“Mini Brains” Playing Video Games

From mouse embryos to mini brains, is this the next development next to AI? Scientists create brain organoids, subject to testing and experimentation outside of the human body.


NIH Image Gallery from Bethesda, Maryland, USA/ Wikimedia Commons

Here are human brain organoids, the new ethical test subjects. Their size is close to a small poppy seed, and their structure, organization and electrical signaling mirror brain tissue. This image highlights defined structured regions in formation through the process of organoid development.

Between your ears, holding endless thoughts and creativity and a whole world inside of you is the world’s most efficient supercomputer, the human mind, which has an endless history.

In dark storage rooms and museum displays, papyrus scrolls from the Ancient Egyptian dating from before the 17throughout century B.C. are filled with the beginnings of neuroscience. Hieroglyphics scattered records describe the symptoms, diagnosis, and prognosis of two patients suffering from severe wounds to the head, and the assessments of their injuries from a battlefield surgeon.

From these relics, we know that prehistoric persons kept the remains of perforated skulls in the revelation of the trepanning process. However, our practiced surgeons today have alternative and more successful remedies. With many innovations, we now have a more complex understanding of the mind, and no longer have to go through the hassle of drilling a hole in our skulls just to cure a migraine. Today, our studies of neuroscience have reached as far as creating a “mini-brain,” or a non-conscious cluster of brain cells, that has paved the way for a new window into the mind, all puns intended.  

Many cultures of brain cells, including mini-brains, gained popularity around 2013, aiding scientists in studying microcephaly, a genetic disorder where babies are born with unusually small brains. These substitutes for the brain allowed scientists to observe interactions with an external environment as well as test different treatments for neurodegenerative diseases, like Alzheimer’s, in isolation. Before this, the most common methods in neuroscience research remained limited to EEGs, fMRIs, and ERPs.

In essence, to witness and record changes, scientists started with detecting abnormalities in the brain waves by measuring electrical activity using electrodes placed outside the skull and continued updating through the treatment. Functional magnetic resonance imaging (fMRI) uses radiology to measure small changes in blood flow and brain activities. Similarly, ERPs and EGGs both measure electrical signals from the surface in response to certain stimuli. But the commonality between these methods is that they all occur outside of the brain, as non-invasive procedures.

In the 1940s, Canadian brain surgeon Wilder Penfield began experimenting with the exposed parts of the brain and mapped the brain’s motor cortex by applying mild electric currents to patients during surgery with electric probes. By internally inspecting and exposing the brain, Penfield witnessed the effects of electrical pulses on muscle movements in isolation and simultaneously probed into the beginning of a new phase of neuroscience. 

One of the key issues with non-invasive types of research amounts to the idea that the brain belongs to a person – someone who is conscious. Along the lines of evolutions, these mini-brains can hypothetically gain consciousness, and fuel the backstory of every legendary superhero, albeit without the guarantee of any heroic figure. To our reassurance, the fear surrounding this concept is less of mad creation and more immersive into the ethicalities and respect for science.

If the cells gain consciousness, it would mean that they can feel pain and understand their environments and surroundings. Because of this, several experiments fail to preserve ethical means and thus could not be performed. Still, scientists persisted, and while the approved guidelines for what experiments can be performed remain a heated debate, all members of the neuroscience community work towards curing the pains of human nature rather than heightening them.

Consciousness itself remains a vague subject, so researchers turn to sentience to provide a baseline for their propositions. What remains in the concept of sentience is that it can’t necessarily be numerically defined and rather works in a range. As British neuroscientist Karl Friston states, “One could certainly measure the neuronal correlates of consciousness, in the usual way. However, one can measure sentience — of a basic sort. If one defines sentient behavior and basic sentience as the internal representation of the consequences of action, then one can measure this kind of sentience in terms of anticipatory behavior.”

In short, any experiments performed measure the capacity to experience feelings or sensations, and follow less sentient subjects. Now, instead of operating on a patient, scientists utilize cultures of isolated brain cells from donors to conduct research. Such cells came in two forms, dissociated networks, and brain organoids. It’s important to realize that while these mini-brains are indeed impressive, they aren’t truly ‘brains ‘ but instead thin layers of tissue grown in sheets or stacked together – reaching only about 1/10th of a millimeter in diameter at largest. Their growth past this point remains inhibited due to the dynamics of the structure. Since the structures lack any circulatory or excretory system, any mini-brain that expands past this point begins dying on the inside, rendering them necrotic. 

More recently, scientists based in Melbourne, Australia conducted another experiment published in the Neuron, in which they describe the process they used to grow human and mouse brain cells, and the following experiments conducted on each. Using human stem cells, they developed brain cells and mouse brain cells derived from mouse embryos, collectively amounting to a total of 800 thousand cells. These cells were placed onto a “DishBrain,’ or a nutrient-filled dish with an electric chip containing electrodes for stimulation. Through a series of pulses, scientists monitored the cells for active responses and were able to communicate through a pattern of electrical signals. These signals were sent to and from a computer that recorded them. Like morse code, the computer program followed patterns to communicate with the cells and transfer, and record information. The team of scientists aimed to test the reaction of the “mini-brains” to different stimuli and compel cells to modify their activity.

To do this, they used the famous arcade game from the 1970s called Pong. The goal of the game is for two players to use paddles that throw a ball back and forth. The key to this experiment was simplicity, as their experiments used isolated factors in the environment to differentiate results, and utilized previous years’ references and research to create a unique form of testing and apply previously identified skills. 

Without stimulation, the cells showed no intention of playing the game, considering their lack of understanding. To remedy this, the scientists motivated them with electrical stimulation: a signal when they got it right, or a chaotic stream of white noise when they got it wrong. Through these signals, they created a reward system aimed at encouraging the cells to hit the ball.

In response, the cells produced electrical activity of their own and expended less energy as the game continued- demonstrating a convincing argument for “learning” how to play. As concluded, the cells tentatively reached for predictable situations in their environment, seeking out the organized electric outburst. While the cells never achieved successful Pong rates, their success rate was noticeably higher than pure random generation. Human brain cells were notably better at playing the game than their mouse brain cells counterparts, and such information replicated features comparable to the learning and memorization process we are most familiar with. As the brain learns how to do certain activities there’s a point where they become second nature; for example in the way that muscle memory works. 

The astonishing part about these cells is that they each observed different learning styles. As mentioned before, the cells of each brain organoid also featured surprising qualities, as they began to morph and resemble different types of differentiated neurons. However, rest assured, our fears of AI running rampant and tyrannically dominating the world won’t come to fruition anytime soon. Still, the experiments proved much more than just playing pong.

As Dr. Adeel Razi, associate professor in research and psychology put it: “While still more evidence is required and several methodological challenges still to be solved, we showed that the human cortical cells showed faster learning than the mouse cortical cells in a dish. This may point towards the (innate) superior capacity of human cells to learn compared to other species,” pointing towards proving some new theories and expanding on a more solid basis of research.

Of all instigators, electricity isn’t the only way that cells respond, but several other factors as well. Cells respond to what they sense, mainly through receptors and ion channels and any changes to the environment via electrochemistry. Many other inputs can be used to invoke a response, since “cell cultures in principle can process and convert various sensory stimuli (visual, auditory, etc) into appropriate electrical signals.

“It is loosely like how the computer works, as we provide input by pressing keys or waving our hands, or touching the screen. These actions/inputs are then converted into electrical signaling for further processing by the computing device to produce the desired output (printing a page, scrolling a web browser, etc),” Dr. Razi continues. Anything from ultrasound to chemicals, to temperature, or even lights in the science of optogenetics are alternatives to trigger reactions in cells.

Following these concepts, Dr. Kagan’s team simulated the effects of alcohol on these brain cells, to simulate the human mind when drunk. After submerging the cells in alcohol, they put them to the test at the Pong game once again and noted that the results of the cells declared clear equivalents to the drunken state of the human brain since the cells were less successful in hitting the pong ball. These replications create a miniature world in which we can more closely analyze the effects of differences in isolation towards the brain. Like the duplication of alcohol’s effects, we can now track the treatment of different medications, aiding in our perception of epilepsy and dementia. 

Without delving too far into philosophical rhetoric, the human mind is everything. At least, to us — holding our perceptions, our thoughts, and incredible neural networks that are arguably superior to other living beings. The integration of our psyche and computational devices of today’s day and age would mean a breakthrough in several dimensions, bolstering a new age of ameliorated technology.

But before we can imagine computer chips with the speed and efficiency of our brains, several more issues need to be addressed, such as the process of cultivating and maintaining brain cultures. Such a procedure becomes a meticulous task with temperature controls, pH levels, infection, contamination prevention, and further circumstances. As one can imagine, the layered protections of our skulls and bodily systems provide much better protection which plastic nutrient plates in laboratories simply cannot replace. In the history of this particular line of research, several breakthroughs have been made and lost as well.

Most researchers may believe that cell plates can only last up to 1-2 months, but the truth of the matter is that cell cultures that are properly maintained can be grown for well over a year. The main issue in this procedure is the sensitivity to osmolarity. Whereas skin cells are constantly exposed to air and different chemicals or factors, brain cells are far more sensitive, and can be overloaded by hypertonic solutions which prematurely kill the cultures. If the cells are exposed to open air, like that which we breathe, they shrivel, as if they are salt on slugs. 

Due to a lack of communication and collaborative efforts, the solution remains ignored, and widely unused – Teflon plates. These Teflon plates protect cell cultures through humidification and prevention of air exposure, creating a sterile environment. It was created by Dr. Steve M. Potter and Jerry Pine, working at the California Institute of technology. Potters’ research group began in 1999, and through years of effort, the “cultivated” cultured networks finally succeeded in 2008. Through several other researchers, contributing from around the world, and throughout several projects, the information grew and gained in popularity, especially prompted by the leading hit publicity of Dr. Kagan’s 2013 paper. While the details are more complicated than reported media coverage, the stir brought in attention from VC venture capitalists and investors who donated more money to the company and further research. 

Like this, many other situations remain unresolved. For one, the number of cells you start with in a brain culture is at most what you will return with at the end of the experimentation. Unlike muscle cells, brain cells divide once after completing differentiation, and thereafter need a constant supply of new cells either from tissue or stem cells. In recent years, a fairly new process of differentiation can be used, while it is incredibly difficult to manage successfully. Another concern is the technological side, as the real-time interface of neural networks requires a very high speed, and technology can tediously take up to 3 months to mature and fire electrical signals consecutively, at a point where researchers can then proceed to experiment on. Additionally, there are still further prompts for the application of research.

Most research is based mainly on computers instead of the crossover of biology and machinery. We often forget the most important step, to reground and return to the basics of psychology and neural networking. Without retouching to base, most research aimlessly fires farfetched possibilities. As Dr. Steve Potter puts it, “It’s kind of like saying, I’m going to look at a bird and I want to make a flying machine – and I think that feathers are the most important thing. So whatever my machine is, it’s gonna have feathers on it, you know?” Thus the manners in which the brain and skull work, axon buildup, and basic information are all critically important to the understanding of the mind, for which the proportionate information of computer technology is more readily available already.  

Despite the plethora of research investments required, the utilization of mini-brains offers a promising world of innovation ahead. Our brains themselves are, as mentioned before, highly efficient supercomputers. The mind uses only about 20 watts of energy, as compared to the most powerful man-made computer that can take a warehouse full of backup and 22 megawatts of energy. You may find computer lagging to be quite a nuisance, but they too are important tools. “Evolution said, delays are useful, we can make use of delays since they model the outside world,” said Dr. Potter. Features like this are unique specifically to the biological system. The brain holds many mysteries, like the purpose of dreams, or our subconscious mind. Inevitability, there’s unlocked potential in need of discovery. 

Now, scientists are testing the waters to a new world of functionalities and a new medium of research and bringing into the future where computers may induce biological phenomenon, in place of artificial intelligence. Each new experiment can hold the cure to diseases, develop the cells to more complex functions and mimic the human mind to finally venture deeper into neuroscience.

Now, scientists are testing the waters to a new world of functionalities and a new medium of research and bringing into the future where computers may induce biological phenomenon, in place of artificial intelligence.