A man strangles his wife while dreaming about fighting off intruders in his sleep. Does that make him mad, bad or innocent? Recent research is helping to unpick these issues, and may help reveal who, if anyone, bears responsibility in such cases.
Last week, British man Brian Thomas appeared in court on a murder charge after strangling his wife as they slept in their camper van.
The prosecution withdrew the charges after three psychiatrists testified that locking him up would serve no useful purpose. The judge said that Thomas bore no responsibility for his actions.
The case has cast a spotlight on the use of such sleepwalking defences in court.
"If you look at the media reports there appears to be an upsurge in the use of the sleepwalking defence," says Michel Cramer-Bornemann of the Minnesota Regional Sleep Disorders Center in Minneapolis.
Thomas had a genuine sleep disorder, but Cramer-Bornemann is concerned that in many other cases, the sleepwalking and other sleep-related defences are misused. Studies on the causes of sleepwalking may eventually make it easier to identify who has a genuine sleep disorder that could occasionally result in violence, and who is making it up.
Last month, Ursula Voss of Bonn University in Germany and colleagues reported that even during lucid dreaming – a state in which some people claim to be able to control their dreams – some areas of the brain associated with intent stayed offline, while other areas associated with consciousness were active.
"As long as you are in a dream, you have no free rein on your actions and emotions," says Voss.
Although this research didn't look specifically at sleepwalkers, it tallies with a previous study by Claudio Bassetti at the University of Zurich in Switzerland, who once managed to manoeuvre a sleepwalker into a brain scanner during a sleepwalking episode.
He found the sleepwalker also showed no activation in the areas of the brain associated with intent, though emotional areas and those associated with movement were active.
"Our judgement is off and our ability to act out emotionally is on," says Rosalind Cartwright of the Sleep Disorder Service and Research Center in Chicago. She believes a confirmed diagnosis of sleepwalking would make a strong defence in court, but says better tests are needed to establish who has a genuine sleep disorder.
That might become easier with the recent discovery that auditory stimuli, such as a dog barking, can trigger sleepwalking in those susceptible to it – particularly if they have been suffering from sleep deprivation.
Antonio Zadra at the University of Montreal in Quebec, Canada, and his colleagues measured the brain activity of 10 sleepwalkers and 10 control subjects to determine what stage of sleep they were in.
They found that sounding a buzzer during "slow wave" sleep triggered sleepwalking in three of the sleepwalkers under normal circumstances, and all 10 sleepwalkers when they had been kept awake for 25 hours prior to sleeping. None of the control subjects were prompted to sleepwalk when the buzzer was sounded.
The study might eventually enable a test for genuine sleepwalkers. "That's a big breakthrough," says Cartwright. Until recently, defence lawyers used evidence of sleepwalking in childhood or a family history of the activity to back up their claims.
However, Zadra cautions that other factors, like having the motivation to commit a crime, must also be taken into account.
"We should not forget that some sleepwalkers can be criminals," he says.