What Isolation Does to the Brain: The Neuroscience of Solitary Confinement

Imagine a room approximately the size of a parking space. You are in it for twenty-two to twenty-four hours a day. There is a slot in the door for food. There is no meaningful human contact. There is nothing to do. This continues not for a day, not for a week, but for months or years at a time.

This is solitary confinement. It is currently used in prisons and jails across the United States, and on any given day, an estimated eighty thousand people are being held in some form of isolated confinement in this country. [1]

For most of its history, the debate around solitary confinement has been framed as a question of prison management. Some inmates are dangerous. Some need to be separated. This is a security issue, not a science issue.

Neuroscience has a different view.

What Solitary Confinement Actually Is

The formal term used by correctional systems is restrictive housing or segregated housing. The conditions vary by facility, but the core features are consistent: extreme limitation of movement, near total absence of social interaction, sensory environments that are either completely bare or filled with constant noise and artificial light, and little to no meaningful activity. [2]

People are placed in solitary for a wide range of reasons, including disciplinary violations, suspected gang affiliation, protective custody, and in some cases simply because a facility has no other place to put them. Stays that were once measured in days have routinely extended to years. Some people have spent decades in solitary confinement without ever being convicted of an additional crime. [3]

The question neuroscience asks is straightforward. What does this actually do to a human brain?

Lens One: The Scientific Evidence

In 1983, a psychiatrist named Stuart Grassian was asked to evaluate inmates at Walpole State Prison in Massachusetts who were being held in solitary confinement. What he found became one of the most important documents in the scientific literature on isolation.

Grassian documented a consistent cluster of psychiatric symptoms appearing in inmates who had no prior history of mental illness. The symptoms included hypersensitivity to stimuli, perceptual distortions, paranoia, difficulty with thinking and concentration, intrusive thoughts, and in some cases frank hallucinations. He called this pattern SHU syndrome, named after the Security Housing Unit where it most commonly developed. [4]

Subsequent researchers replicated and expanded his findings. Craig Haney at the University of California Santa Cruz conducted extensive interviews with inmates at Pelican Bay State Prison and found that the vast majority of people held in long-term solitary developed serious psychiatric symptoms, regardless of their mental health status before isolation. [5]

A 2020 study published in The Lancet examined health outcomes for formerly incarcerated people and found that time spent in solitary confinement was independently associated with significantly elevated risk of self-harm and suicide, even after controlling for other variables. People who had been held in solitary were more than six times as likely to die by suicide in the first year after release than those who had not. [6]

These are not minor side effects. These are the outcomes of a practice applied to tens of thousands of people every year.

Lens Two: The Neuroscience

To understand why isolation produces these effects, you need to understand something fundamental about what the human brain is built for.

The human brain is a social organ. This is not a metaphor. Human beings evolved in intensely social groups, and the brain developed in that context. Social connection is not a preference the brain has on top of its other functions. It is woven into the architecture of how the brain regulates stress, processes threat, and maintains stability.

Research by neuroscientist Naomi Eisenberger at UCLA demonstrated that social pain and physical pain activate the same neural pathways in the brain. When a person experiences rejection or isolation, the anterior cingulate cortex and the anterior insula respond in the same way they respond to a physical injury. [7] The brain does not treat loneliness as an inconvenience. It treats it as damage.

When social contact is removed entirely, the brain's threat response systems activate and stay activated. Cortisol, the primary stress hormone, remains chronically elevated. Chronic cortisol elevation over long periods has well-documented effects on the brain: it damages neurons in the hippocampus, which is central to memory and learning, and it disrupts the functioning of the prefrontal cortex, which governs judgment, impulse control, and emotional regulation. [8]

Sensory deprivation compounds this further. The brain requires input to function normally. In environments where stimulation is severely reduced, the brain begins generating its own, which is the neurological mechanism behind the hallucinations and perceptual distortions that Grassian documented. This is not mental illness appearing from nowhere. It is the brain responding, in a completely predictable way, to conditions it was never designed to survive. [4]

The implications of this are serious. Solitary confinement is used in part to manage people who are considered dangerous or disruptive. But the neuroscience shows that the conditions of solitary confinement actively degrade the very brain systems responsible for emotional regulation and behavioral control. The practice that is supposed to make dangerous people more manageable is neurologically designed to make people less stable, less controlled, and more likely to decompensate.

Lens Three: The Legal Interpretation

The Eighth Amendment to the Constitution prohibits cruel and unusual punishment. The question of whether solitary confinement meets that standard has been litigated for decades, with inconsistent results.

In Madrid v. Gomez in 1995, a federal judge found that conditions at Pelican Bay's Security Housing Unit were unconstitutional as applied to mentally ill inmates, writing that solitary confinement may well hover on the edge of what is humanly tolerable. The same judge declined to find the conditions unconstitutional for the general population, reasoning that extreme measures could be justified by security needs. [9]

The Supreme Court addressed procedural protections for solitary confinement in Wilkinson v. Austin in 2005, ruling that inmates must be given some process before being placed in the most extreme isolation conditions. The Court did not rule on whether solitary confinement itself violated the Eighth Amendment. [10]

That core constitutional question has never been definitively resolved by the Supreme Court. Lower courts have continued to disagree. In the meantime, the practice has expanded.

Some progress has come through legislation and settlement rather than constitutional ruling. Colorado effectively banned long-term solitary confinement in 2017 through a combination of legislative reform and a settlement agreement. New York passed the HALT Solitary Confinement Act in 2021, limiting solitary stays to fifteen days and establishing alternative therapeutic housing units. Several other states have passed similar restrictions. [3]

The United Nations Standard Minimum Rules for the Treatment of Prisoners, known as the Nelson Mandela Rules, define solitary confinement exceeding fifteen consecutive days as cruel, inhuman, or degrading treatment and prohibit it under international human rights standards. The United States has not adopted those standards domestically. [11]

Putting All Three Lenses Together

The scientific evidence is extensive and consistent: solitary confinement produces severe psychiatric harm in a large proportion of the people subjected to it, including people with no prior mental illness. The neuroscience explains why this happens at the level of brain architecture: the human brain requires social contact and sensory input to function, and removing both triggers biological damage to the systems that regulate emotion, behavior, and reality. The law has acknowledged the problem in pieces, through individual court decisions and some state reforms, but has not produced a national standard that reflects what the science has established.

The result is that a practice with a documented neurological mechanism for producing psychiatric disability continues to be used at scale, in facilities across the country, on people whose brains are being measurably damaged by it.

There is a particular cruelty in what the neuroscience reveals here. Solitary confinement is often justified as a tool for managing people who pose a danger. But by degrading the prefrontal cortex and disrupting stress regulation, it predictably produces people who are more impulsive, more reactive, and less capable of the emotional control that safe reintegration requires. The stated justification and the actual outcome point in opposite directions.

What we do to people in solitary confinement, we do knowing what it will cost them. The science has been available for decades. The choice to continue is not a result of not knowing. It is a result of deciding that knowing is not enough reason to stop.

The next post in this series will look at traumatic brain injury, and the question of what happens when the organ responsible for judgment and behavior is physically damaged before the crime ever occurs.

Sources

[1] The Sentencing Project, solitary confinement statistics: sentencingproject.org

[2] Vera Institute of Justice, "Solitary Confinement: Common Misconceptions and Emerging Safe Alternatives": vera.org

[3] American Civil Liberties Union, "Alone and Forgotten": aclu.org/report/alone-and-forgotten

[4] Grassian, S., "Psychopathological effects of solitary confinement," American Journal of Psychiatry, 1983: ajp.psychiatryonline.org

[5] Haney, C., "Mental Health Issues in Long-Term Solitary and Supermax Confinement," Crime and Delinquency, 2003: journals.sagepub.com

[6] Zlodre and Fazel, "All-Cause and External Mortality in Released Prisoners," The Lancet, 2012: thelancet.com

[7] Eisenberger, N.I., Lieberman, M.D., and Williams, K.D., "Does Rejection Hurt? An fMRI Study of Social Exclusion," Science, 2003: science.org

[8] NIH, chronic cortisol and hippocampal damage: pmc.ncbi.nlm.nih.gov/articles/PMC3181836

[9] Madrid v. Gomez, 889 F. Supp. 1146 (N.D. Cal. 1995): law.justia.com

[10] Wilkinson v. Austin, 545 U.S. 209 (2005): supreme.justia.com/cases/federal/us/545/209

[11] United Nations Standard Minimum Rules for the Treatment of Prisoners (Nelson Mandela Rules), 2015: unodc.org/documents/justice-and-prison-reform/Nelson_Mandela_Rules-E-ebook.pdf