AI-powered brain emulation is changing our definition of death

By Philip Jaekl

Our definition of death is changing. Until the concept of brain death was realised in the 20th century, death was primarily presumed upon the absence of a pulse and breathing. Prior methods for detecting the presence of a heartbeat were not well founded and not always effective.

This led to some rather strange additional criterion. In the 18th century for example, a variety of odd methods were engaged to hopefully determine if someone was dead. Expiry was declared if there was no response to tickling with a feather quill, whipping with nettles, mouth washing with urine, sticking needles under the toenails or even a tobacco smoke enema. The belief that breathing imparts vitality is also documented in the 18th century. For example, in a 1791 Annual Register which instances the misconception that cats can steal babies breath – a misconception that is still held today. It states “...on the coroner’s inquest, the child died in consequence of a cat sucking its breath.”

Until the 19th century it was held that consciousness was supported by ‘animal spirits’ – the pumping of fluids in the nervous system, controlled by the brain. Then it was discovered that nervous system activity was sustained electrically.

The famous Italian physician and scientist, Luigi Galvani, and his followers conducted experiments to investigate connections between life, death and electrical conduction through nerves. Their explorations culminated in experiments conducted in front of large audiences where they attempted to reanimate executed prisoners. Such experimentation was highly controversial as critics believed that life and death should ultimately be in the hands of God, not humans. This controversy would, however, inspire Mary Shelley’s Frankenstein.

Despite these advances in nervous system understanding, knowledge of the now well-known process of brain death had yet to unfold. Misattributions of death were still relatively common. So much so that various ‘escape hatch’ coffin designs were available for those with taphephobia – fear of being buried alive.

Now, thankfully, there is a deeper scientific understanding of heart and lung functioning and evidence-based methods of cardiopulmonary resuscitation. Since the invention of bypass machines in the 1950s, which can artificially maintain circulation and respiration, death has come to be defined by an absence of brain activity – brain death. Again, the concept appears to be straightforward: no brain activity and no pulse equals no life. But it’s far from simple.

Today we know that brain activity does not cease at once and that current methods of measurement, such as electroencephalography, are not always definitive. For example, patients who suffer cardiac arrest, are often hypothermically treated to slow down their metabolism and thereby prevent brain damage due to lack of oxygen. Although they can achieve a full recovery, they may undergo a temporary but complete absence of monitored brain activity in the cerebral cortex while their brainstem – for which activity is difficult to accurately monitor – may continue to maintain some basic functions.

Take, for example, the ambiguity of a 2011 case involving a 55 year-old man who was pronounced brain dead after cardiac arrest. The case report reads: “Twenty-four hrs after brain death pronouncement, on arrival to the operating room for organ procurement, the patient was found to have regained corneal reflexes, cough reflex, and spontaneous respirations” (these functions are controlled by the brainstem). It goes on to state, “The care team faced the challenge of offering an adequate explanation to the patient's family and other healthcare professionals involved.”

“We don’t have a really great way of determining when brain function is gone and when that is irreversible. We can do a pretty good job but we’re not always perfect,” says Adam Webb, a neurologist at Emory University Hospital in Atlanta, who experienced first-hand the apparent reanimation of this patient. “And this case report shows that there are things that can confound our ability to do that and become overconfident, or that can cloud the presence of brain function temporarily. After cardiac arrest it can be very muddy and hard to tell.” The case report was written to promote discussion and thereby find alternative or advanced methods.

The ambiguity of measured brain activity as a criterion for death is also realised by its dependence on jurisdiction. A 2015 study concludes that “substantial differences in perceptions and practices of brain death exist worldwide”. Consider, for instance, the 2013 case of Jahi McMath, declared dead at the age of 13 after an unsuccessful surgery in a hospital in California.

Yet, while remaining on life-support she showed signs of responsiveness and was transported to a hospital in New Jersey, where it is possible to declare a religious exemption from life support withdrawal. Thus, she was officially dead in California but not in New Jersey.

Her condition continues to be debated by neurologists today, five years later, as she continues to remain on life-support. The family has since taken over 50 videos that they claim demonstrate her ability to consciously answer questions, often by leg or finger motion. Yet, analysis of her neural activity and neural structure remains inconclusive in terms of whether or not she should be declared brain dead.

But even permanent brain death, which is considered undisputed, does not hold as an adequate criteria for some. The field of cryonics rests on the belief that if one’s brain is frozen and kept intact, future scientific advances will enable reanimation. Although often considered outlandish, new advances are being made that involve a process called aldehyde-stabilised cryopreservation.

Using this method, a pig’s brain was recently preserved in fidelity high enough to maintain its connectome – the connections between neurons vital for communication across networks thought to preserve aspects crucial to human personality, memory, perception and identity.

Cryonics assumes a vital connection between brain and mind: no functional brain equals no functional mind. Novel approaches to neurobiology based on information-theory, however, go a step further, making the bold claim that it is possible to accurately emulate the brain’s connectivity, and therefore the mind, in the form of digital information.

According to this perspective, permanent death occurs only when the neural connections that support one’s memory, personality and self are annihilated. It is underpinned by the emerging possibility that one’s connectome can be scanned entirely, transformed into a digital code and then ‘uploaded’ to any new, viable substrate.

This view on existence advocates the possibility of existence of consciousness as a primary criterion for being. “The way we think about life and death doesn’t currently have much to do with consciousness,” says Randal Koene, neuroscientist and neuroengineer co-founder of Carboncopies, an organisation for the advancement and development of substrate independent minds. “If in the future we were to talk to an artificial intelligence and, for all we know, the AI genuinely seems conscious – aware and intelligent – how relevant is the question of whether it’s alive or dead? Must we call the AI alive? Or is it a dead thing that is conscious?”

The boundary between life and death seems like it will never be clear. Yet the connectome and the information it contains regarding critical aspects of consciousness mark a digital-age transcendence of current thinking regarding biological brain and heart activity as supporting life. In the coming decades, defining death will likely become more and more difficult. /wired.co.uk