of sleepwalking cases were associated with provocations including noise, touch and/or close proximity. The victims were rarely sought by sleepwalkers.

Two further studies tried to identify risk factors for the occurrence of violence in disorders of arousal. Moldofsky et al. (1995) retrospectively compared clinical and polysomnographic characteristics of subjects with sleepwalking and sleep terrors and found that violent behaviour directed towards other people occurred preferentially in males and was significantly associated with more stressors, a disturbed sleep schedule, excessive use of caffeinated beverages, drug abuse and less stage 4 non-REM sleep. The retrospective nature of this study and the fact that occurrence of violence and associated factors was based uniquely on the individual's subjective recall represents major limitations of this study. Guilleminault et al. (1995) retrospectively reviewed a series of 41 adult individuals with nocturnal wandering of different aetiologies. Compared with non-violent individuals, the 29 violent patients were predominantly male (65 versus 42%) and included the only two subjects with temporal lobe epilepsy.

Functional neuroimaging studies performed on sleeping subjects reveal that during non-REM sleep, the prefrontal cortex is hypoactive in comparison with the resting wakeful state (Maquet et al., 1996, 1997; Braun et al., 1997; Andersson et al., 1998; Kaufmann et al., 2006). The prefrontal associative cortices are functionally related to executive functions including planning, attention and judgement (Cummings, 1993). A single photon emission computed tomography study performed during an episode of sleepwalking documented hyperperfusion of the posterior cingular cortex and cerebellar vermis in addition to decreased cerebral blood flow in the frontal and parietal association cortices (Bassetti et al., 2000). In line with this observation, a recent intracerebral electroencephalogram (EEG) study showed that during confusional arousals, the motor and cingulate cortices are activated and display the same activity as during wakefulness, whereas the frontoparietal associative cortices exhibit an enhancement of delta activity characteristic of sleep (Terzaghi et al., 2009). These findings suggest that disorders of arousal share characteristics of both sleeping and waking states, resulting from the selective activation of thalamo-cingulo pathways implicated in the control of complex motor and emotional behaviour, and from hypoactivation of other thalamocortical pathways, including those projecting to the frontal lobes. This constellation of regional brain activity during non-REM sleep shares a number of similarities with the presumed anatomical substrates of violence occurring during wakefulness. Dysfunction in the medial prefrontal and orbitofrontal region areas is indicated by impaired emotional recognition of faces, errors in odour identification and disadvantageous decisions in gambling tests, all of which have been observed in patients with impulsive aggressive disorder (Best et al., 2002; Dileo et al., 2008). Further evidence for the involvement of these areas is suggested by the finding that traumatic brain injuries are associated with violence if localized to the frontal ventromedial or orbitofrontal regions (Grafman et al., 1996). Individuals with intermittent explosive disorder exhibit exaggerated amygdala activity and diminished activation of the orbitofrontal cortex in response to faces expressing anger, and fail to demonstrate coupling between these two structures (Coccaro et al., 2007). According to functional models (Blair, 2004), the amygdala and the orbitofrontal lobe act on subcortical systems mediating reactive aggression. Although the role of the amygdala is to up- or down-regulate their response to threat, the orbitofrontal cortex exerts its modulating activity in response to social cues. It is thus conceivable that unwarranted aggressive behaviours occurring during parasomnias are favoured by the unrestrained influence of the amygdala that is no longer kept in balance by proper introspection of executive function normally maintained by the dorsolateral prefrontal cortex, since it remains hypoactive.

In summary, arousal disorders represent a dissociated state with features of both wakefulness and sleep. Sleep violence in this setting seems to be favoured by hypoactivity of prefrontal associative cortices that are functionally related to planning, attention and judgement. It is further modulated at different levels—the predisposition for arousal disorders is mainly influenced by genetic and maturational factors. The occurrence of an arousal disorder on a particular night depends on a delicate balance between slow-wave sleep pressure and arousing factors. Violent acts occur more frequently in males, can be elicited by provocation (noise, touch) and are probably influenced by dream imagery and mental content.

Rapid Eye Movement Sleep Behaviour Disorder

RBD is characterized by the loss of normal muscle atonia and an increase of phasic muscle activity during REM sleep, and is associated with altered dream content and acting out of dreams. It is more prevalent in males after the age of 50. In at least 50% of cases, it represents the first manifestation of a neurodegenerative disorder including Parkinson's disease, multisystem atrophy and dementia with Lewy bodies (Iranzo et al., 2006b). RBD may also occur in the course of narcolepsy (Schenck and Mahowald, 1992; Nishino and Kanbayashi, 2005), secondary to brainstem lesions (Kimura et al., 2000; Plazzi and Montagna, 2002; Zambelis et al., 2002; Provini et al., 2004; Tippmann-Peikert et al., 2006; Mathis et al., 2007) or limbic encephalitis (Iranzo et al., 2006a; Lin et al., 2009). Dreams of patients with idiopathic and secondary RBD have a more aggressive content compared with dreams of normal subjects, despite non-elevated or even lower levels of daytime waking aggressiveness (Fantini et al., 2005). Likewise, among patients with Parkinson's disease, those affected by RBD have more aggressive dreams compared with patients without RBD, irrespective of gender (Borek et al., 2007).

Violence in Rapid Eye Movement Sleep Behaviour Disorder Violence with RBD results in injury to the patient or to the bed partner in 32–69% of cases (Comella et al., 1998; Olson et al., 2000; Scaglione et al., 2005). Harm to the patient himself generally occurs when the patient hits the furniture or wall, or falls out of bed, and can result in serious injuries, including subdural haematomas, high cervical fractures and other fractures (Dyken et al., 1995; Comella et al., 1998; Olson et al., 2000). Attempted assault of sleep partners has been reported to occur in 64% of cases, with injuries in 3% (Olson et al., 2000). It may be potentially lethal to the affected person and to the bed partner (Schenck et al., 2009). Unlike violence related to disorders of arousal, in RBD the individual is readily oriented upon awakening and can generally recall vivid dream imagery related to the violent act. Figure 1 shows a practical example of sleep violence in the context of RBD.

Figure 1.

An 82-year-old male with a 2-year history of symmetrical extrapyramidal symptoms and cognitive dysfunction, attributed to vascular leucencephalopathy. According to his wife, he displayed frequent vocalizations and violent behaviour during sleep. Polysomnographic recordings (D) show loss of normal muscle atonia during REM sleep and increased phasic muscle activity, as well as acting out of dreams with kicking (A), hitting (B) and menacing gestures (C). Based on these features, the diagnosis of RBD was established.

Lesion studies performed on cats (Jouvet and Delorme, 1965; Hendricks et al., 1982; Lai and Siegel, 1988, 1990, 1997; Shouse and Siegel, 1992; Morrison, 1998) have suggested that absence of motor activity during normal REM sleep relies on two main mechanisms: (i) active inhibition of spinal motor neurons (i.e. REM atonia), and (ii) reduced drive of locomotor centres (i.e. leading to reduced phasic muscle activity). In the rat, potential neuroanatomical substrates for the control of REM sleep have been identified (Boissard et al., 2002, 2003; Lu et al., 2006): 'REM off' regions localized in the ventrolateral part of the periaqueductal grey matter and the lateral pontine tegmentum would interact in a mutually inhibitory way with 'REM on' regions (i.e. the sublaterodorsal nucleus and precoeruleus region) that project to the medulla and spinal cord, with a net inhibitory effect on motor neurons. Based on lesion and neuropathological studies as well as pharmacological effects, similar mechanisms have been postulated to be operative in humans (Boeve et al., 2007). The precise nature and localization of the locomotor generators, which are presumed to project to the spinal motoneurons, either directly or indirectly via other brainstem nuclei, has yet to be defined. The observation that limbic encephalitis in humans, without evidence for brainstem dysfunction, is associated with RBD (Iranzo et al., 2006a) and the altered dream content in RBD (Fantini et al., 2005; Borek et al., 2007) suggest a role of the limbic system in the pathogenesis of RBD, possibly by modulation of connections between the limbic system and the brainstem regions responsible for REM sleep atonia. Studies using single photon emission computed tomography imaging showed hypoperfusion in frontal and temporo-parietal cortical areas in patients with RBD compared with control subjects, providing further evidence for involvement of supratentorial structures in the pathophysiology of RBD (Shirakawa et al., 2002; Mazza et al., 2006).

In summary, disorders causing RBD affect brainstem structures or supratentorial limbic regions. Dream enactment occurs in the presence of increased activity of locomotor centres and incomplete REM sleep associated muscle atonia. Violence in this setting is related to enactment of aggressive dream content and shows a strong male predominance.

Parasomnia Overlap Disorder and Status Dissociatus

Parasomnia overlap disorder is diagnosed when patients fulfil criteria for both RBD and disorders of arousal (American Academy of Sleep Medicine, 2005). In status dissociatus, features of each one of the three states (REM, non-REM and wakefulness) are simultaneously present, in the absence of identifiable conventional sleep stages (American Academy of Sleep Medicine, 2005). Both conditions can be associated with violent behaviour (Mahowald and Schenck, 1991; Schenck et al., 1997).


In nocturnal frontal lobe epilepsy, seizures can occur exclusively during sleep and can be associated with violence. Paroxysmal arousals consist of abrupt arousals from sleep with stereotyped motor phenomena, including head movements, frightened expression and dystonic limb posturing. Episodes of the so-called nocturnal paroxysmal dystonia involve more complex motor activity such as bipedal automatisms (kicking, cycling), rhythmic movements of the trunk (sometimes with semi-purposeful repetitive movements mimicking sexual activity) and limbs, as well as tonic and dystonic posturing. Autonomic activation with tachycardia and hypertension generally accompanies the episodes. Epileptic nocturnal wanderings generally start with an abrupt arousal, proceed through the stage of paroxysmal dystonia and eventually culminate in perambulation. Alerting stimuli have been shown to trigger seizures in patients with autosominal dominant nocturnal frontal lobe epilepsy (El Helou et al., 2008). On awakening, individuals are generally rapidly oriented, unlike in other seizure types, and there is no prolonged post-ictal confusion (Provini et al., 1999). Clinical manifestations of seizures are often very similar to the motor activity observed during disorders of arousal, making a distinction between the two conditions difficult (distinguishing features are summarized in Table 3). Nocturnal hypermotor seizures (Nobili et al., 2004; Ryvlin, 2006) including epileptic nocturnal wanderings (Nobili et al., 2002) can occasionally originate from the temporal lobe.

Violence in Epilepsy Injury resulting from nocturnal seizures can be accidental and related to hyperkinetic features (Schenck et al., 1989a; Fig. 2), although compared with seizures during wakefulness, the injury potential of seizures occurring solely during sleep is probably lower, as the bed represents a relatively safe environment (Wirrel, 2006). Ictal aggression is exceptional (Rodin, 1973; Delgado-Escueta et al., 1981; Ramani and Gumnit, 1981; Marsh and Krauss, 2000; Tassinari et al., 2005a; Reuber and Mackay, 2008). It can take the form of biting, grasping, hitting, threatening, screaming, facial expressions of anger, pushing, shoving and spitting (Marsh and Krauss, 2000; Tassinari et al., 2005a) and has been reported to occur during episodes of nocturnal epileptic wandering (Maselli et al., 1988; Schenck et al., 1989a; Guilleminault et al., 1995; Plazzi et al., 1995). The violent act generally starts abruptly, is of short duration, is not intentionally directed towards others and does not involve intricate skills or purposeful movements. Some behavioural patterns, such as ictal grasping or repetitive limb movements, although not directed towards others, can be misinterpreted by bystanders as threatening, intentional gestures. Ictal aggression is recurrent, out of character and stereotyped for a given patient, and is frequently associated with amnesia (Marsh and Krauss, 2000). These kinds of epileptic manifestations seem to be more common in males (Rodin, 1973; Delgado-Escueta et al., 1981; Marsh and Krauss, 2000; Tassinari et al., 2005a) and are more frequently observed in seizures originating from the non-dominant hemisphere (Tassinari et al., 2005a). Peri-ictal aggression occurs in the pre-ictal, or more commonly in the post-ictal period, in a setting of confusion and abnormal mood (depression, psychosis or delirium) (Fig. 3). More directed violence (towards others) can occur as a reaction to stimuli of the patient's environment; such as the act of restraining the patient (Delgado-Escueta et al., 1981; Tassinari et al., 2005a).

Figure 2.

A 36-year-old female with nocturnal frontal lobe epilepsy since the age of 13. Seizures occurred up to 10 times a night and were characterized by an abrupt awakening from sleep with a feeling of chest tightness and dyspnoea, and progressed to tonic posturing of the limbs and hyperkinetic features, including elevation and extension of the left arm (B) and pedalling (A and C). The patient was fully aware during the seizure but was unable to speak. She had sustained several injuries including fractures of her fingers when hitting objects during the seizures. The red arrow on the polysomnographic recording (D) shows the beginning of a seizure arising from slow wave sleep. Intracranial EEG recordings indicated that seizures originated from a cortical dysplasia in the left frontal cingulum.

Figure 3.

A 32-year-old male with frontal lobe epilepsy since the age of 18 years. Seizures started with an aura characterized by cephalic tingling and fear followed by left head deviation associated with clonic jerks (first in the left arm, then rapidly bilaterally). Consciousness was preserved. Nocturnal seizures were occasionally followed by a secondary generalization with a post-ictal phase characterized by violent movements and auto-aggressive behaviour. The patient could throw himself against the wall (first line of photogram), fall out of bed (second line) or leave the room (third line). Intracerebral EEG recordings showed that seizures originated from a cortical dysplasia in the right frontal superior gyrus.

Intracranial EEG investigations have demonstrated that seizures characterized by violent complex behaviours and intense fear are associated with a predominantly bilateral dysfunction of a network involving the cingulate, orbitofrontal and temporal regions (the amygdala, hippocampus and lateral temporal cortex) (Tassinari et al., 2005a). As with disorders of arousal, these potential anatomical substrates are similar to those postulated to underlie violence during wakefulness. Additional observations indicate that arousal disorders and seizures in nocturnal frontal lobe epilepsy might share a final common pathway. Both conditions have a similar clinical semiology. Patients with nocturnal frontal lobe epilepsy have a high prevalence of parasomnias in the personal or family history compared with large control populations (Provini et al., 1999; Bisulli et al., 2010). Finally, single photon emission computed tomography imaging findings documented cingular and cerebellar hyperperfusion in both sleepwalking and epileptic paroxysmal arousals (Bassetti et al., 2000; Vetrugno et al., 2005). It has been postulated that the final common pathway between arousal disorders and seizures in nocturnal frontal lobe epilepsy might imply activation of so-called central pattern generators in the brainstem that subserve stereotyped innate motor behaviour necessary for survival, such as locomotion, swimming, sexual activity, other rhythmic motor sequences and aggressive behaviour (as an innate action pattern aimed at the defence of the peri-personal space) (Tassinari et al., 2003, 2005b, 2009). However, solid evidence for the existence of central pattern generators in humans is still lacking.

In summary, ictal sleep-related violence results from direct activation of specific brain regions in the frontal and temporal lobes. Alternatively, violence can result from hyperkinetic features or occur in the post-ictal period, in a setting of confusion or abnormal mood. Nocturnal frontal lobe epilepsy and arousal disorders have similar patterns of brain activation and share common favouring factors (sleep deprivation, provocation), suggesting a common final pathway.

Nocturnal Dissociative Disorders

Dissociative disorders are defined as a disruption of the usually integrated functions of consciousness, memory, identity or perception of the environment that occur without the conscious awareness of the part of the individual (American Psychiatric Association, 2000). They are sometimes associated with violence (McCaldon, 1964) and may arise exclusively or predominantly from the sleep period (Schenck et al., 1989b; Fig. 4) during well-established EEG wakefulness, either at the transition from wakefulness to sleep or within several minutes after awakening from stages 1 or 2 of non-REM or from REM sleep (American Academy of Sleep Medicine, 2005). Patients are generally female, have a history of sexual abuse and post-traumatic stress disorder and often have additional dissociative episodes during daytime (Schenck et al., 1989b).

Figure 4.

Fifty-three minutes after sleep onset in the sleep lab, a 19-year-old male suddenly begins to growl and then 28 s later he leaves the bed and crawls away in the manner of a large jungle cat, as noted by the sleep technologist. A nine channel EEG indicates a corresponding wakeful state. For 4 years, once or twice weekly, he had acted like a large jungle cat (with deep, persistent growling) for 30–60 min during the nocturnal sleep period at home. His parents commented on his 'super-human strength' during these episodes, such as leaping far from his bed, lifting a marble table with his jaws, lifting a mattress with his jaws and then dragging it across a room. He frequently left imprints of his teeth on wooden furniture. He often bruised and lacerated himself all over his body from aggressive and violent behaviour during these episodes, and also repeatedly injured his lips and gums from biting sharp objects. He has never had a dissociative episode during the daytime, so his case represents an exclusively nocturnal, sleep-related animalistic dissociative (multiple personality) disorder. Reprinted from Schenck et al. (1989b), with permission.

Factitious Disorder

In factitious disorder (also named Münchhausen syndrome), an individual presents with an illness that is deliberately produced or falsified for the purpose of assuming the sick role or produces symptoms in another person (factitious syndrome by proxy) (American Psychiatric Association, 2000). Both disorders may occur during the sleep period and can be associated with violence (Griffith and Slovik, 1989; Mahowald et al., 1992).


In contrast to factitious disorder, malingering is not a mental illness and intentionally produced symptoms or signs are motivated by external incentives for the behaviour (i.e. economic gain, avoiding legal responsibility, improving physical well-being) (American Psychiatric Association, 2000). Again, symptoms can be referred to the sleep period, but when observed, always occur during well-established EEG-wakefulness.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: