Anechoic Chamber Safety – Sensory Deprivation and Institutional Duty of Care

In February 2020, a university department dedicated to the study of the human mind published a safety checklist for its soundproof anechoic chamber. The register ran to fire risks, heavy doors that could trap a student inside, and rising room temperatures in a sealed space. It did not mention that, according to more than fifty years of clinical data, the silence inside that same room can trigger hallucinations and paranoia in under fifteen minutes. The department was the University of Melbourne’s School of Psychological Sciences. The document was version 1.5 of its MSPS OHS Risk Register.

That gap is the investigation.

The Hardware Baseline

In January 1944, the United States Navy sealed a seventeen-year-old seaman named Nathan Schnurman into a ten-foot by ten-foot chamber to test protective clothing. He spent one hour a day in there for five days, breathing sulphur mustard and Lewisite. Schnurman asked to come out but was refused.

On the fifth day, he collapsed into cardiac arrest inside the chamber. The Physicians Committee for Responsible Medicine logged the case in its Ethical Science report, but the record only reached the public in 1993, forty-nine years after the door was first shut.

The thing to notice in the Schnurman file is what the clipboard on the outside of that door was measuring. Not his pulse or his distress but his clothing. Schnurman was inside a sealed box as the instrument by which the fabric was being tested, and when the instrument failed, the test continued. That same chamber-as-apparatus frame is what acoustic engineers bring to the rooms in the rest of this investigation. The stimulus changes from gas to silence. A locked door stays locked for the same reason.

The NASA Failure Rate

On 1 March 1967, a research team at Yeshiva University, Fisher, Richlin, Weinstein and Weisinger, published a NASA-funded report titled The Effects of Sensory Deprivation on Sensory, Perceptual, Motor, Cognitive, and Physiological Functions. It carries the NASA Technical Report identifier 19670009327. The design was simple. One hundred and thirty subjects were scheduled for seventy-two hours of isolation. The stated aim was to understand what silence and stillness would do to the human nervous system over time.

The outcomes are stacked on page 19 of the report.

Eighty per cent of subjects reported general anxiety. Forty-one per cent reported visual hallucinations. A further five per cent reported auditory hallucinations. Ninety per cent reported severe time disorientation. These are not fringe outcomes. The central findings in a table that sat in the public technical record for fifty-nine years before this investigation went looking for it.

The most uncomfortable figure on the page is the withdrawal rate. Fifty-three per cent of the one hundred and thirty subjects could not complete the seventy-two-hour isolation. They asked to come out early, citing claustrophobia, restlessness and severe time disorientation.

More than half of a healthy volunteer pool walked away from a silent room before the clock ran down. That number has been on NASA’s own server, in a scanned PDF of the original typescript, since the study was published.

The 15-Minute Trigger

In October 2009, researchers Oliver Mason and Francesca Brady of University College London’s Department of Clinical, Educational, and Health Psychology published The Psychotomimetic Effects of Short-Term Sensory Deprivation, catalogued on PubMed as identifier 19829208. The method involved placing nineteen participants in an anechoic chamber (a room designed to completely absorb sound reflections) in complete darkness for fifteen minutes.

Fifteen minutes is the length of a coffee break. It is shorter than a GP appointment. Inside that window, the Mason and Brady subjects reported perceptual disturbances, paranoia and anhedonia, a term clinicians use for the loss of the ability to feel pleasure. Visual and auditory hallucinations were logged, with stronger effects among participants already prone to hallucinations. The point of the study was not to prove that long isolation is dangerous. Clinicians already knew that. The point was to show that the trigger fires almost immediately.

The legal frame was already in place before the clinical trigger was measured.

On 2 March 1972, the UK government published the Parker Report into interrogation methods used on detainees in Northern Ireland. Those techniques included hooding and subjection to noise and silence, grouped under the heading of sensory deprivation. On the day of publication, the Prime Minister announced their abandonment. Some years later, the European Court of Human Rights ruled, in Ireland v. The United Kingdom, that those techniques amounted to inhuman and degrading treatment. The same class of stimulus the Mason and Brady study applied for fifteen minutes, to consenting volunteers, had already been classified as abusive when applied to prisoners.

The 45-Minute Marketing Myth

On 3 April 2012, the Daily Mail ran an article on Orfield Laboratories in South Minneapolis, a facility that had held the Guinness World Record for the quietest place on Earth since 2004. The piece quoted the facility’s founder, Steven Orfield. A single line travelled: the longest anyone could bear the chamber was forty-five minutes, after which subjects became disoriented by the sound of their own bodily functions and began to hallucinate. Catchy figures travel well. The claim propagated across Vice, Good.is and dozens of aggregator sites, cementing the chamber’s reputation as a test of endurance.

At first glance, the forty-five minute claim looks like clinical data. Clinicians have shown that short exposures are risky. A stated hard limit sits plausibly alongside that.

What turned the picture was the current Orfield Laboratories website. The visit page offers one-hour Group Sessions of Silence and Private Group Sessions of Silence, both lasting a full sixty minutes. A separate disclaimer states plainly that there is no Guinness World Record for time spent in the chamber, and that previous press stories regarding a forty-five minute limit were incorrectly reported.

The forty-five minute limit was not a medical finding. A press narrative had outlived its usefulness. When Orfield reclaimed the Guinness record in November 2021 with a new measurement of minus 24.9 A-weighted decibels, the facility was already pivoting towards paid therapeutic sessions for customers with sensory processing disorders. An hour-long session cannot be sold if forty-five minutes is the breaking point. So the figure went, quietly, on the facility’s own site.

Earlier press stories remain online. Two versions of the claim now sit in separate corners of the internet, pointing in different directions.

The Administrative Blind Spot

On 14 February 2020, the University of Melbourne’s School of Psychological Sciences published the MSPS OHS Risk Register, version 1.5. The document is a PDF hosted on the school’s own web server. It runs to a single page of hazard rows, control measures, and sign-off boxes. A department that studies the human mind for a living owns the chamber the register covers.

The hazards it lists are physical. Entrapment from heavy soundproof doors. Failure to hear emergency alarms through thick acoustic walls. Exposure to fire. Rising internal temperatures in a sealed, ventilation-limited space. The controls are equally physical. A warning light system wired to the building’s emergency panels, quick-release doors and a direct telephone landline inside the chamber. A buddy system, so that no student enters alone. Mechanical ventilation to keep the air moving. Every item reads as competent occupational health thinking about a soundproof room.

The omission sits in what is not there. There is no row for psychological distress. No row for disorientation. Nothing for anxiety, nothing for hallucinations. No duration limit, no mandatory break, no psychological pre-screening. A psychological science department audited an acute sensory deprivation environment without logging a single psychological hazard, on a document where those columns would have been entirely natural. The clinical literature that would have populated those columns was being produced, in part, by people working in the same kind of building as the authors.

The process, as far as the document can reconstruct it, is template-led. Occupational health and safety auditors work from pre-written hazard libraries. Those libraries contain fire, entrapment, electrical and temperature categories. They do not contain an entry for the acute absence of sensory stimulus, because no generic workplace produces that condition. When the auditor walked into the chamber with the standard template, the template returned the hazards it could see. A soundproof door registered as a mechanical trap. The silence itself did not register at all. This is the researcher’s reading of how a register ends up shaped this way. Whether the authors of the Melbourne document were ever shown the Mason and Brady paper is not known.

The Regulatory Vacuum

Microsoft’s Building 87 in Redmond, Washington opened in 2015. It contains multiple anechoic chambers, one of which held the Guinness World Record for the quietest environment on Earth at minus 20.6 decibels, a reading closer to the theoretical limit of Brownian motion than to ordinary human hearing. Microsoft uses the facility for characterising microphones and evaluating the Cortana speech interface. Human factors work there is led by Principal Human Factors Engineer Hundraj Gopal and Senior Engineer LeSalle Munroe, as described in the Brüel & Kjær profile of the room.

Engineers inside that room spend working hours in an environment that is quieter than the one the Mason and Brady experiment used to trigger psychotic-like experiences in fifteen minutes.

What happens on the ethical oversight side of that arrangement is not documented in the public record. A clinical trial running experiments inside such a chamber would require Institutional Review Board approval and specific waivers covering hallucinations and spatial disorientation. No equivalent paperwork for Building 87 engineers can be found in public sources. Whether that paperwork exists behind the corporate firewall is a separate question, and the answer is not yet on the record. The available evidence does not confirm whether engineers testing microphones are subject to different ethical oversight than clinical researchers running shorter experiments in comparable rooms. Veriarch approached Microsoft for comment but received no response.

Above the facility-by-facility level, the regulator is silent. Section III, Chapter 5 of the OSHA Technical Manual lays out detailed standards for workplaces exposed to noise above 85 A-weighted decibels, covering hearing loss, monitoring and hearing protection. The equivalent standard for the absence of sound does not exist. OSHA does not audit facilities for negative decibel exposure because negative decibel exposure is not a category the federal safety rules recognise. A worker who spends a shift next to a compressor is covered. A worker who spends a shift inside a minus 20.6 decibel chamber is covered only by whatever the employer volunteers.

The Panic Button Paradox

In 2010, Vaughan Bell of King’s College London’s Institute of Psychiatry published a critique of the Mason and Brady study in the Journal of Nervous and Mental Disease. His argument was narrow and awkward.

The 2009 experimental design had included a physical panic button inside the chamber, available to any subject who needed out. Bell pointed to earlier studies showing that the mere presence of a panic button raises baseline anxiety, because the button tells the subject, silently, that the room is considered dangerous. On Bell’s reading, the hallucinations were at least partly a product of the anxiety the button induced, not the silence itself.

Bell’s critique did not sink the finding. Later studies replaced the panic button with a one-way microphone, allowing researchers to monitor subjects without alerting them to an emergency protocol within the chamber. Those studies, summarised in the PMC research article catalogued as 4354964, still recorded a significant increase in psychotic-like experiences during sensory deprivation. Silence did the work. A button on the wall was a compounding factor rather than the cause.

What Bell’s observation means is that the safety design is awkward. The University of Melbourne registers orders a direct emergency landline inside the chamber, along with a buddy system and warning lights. In one direction, the handwritten logic is sound. If a student locked inside a soundproof room has no way to call for help, that is a safety failure. In the other direction, the logic runs straight into Bell. An emergency phone on the wall signals danger to the subject on its own. A person inside is told, in effect, that the room is the kind of place where a phone might be needed. The safety device and the psychological trigger are the same object.

Sources

Sources include: the Physicians Committee for Responsible Medicine’s ‘Ethical Science’ report regarding the 1944 Schnurman exposure; NASA Technical Report 19670009327 on 72-hour isolation outcomes; the 2009 Mason and Brady study on short-term sensory deprivation (PubMed identifier 19829208) and Vaughan Bell’s 2010 critique in the ‘Journal of Nervous and Mental Disease’; the 1972 Parker Report and the European Court of Human Rights ruling in ‘Ireland v. The United Kingdom’; 2012 press archives from the ‘Daily Mail’ alongside current booking disclaimers from the Orfield Laboratories official website; the University of Melbourne School of Psychological Sciences MSPS OHS Risk Register, version 1.5; the Brüel & Kjær technical profile of Microsoft Building 87; and the OSHA Technical Manual (Section III, Chapter 5).