New Delhi: Somewhere in the cells of our brains lies the key to what we call sentience, or the ability to sense, feel, and be conscious of what is happening around us. To better understand how the brain works, scientists have now cultured human and mouse neurons in a dish and trained them to play a computer game outside the body.
The game, called Pong, is based on table tennis. It requires the player to move a paddle up and down and hit a ball across the board. One of the earliest computer games, Pong is credited with contributing to the growth of video game parlours during the 1970s and 1980s.
In the new study, carried out by an international team led by scientists in Melbourne, 80,000 neurons were mounted on an array of tiny electrodes in a dish, called DishBrain, and trained to return the ball in Pong.
“Harnessing the computational power of living neurons to create synthetic biological intelligence (SBI), previously confined to the realm of science fiction, may now be within reach of human innovation,” the team writes in the journal Neuron.
But does the ability of neurons to play Pong really amount to sentience, and why does it matter? While the answer to the first question depends on how one chooses to define sentience, such experiments can have implications for medicine by telling us more about how drugs affect the brain.
The lessons
DishBrain has been developed by Cortical Labs, a biotech start-up based in Melbourne. It works by an exchange of signals via the electrodes.
“We apply electrical stimulation to the cells via a microelectrode array that represents where a simulated pong ball is relative to a paddle. We then measure how the cells are responding,” Brett Kagan, chief scientific officer with Cortical Labs, told HT in an email response.
The human cells performed slightly better than the mouse cells. Kagan called it a small but significant difference and said further research is necessary to understand this better.
To give them an incentive to play, the cells were sent unpredictable electric signals if they missed the ball. If they did return it, the signals would be predictable. Eventually, the neurons learnt to work towards a goal: To make their environment predictable with their actions.
This is called the “free energy principle” developed by professor Karl Friston of University College London, one of the co-authors of the study. Simply put, it proposes that the cells try to minimise unpredictability, Friston said.
“Just learning to return the ball means that the sensory input becomes more predictable, because you know where the ball will be when you are playing the game properly. If you miss the ball, it reappears in an unpredictable position and, under some protocols, results in unpredictable electrical stimulation. In summary, one can describe the sentient behaviour very simply as learning to minimise unpredictable, surprising sensory input,” Friston told HT.
Is it sentience?
In the scientific sense, yes. “Sentience per se is just ‘sense making’, which is an apt description of the elemental kind of sentient behaviour evinced by the dish experiments,” Friston said.
And Kagan added, “The word sentience can be used in different ways and for us, we do not mean that the cells are conscious or have any subjective awareness in a way we would understand it, only that they respond in interesting ways to incoming information.”
Will the behaviour of neurons in a dish tell us more about how neurons in a brain are working? “It is very likely,” Kagan said, “but it needs more research to be 100% sure.”
Why it matters
The scientists will next investigate what happens when DishBrain is affected by medicines and alcohol. They plan to get the cells “drunk” on alcohol and see if they play the game more poorly, just as when people drink.
Another team of scientists, meanwhile, has reported in Nature that they transplanted human neurons into rats’ brains and observed their behaviour, the long-term aim being to better understand how drugs affect the brain.
Stanford Medicine researchers first connected living human neurons and supporting brain cells with the brain tissue of rats that were two to three days old. With time, the human cells grew and formed hybridised circuits, forming working connections with indigenous circuits.
The team studied the response to a rare genetic condition called Timothy syndrome. They transplanted an organoid generated from a patient’s cells into one side of the rat’s brain, and an organoid from a healthy individual on the other side. Five to six months later, they observed marked differences between the two sides’ electrical activity.
“We can… use this new platform to test new drugs and gene therapies for neuropsychiatric disorders,” the study’s senior author, Dr Sergiu Pasca, said in a statement released by Stanford Medicine.