Skip to main content

Verified by Psychology Today

Sleep Driving and Sleep Killing

Exploring the Kenneth Parks case.

People do bizarre things while asleep. Some talk and laugh in their sleep. Some kick their partners due to restless leg syndrome. Some snore and temporarily stop breathing.

And then there are people who get out of their beds while still asleep and complete complicated actions. We call it sleepwalking. But walking is not all they do.

People can do astonishing and complicated things in their sleep. One Australian woman would regularly get out of bed at night and have sex with strangers. Former chef Rob Wood cooked spaghetti bolognese and fish and chips during his sleep. And some people kill.

The “sleepwalking defense” has been used on several occasions but only a few have been acquitted on those grounds. The first was Albert Tirrell. He slit the throat of a prostitute in Boston in 1846, set fire to the brothel and took off to New Orleans. The defense was able to convince the jury that Tirrell was a chronic sleepwalker and could have committed the crimes while sleepwalking. He was acquitted but most of the contemporaries analyzing the case didn’t believe the defense lawyer’s story.

The perhaps most interesting and convincing case of a killer who got acquitted on the grounds that he was sleepwalking was that of Kenneth Parks. In many ways, this case is a beautiful portrait of a superhuman mind run amok.

The Kenneth Parks Case

One of the most talked-about trials in history was the R. v. Parks, [1992] 2 S.C.R. 871. Kenneth James Parks, a 23-year-old man from Toronto with a wife and infant daughter, was suffering from severe insomnia and anxiety owing to unemployment and gambling debts.

The summer before, Kenneth had repeatedly placed bets on horse races, which had caused him severe financial problems. To obtain more money for gambling, he stole $32,000 from his employer Revere Electric. Kenneth kept losing money, and when the company found out about the theft in March 1987, he was fired. Court proceedings were brought against him, and his personal life suffered.

In the early morning of May 1987, Kenneth got out of bed and drove 23 km from Pickering to the house of his wife’s parents, Barbara Ann and Denis Woods, in the Toronto suburb of Scarborough. After fetching a tire iron from the car trunk and using his key to enter the house, he proceeded to the bedroom of the house and choked his father-in-law unconscious. He then beat his mother-in-law with the tire iron and stabbed her repeatedly with a kitchen knife. He also stabbed his father-in-law.

Barbara was later found in a room five to six feet away from the bedroom. She had been stabbed in the chest, shoulder, and heart. She had sustained blunt-force injuries to her nose, eye, and skull that caused a subarachnoid hemorrhage. Though Dennis was unconscious, his wounds were less severe. That night, Kenneth also picked up the phone in the kitchen and set it down again, off the hook. He ran upstairs to the teenage daughters’ bedrooms. But he stopped outside the door, just stood there, then ran down again and left.

After the killing, Kenneth then drove to the police station. He arrived at 4:45 a.m., covered in blood, and said “I just killed someone with my bare hands; oh my God, I just killed someone; I’ve just killed two people; my God, I’ve just killed two people with my hands; my God, I’ve just killed two people. My hands; I just killed two people. I killed them; I just killed two people; I’ve just killed my mother- and father-in-law. I stabbed and beat them to death. It’s all my fault.”

The police said that he seemed distressed and was shaking. He did not appear to be in pain, despite having cut tendons in both hands. This is an example of dissociative analgesia, a profound blunting of pain sensation in the absence of painkillers. Dissociative analgesia can occur during states of sleepwalking but also after drug use and in states of shock or great distress.

After careful examination of the case, the experts could find no other explanation of the crime than sleepwalking. Kenneth underwent a series of sleep tests and psychological tests. The electroencephalography (EEG) scans showed that Kenneth had some abnormal brain activity during deep sleep, periods of partial awakenings, which is indicative of parasomnia. Since there allegedly is no way to fake one’s own EEG results, and Kenneth had appeared to feel no pain when he arrived at the police station, it was determined that he was sleepwalking when he attacked his in-laws.

The experts described Kenneth’s actions as the result of many circumstances converging: He had plans to fix his in-laws’ furnace, he was familiar with the route he would take to get to their house, and he was restless from anxiety and worried about his upcoming trial. The experts thought that it suddenly occurred to Kenneth in his sleep that he should fix his in-law’s furnace. He then got up and drove to the house but was startled by his in-laws. He attacked both of them without knowing what he was doing.

Sleepwalking doesn’t automatically lead to a full acquittal. An involuntary act entitles an accused to an unqualified acquittal only if the automatic condition did not originate in “a disease of the mind” that has made the person insane. In the latter case, the accused is not entitled to a full acquittal, but only to a verdict of insanity. “Disease of the mind” is not a medical term but a legal term. Because it is a legal term, a trial judge cannot rely blindly on medical opinion but must also consider the likelihood of recurrence and the cause of the act. A condition likely to present recurring danger should be treated as insanity. A condition stemming from the internal makeup of the accused, rather than external factors, should also lead to a verdict of insanity. These two conditions might seem sufficient to justify less than full acquittal of sleepwalkers who kill. But the defense at Kenneth’s trial argued that a combination of external factors caused the killing and that it was unlikely that a similar combination of external factors would occur again in the future.

In the medical review, it was concluded that the legal defense was, therefore, one of homicide during non-insane automatism as part of a presumed episode of somnambulism... the defendant did not have any preexisting "disease of the mind" within the meaning of... the Canadian Criminal Code. There was no evidence for psychosis or other mental pathology. Moreover, it was believed that the clustering of such a number of triggering factors was extremely unlikely to occur again so that the possibility of recurrence of sleepwalking with aggression was considered extremely remote.” Accordingly, Kenneth was acquitted of the murder of his mother-in-law and assault on his father-in-law. He was free to go.

On the face of it, the case appears straightforward. However, several of the details of the case were ignored during the trial. Kenneth was known for having some uncontrollable urges. Owing to his uncontrollable urges, he acquired a gambling debt and lost his job. His gambling misdeeds caused him to suffer from extreme anxiety that caused insomnia. He used to have a good relationship with his in-laws in part because when he first met their daughter, she was a run-away, and he convinced her to return. They continued to be grateful for his intervention. But after losing his job Kenneth stopped visiting them. He was ashamed and feared their abandonment. He was unable to find a new job. He did a few jobs on the side as an electrician and continued to gamble. He was charged with fraud but was out on bail at the time of the attacks and was awaiting his trial. His wife Karen wanted him to get help for his gambling and said she would leave him if his gambling continued. They had many fights over this. Under pressure from his wife, Kenneth arranged to join Gamblers Anonymous. On the day of the killing, Kenneth was supposed to go to a barbeque at his in-laws' house and tell them about his gambling, his debt, and his job loss. Could these factors have motivated Kenneth to drive 23 km to kill his in-laws in a fit of temporary insanity? His impulsivity does suggest a tendency toward psychosis.

The evidence collected prior to the trial suggests that Kenneth was indeed a sleepwalker. His wife Karen had no recollection of his sleepwalking but only of him talking to her in his sleep and also of him being a deep sleeper and being difficult to awaken. The EEG readings during his sleep tests were irregular, suggesting parasomnia. But irregular readings can also occur during sleep when people are under great stress and during substance abuse. Kenneth’s mother reported that when he was 13 to 14 years old she went in to check on him and his legs were going out of the 6th-floor window. This is the only case of sleepwalking that she could recall. However, Kenneth’s grandfather was a sleepwalker. He would walk around in the house and sometimes cook food without eating it.

The defense attorney alleged that Kenneth didn’t remember the details of the killing. But he said he remembered his mother-in-law’s face after he killed her. He apparently also realized that he had killed her when he arrived at the police station. Was Kenneth really fully asleep the whole time, or could he have been conscious the whole time but repressed the horrific memories almost immediately?

There is lots of evidence that supports the latter hypothesis. Sleepwalking occurs in the deep stage of sleep when slow brain waves (50 percent + delta waves) begin to appear. Because of the slow brain waves, people who are asleep are not normally consciously aware of sensory input from their surroundings. During sleep, there is also a gating mechanism that blocks input from the cognitive brain to the motor system. The chemical messenger gamma-aminobutyric acid (GABA) acts as an inhibitor that stifles the activity of the brain’s motor system.

However, in parasomnia, there is a defect in the gating mechanism that allows substantial input to the motor system. Owing to the failure of the gating mechanism the brain issues commands to the muscles during sleep. In children, the neurons that release this neurotransmitter are still developing and have not yet fully established a network of connections to keep motor activity under control. Sometimes the gating mechanism remains underdeveloped, or it functions less effectively owing to sleep deprivation, fever, anxiety, or drugs. In those cases, sleepwalking can persist into adulthood.

While slow delta brain waves occur during sleepwalking, it has also been found that there is a significant amount of high oscillation waves, just like in people who are fully awake. Sleepwalkers have their eyes open: They can see their environment but not consciously. While sleepwalkers are in a state of deep sleep, the part of the brain in charge of motion is awake. Only the part of the brain that correlates with awareness and cognition remains asleep. Sleepwalkers are essentially awake and asleep at the same time. As the cortex, which is the part of the brain that controls thinking and voluntary movement, is asleep during slow-wave sleep, the movements sleepwalkers make are controlled by other parts of the brain and are more or less reflexive.

The sleepwalker’s brain processes visual and auditory stimuli from his surroundings, but the processing of these sensory stimuli does not give rise to stable neuron activity. Because of this, the brain signals are not as strong as during waking states. This is why people normally can only complete tasks they have done hundreds of times before. To navigate safely in their sleep, sleepwalkers must be in their natural surroundings. When sleepwalkers go on vacation they often get hurt because the brain assumes that they are in their familiar surroundings. When they stay at new places sleepwalkers bump into walls or stub their toes. In April 2007 17-year-old Canadian tennis player Peter Polansky broke a window and crawled out through the broken shards and fell down while sleepwalking in his Mexico City hotel room.

These considerations cast doubt on the claim that Kenneth was fully asleep while he carried out his highly complicated actions. In criminal cases, it is often alleged that people who perform exceedingly complex actions in relatively unfamiliar surroundings and commit horrendous acts are sleepwalking. But equally complex actions in relatively unfamiliar surroundings are unheard of in non-criminal cases. It is unlikely that enough visual information about the environment can be unconsciously processed for a person to be able to complete extremely complex actions in relatively unfamiliar surroundings.

Kenneth hadn’t been to his in-law’s house since he lost his job two months before the killing. He would have had to make the drive unconsciously in relatively unfamiliar surroundings in the dark at night. During his drive, he encountered several major intersections that he would have had to maneuver unconsciously. This seems like an impossible task. Drivers completely lost in thought are not fully unconscious, they have at least some minimal awareness, yet they miss exits or find themselves having taken the most familiar route rather than one they planned. They suddenly find themselves at their old house, their child’s school or their workplace rather than the place they meant to go to.

Sleepwalkers don’t normally take long trips during their sleep. Kenneth’s grandfather who was a sleepwalker never left the house when he was sleepwalking. Though there were several people with parasomnia in Kenneth’s family, few were sleepwalkers and only one, a second cousin, left the house during her sleepwalking. She made it outside during her sleep and just sat there. No one went on long trips or ended up carrying out highly complex actions that appear to require conscious control. As district attorney Chris Frisco said in the Stephen Reitz case in which Stephen Reitz stabbed his wife while allegedly being asleep, to carry out relatively complex acts such as figuring out which end of the knife to hold and which end to use to stab a person with, you have to have some conscious control of what you are doing.

It is also highly dubious that Kenneth could have stayed asleep during the whole ordeal. It’s a myth that you cannot awaken a sleepwalker. It may be unwise to awaken them because they may end up confused or terrified. But it is no harder to awaken a sleepwalker than to awaken any other person in a state of deep sleep. When Kenneth arrived at the police station, severe cuts to his hands attested to a struggle with his in-laws. The police later described signs of a great struggle in the bedroom: The bed was disheveled, the pillows were soaked in blood, and the mattress was moved around so that the headboard was tipped forward. His mother-in-law Barbara was found in a room 5-6 feet from the bedroom. It is just plainly implausible that a severe struggle with his in-laws, them screaming at him, asking him what he was doing, pleading with him, failed to wake him up.

It is true that the brain may unconsciously go into defense mode if startled, but Kenneth fetched a tire iron from his car trunk and used his key to let himself into the house. He choked his father-in-law, beat up his mother-in-law with the tire iron, and stabbed both of the in-laws repeatedly. Barbara sustained six stab wounds through her chest, one through her shoulder blade, and a fatal wound through her heart. These gruesome acts are not simply the result of the brain unconsciously going into defense mode after being startled. The struggle by itself should have been enough to wake up a person in deep sleep in the next room.

At the time of the attacks, Kenneth and his wife Karen were having frequent arguments about his gambling, and Kenneth was asked to sleep on the couch. But that night after watching television together on the couch, Karen had invited him to sleep with her in the bedroom when she went to bed after midnight. Kenneth refused, saying he preferred to wait until he had gotten help for his gambling problem, and besides, he said, he wasn’t sleepy.

At the trial Kenneth reported the night’s events as follows: He fell asleep on the couch sometime after midnight, and his next recollection was seeing his mother-in-law’s face, she had her eyes and her mouth open. Kenneth described it as a very sad face. The fact that Kenneth actually remembers his mother in-law’s face is an important detail overlooked by the prosecution. It suggests that his subsequent dissociative analgesia at the police station, which was a core piece of evidence in the trial, most likely was not due to being in a sleepwalking state. The dissociative analgesia was more likely the result of being in a state of shock or great distress. Kenneth said that after seeing his mother-in-law’s face, he just sat there. He then heard the kids yelling. He recalled thinking the kids were in trouble and needed help. He said he yelled “kids, kids, kids” and went upstairs. The kids didn’t hear him yell. They heard only grunting noises of the kind sometimes made by a person during sleep. But this is very odd. If Kenneth had really been sleepwalking at the time at which the kids were yelling, he would not have been able to recall the moment so clearly afterward. If, on the other hand, he was awake at the time, then he wouldn’t have made the grunting noises, he would have yelled just as he remembered he did.

The evidence suggests that Kenneth’s actions were not entirely automatic. He must have been at least partially conscious but owing to the gruesome nature of his actions he might have repressed his memories of the details. Paula Pinckard who shot her 11-year-old daughter Aubrey to death before shooting herself in their Rock Hill home in March 2000 during a Prozac-induced psychosis had to undergo hypnosis to remember any of the terrible events. Like Kenneth, she was not showing any pain symptoms from her wounds when the paramedics showed up. She was charged with first-degree murder and the death penalty but the judge found her Not Guilty by Reason of Insanity and committed her to the custody of the State Department of Health and Hospital, Forensic Division.

People who are insane cannot intend to perform a criminal act because they either don’t know the act is wrong or they cannot fully control their actions. It is plausible that owing to temporary insanity Kenneth was not fully in control of his actions even if he was conscious of them. His actions then were voluntary but unintentional.

If indeed Kenneth went temporarily insane at the moment of the killings, he still is not guilty by reason of insanity. But morally he nonetheless is not without fault. After starting to gamble heavily, Kenneth stopped socializing, he began to suffer from pressure headaches, and his weight increased from 240 to 310 pounds. He suffered from insomnia. After going to bed at 11 p.m. he would watch television until 2 a.m. and then fall asleep and sleep four to six hours. His sleep condition improved after his daughter was born in December 1986. Then in March 1987, his mental health deteriorated. He had trouble sleeping on most days. He would wake up and feel like there was heavy breathing and pressure on the chest. He saw a doctor and was given treatment for asthma. But the symptoms are more likely symptoms of a severe anxiety disorder. There were clearly underlying emotional problems that he should have sought treatment for.

Kenneth’s social history testifies to this hypothesis. His parents separated when he was 5 and his mother subsequently remarried. He had no contact with his father until he was 18. He never had a close relationship with his mother. He was a troubled child. He didn’t do well in elementary school. In Junior High School he was arrested once for petty theft. At age 16, his parents moved away and he moved in with his grandmother. It is plausible that an underlying emotional conflict gave rise to Kenneth’s uncontrollable gambling urges, anxiety disorder, and later manslaughter.

Hallucinations During Sleepwalking

There is another possibility entirely consistent with the events in the Kenneth Parks case. A common assumption about sleepwalkers is that they are deeply asleep and therefore act as automatons. There are, however, lots of verbal reports from sleepwalkers suggesting that hallucinations, or dream-like states of mind, often take place during sleepwalking.

Alyson Bair, a 31-year-old woman from Idaho, was having a nightmare that she was in a deep river and was getting tired. In her dream, she realized that she was drowning. Suddenly she woke up and realized that she actually was drowning. She had been sleepwalking and had ended up in the river outside her home.

"I thought I was dreaming, but then I realized I wasn't and I was scared," Bair told She was eventually able to make her way back to the riverbank and stay put until she was found the next morning.

Another case in which a sleepwalker thought she was dreaming was that of 36-year-old Welsh Australian artist Lee Hadwin. He would get up at night in his sleep and produce surrealistic and fantastical artworks. Usually, he would not have any recollection of having completed these works overnight but would find them there the next day. He has had several requests from galleries and museums and now goes by the same “Kipasso.” Apparently Lee Hadwin has no interest in painting or any ability to do it during the day.

Though he rarely has any recollection of what he has been doing during his nightly art sessions, he reports that he does frequently have very vivid dreams of painting while asleep. He was being sleepwalking since he was a child. In most children sleepwalking ceases with age but for Lee Hadwin, it continued into a way of living that soon will make him more money than his day job.

Contrary to the defense’s story in the Parks case, Kenneth Parks may also have had several hallucinations during his sleepwalking. He said he yelled “kids, kids, kids” after hearing the kids yelling. But they only heard him make grunting noises. He also saw his in-law’s face as sad, which may have been a hallucinatory state of mind. It is thus possible that Parks’ verbal reports reflect hallucinations rather than an attempt to mislead the jury.

If indeed sleepwalkers don’t act as complete automatons during sleepwalking, this raises the question of to what extent they can actually be held responsible for their actions, even if they really are truly asleep and truly sleepwalking while committing their crimes or masterpieces.

We are in fact very rarely fully in control of our actions. Studies revealing our limited rational behavior are not hard to come by. Dan Ariela (2008: 177), for example, reports on studies showing that in countries in which people applying for a driver’s license are asked to check a box if they want to join an organ donor program, they don’t check the box and they don’t join. In countries in which people applying for a driver’s license are asked to check a box if they don’t want to join an organ donor program, they don’t check the box and they join. The reason people make these decisions is that they don’t want to deal with the complex issue of organ donation and so remain content with the option already chosen for them.

To test whether experts make similarly bad decisions, Ariela ran a parallel study with physicians. The physicians were divided into two groups. Both groups were presented with a case of a patient with hip pain. The physicians were told that they had sent him off to hip replacement surgery after experimenting with several medications. The first group was then asked what they would do if they suddenly realized that they had not tested the efficacy of ibuprofen. The majority of participants in this group chose to cancel the hip replacement surgery in favor of experimenting with this medication. The second group was asked what they would do if they realized that two medications hadn’t been tested. The majority of participants in this group chose to let the patient have hip replacement surgery. Apparently, the increased complexity of canceling the surgery made them chose not to do so.

The real reasons for acting the way we do need not be accessible to introspection. When we introspect it seems that we were acting for rational reasons. The reasons we cite as reasons for action, however, are in many cases pure constructs. In one study, Johansson and colleagues presented pairs of images representing female faces to participants. The participants were asked to pick the face they found more attractive. In three of the 12 trials, the researchers contrived to treat the photo that the participants did not pick as though they did pick it. When people were asked to explain their choices for three of the non-manipulated trials and the three manipulated trials, the explanations were nearly indistinguishable.

These types of studies cast doubt on the extent to which we act on the basis of reasons to which we have conscious access. The perhaps most impressive case of the apparent failure of being the agent of our actions is the case of slow walkers. In one study, participants were given a puzzle and asked to construct grammatical sentences out of randomly ordered words. One group received a version containing words associated with stereotypes of the elderly, such as wrinkled, gray and Florida. The other group received a test containing only age-neutral words, such as private, thirsty and clean. The study showed that participants who were given the puzzle with the geriatric words were walking significantly slower afterward than the participants who were provided with the age-neutral vocabulary.

If indeed agents have conscious awareness during sleepwalking and they have a motive and cannot be acquitted on the grounds of insanity, the question becomes how much the actions of sleepwalkers differ from actions we all engage in. We might never have full autonomy when we act. It is possible that sleepwalkers who commit crimes or produce artwork or food have some autonomy and hence only differ from people who are awake in degree rather than in kind. If indeed this is the case, then maybe full acquittal for crimes committed while people are sleep should never be an option.

The cases of extreme sleepwalking testify to the amazing capabilities of the mind. Even when you sleep your mind can go on a mission and complete the real-life projects you didn’t have the balls to complete while awake.

By Berit Brogaard and Kristian Marlow

More from Psychology Today
4 Min Read
Certain elements of temperament, including shyness, can be detected as early as 4 months of age, suggesting it's largely inborn.

More from Berit Brogaard D.M.Sci., Ph.D

More from Psychology Today