Accuracy and safety seem to be the dilemma of brain-computer interface - how to achieve a balance between minimum damage to the brain and maximum utilization of the brain is the core challenge of brain-computer interface technology
When machines penetrate into the human brain step by step , discussions on the relationship between humans and machines, such as am I still me, is a machine still a machine, and are human beings still human beings, will continue.
Article | "Lookout" News Weekly reporter Zhu Han
The technological explosion period of brain-computer interface is coming.
In mid-June, a study from Southeast University and other institutions showed that scientists used non-invasive brain-computer interfaces to control metamaterials with their thoughts and transmit their thoughts remotely. This result immediately attracted attention after it was published in "Light: Express".
Mind control is an enduring classic scene in film and television works. From the flying swords in Chinese martial arts dramas to the future world where human-machine integration is extremely common in the movie "Alita ", they all reflect human beings' expectations for expanding the boundaries of capabilities through machines.
brain-computer interface technology makes mind control no longer limited to imagination and science fiction. It is understood that as a cutting-edge disruptive technology that integrates life sciences and information technology, brain-computer interface has developed for nearly 50 years and has gone through the stage of scientific fantasy and proof of concept. It is gradually being used in the treatment of brain diseases such as epilepsy through clinical research. As well as allowing the visually impaired to "return" to light, allowing the disabled and paralyzed patients to regain their ability to move.
brain-computer interface can realize more scientific imagination, but it still needs technological breakthroughs. But what is certain is that the pace of connecting human brains and computers is getting faster and faster.
Facing the futuristic research of brain-computer interface and the possible era of human-computer symbiosis, people want to know: Where is the application boundary of brain-computer interface technology ? After "opening" our brains, will we be controlled by the outside world? Should human brains be "transported"?
Fantasy into reality
Brain-computer interface technology is a kind of communication and control channel established between the brain and external devices , using the brain's bioelectrical signals to directly control external devices, or using external stimulation to regulate brain activities, thereby enhancing , technology to improve and extend brain function.
The exploration of this technology began in the 1970s. In the past 50 years, many universities, research institutions, and enterprises around the world have deployed brain-computer interfaces, rapidly advancing from scientific principles, technology optimization, clinical transformation, and commercial applications.
The reason why many researchers are diligently pursuing the brain-computer interface is that it is considered to be the most effective tool for neural repair and an effective means to solve the neurological damage of patients with paralysis, stroke, Parkinson's disease, etc. In addition, it It is also the core key technology for comprehensively analyzing and understanding the brain, and is an important tool for the most cutting-edge research in international brain science.
As we all know, the brain is the most complex organ in the human body, and it is also the "ultimate territory" for understanding nature and human beings themselves. In the brain, more than 100 billion neurons form more than 100 trillion neuron connections, making "reading" the brain a worldwide problem. One of the purposes of the
brain-computer interface is to read, or even "understand" brain information.
It is understood that the technical process of brain-computer interface is mainly divided into two steps: collecting and interpreting brain waves . Interpreting brain waves is a key link in building a brain-computer interface system. Experts interviewed by
said that the process of interpreting brain waves is very cumbersome and requires a mapping model between brain signals and thinking tasks to be established in advance. Brain-computer interface systems use mapping models to process online recorded EEG signals in real time and convert them into machine instructions to control external devices. At the same time, the brain receives feedback results from the brain-computer interface in real time.
In terms of signal collection, depending on whether the EEG signal collection device needs to be implanted into the human brain, brain-computer interfaces can be divided into two categories: invasive (implanted into the human brain) and non-invasive (covered on the scalp).
Generally speaking, the signal quality obtained by invasive brain-computer interface is higher and the signal interpretation is more accurate. Since the 21st century, with the advancement of neuroscience , computing science, and materials science, invasive brain-computer interface technology has become a hot and difficult topic in global research.
In recent years, multiple research teams such as the University of Pittsburgh and California Institute of Technology have successfully used invasive brain-computer interfaces to allow paralyzed patients to use their brains to directly control robotic hands to complete grasping and stirring movements, and directly convert brain waves into text and speech. wait.
Domestic brain-computer interface research has also achieved certain results in the fields of medical and health. In 2020, the Zhejiang University team completed my country's first clinical translation study of invasive brain-computer interface. A 72-year-old high-level paraplegic patient controlled a robotic arm through the brain-computer interface and regained the ability to eat, shake hands and other movements.
"Behind every tiny movement is a series of complex processes such as signal sending, transmission and decoding." Zhang Jianmin, director of Neurosurgery at the Second Affiliated Hospital of Zhejiang University School of Medicine, said.
Accuracy and safety: The dilemma of brain-computer interface
Brain-computer interface carries the expectations of patients for functional rehabilitation and improvement of quality of life, but the technology itself still has many things to improve.
Zhang Jianmin introduced that the ultimate goal of any basic medicine is to solve practical problems for patients. Brain-computer interface research is of great significance to the clinical treatment and rehabilitation of patients with hemiplegia and aphasia . However, in order for related technologies to be truly applied to real life, some technical bottlenecks must be overcome. For example, how to solve the signal attenuation after implanting electrodes, how to minimize the damage to the brain caused by chip implantation, etc.
It is understood that after a period of time after the electrode is implanted in the brain, the electrode will be wrapped by glial cells , and the monitored neuron activity will become less and less or even disappear. U.S. public information shows that the most commonly used electrode for brain-computer interfaces is the Utah electrode, which has a lifespan of about 2 to 5 years after being implanted in the brain. Thereafter, if the brain-computer interface function is to be maintained, the electrodes must be re-implanted, which also means that the patient will experience the risk of re-operation. The damage caused by
chip implantation to the brain cannot be underestimated. Invasive brain-computer interfaces can deeply collect neural signals and obtain more high-quality EEG signals, but they are traumatic to the human body and have a high risk of rejection. They may bring risks of surgical injury and long-term implantation, safety risks of physiological rejection of equipment, and signal transmission. Or the risk of errors caused by software algorithms, etc. If
adopts a relatively mild and safe non-invasive technical route, it may be difficult to achieve continuous, stable and accurate capture of EEG signals.
"If you imagine the human brain as a chorus performing in an opera house, and each member is a nerve cell, then using non-invasive brain-computer interface technology is equivalent to letting the audience listen to the chorus outside the opera house. The audience I can’t hear a certain singer’s voice clearly,” said Xu Kedi, a professor at the Qiushi Institute for Advanced Study at Zhejiang University.
Accuracy and safety seem to be a dilemma for brain-computer interfaces - how to achieve a balance between minimizing damage to the brain and maximizing utilization of the brain is the core challenge of brain-computer interface technology.
A high-level paraplegic patient controls a robotic arm through a brain-computer interface and regains the ability to eat, shake hands and other movements (photographed in January 2020) Photo by Lu Shaoqing
Who will have the brain signal
People are hesitant about brain-computer interface technology, and it is not just Personal safety. Brain-computer interfaces also face legal and ethical dilemmas such as privacy leaks and mind manipulation. Questions include: If brain-computer interface devices are used on a large scale, who will own our brain signals? Who has the right to read, write, or even tamper with brain signals?
In 2020, at the "Web Summit", Europe's largest technology conference, held in Lisbon, Portugal, Rafael Yuste, founder of the neurorights advocacy organization "Morningside Group" and professor at Columbia University in the United States ) said that advances in brain science and technology, from deep brain stimulation to wearable instruments, make it increasingly possible for external forces to manipulate the human brain.
Researchers have demonstrated that brain imaging and neural signals can be used to determine thinking and intervene in behavior. American scientists published a paper in 2013 stating that people's political views can be inferred from brain activity, such as distinguishing between Democrats and Republicans.
In terms of animal experiments, researchers from Zhejiang University found that stimulating specific areas of the brain can change the behavior of animals. For example, making animals become more aggressive, or changing the animal's route, etc.
In the commercial field, some wearable devices that detect brain activity can also be used to track emotions, emotions or judge mental states. For example, the "XWave" headphones developed by an American company have been allowed to plug into Apple phones and read brain waves. The product claims to be able to detect weak brainwave signals through the surface of the skull and forehead, and convert these analog signals into digital signals to detect concentration and meditation levels.
Scholars are worried that EEG signals contain private information such as health status, life experience, property, marriage and love, social relationships, beliefs and psychology. Brain-computer interface technology with the ability to "read minds" can easily lead to the leakage of data, privacy, etc.
360 Group Chairman and CEO Zhou Hongyi made it clear in an interview with the media that the brain-computer interface may have opened "Pandora's Box" and advocated that "human beings should strictly limit the scope of use of the brain-computer interface."
In addition, brain-computer interface technology It may also pose ethical challenges such as mind manipulation.
Professor of the Department of Philosophy of Peking University Chen Shaofeng believes that the causal impact of direct physiological manipulation of people's bodies, intentions, memories and emotions through brain-computer interfaces is unprecedented, which makes it possible to identify autonomous decision-making or decision-making from an objective perspective. Machine decision-making becomes difficult. Commercial analysis of brain signals may allow companies to reversely manipulate users' ideological tendencies and purchase choices by decoding users' neural signals.
Chen Shaofeng further said: "Should we still transfer part of the decision-making power to the brain-computer interface algorithm? How to determine whether the behavioral subject is controlled by himself or the device? When the user's brain signal conflicts with the results of the machine algorithm, who is the final arbiter? It needs to be redefined at both the moral and legal levels. "The ethical impact of
brain-computer interface technology also includes discussions on the boundaries of technological enhancement and human empowerment.
Some experts believe that once brain-computer interface technology matures, human beings will have a new, electronic "super power". Ordinary people without "plug-ins" may be at a competitive disadvantage, exacerbating social class differentiation. Can this digital divide be accepted by people?
How to avoid the "Collingridge Dilemma"
Various legal and ethical doubts about brain-computer interfaces in academia and industry reflect the lag in legislation and regulations related to brain-computer interfaces.
This is a worldwide problem. According to media reports, Spain, the United States, France, , Argentina, , etc. have all begun to study this issue, and some countries have advanced targeted relevant legislation. For example, in 2021, Chilean lawmakers passed a law establishing the rights to personal identity, free will, and mental privacy, becoming the first country in the world to legislate neurotechnology that can manipulate the brain.
In February 2022, the "Global Cyber Security Policy Legal Development Annual Report (2021)" jointly released by the Cyber Security Law Research Center of the Third Research Institute of the Ministry of Public Security and the 360 Group Legal Center predicted that one of the seven major trends in global cyber security policy legislation , that is, " quantum computing , brain-computer interface and other fields will enter the field of legal regulation."
According to Yi Haoxiang, CEO of the brain computer company Enter Technology, in terms of privacy protection, Enter Technology has proposed some industry standards. For example, technology usage rights and data ownership belong to users under any conditions; service providers must obtain written permission from users before providing services, and cannot store or download any data involving users’ personal information. However, these are still voluntary actions of enterprises.
In Chen Shaofeng's view, "Only by implementing ethical solutions and setting up technical red lines before putting them into widespread application can we avoid entering the 'Collingridge Dilemma' where a 'lost sheep' cannot be 'fixed'."
"Colin "Gridge's Dilemma" is a view proposed by British technology philosopher David Collingidge in "The Social Control of Technology" in 1980. It means that the social consequences of a technology cannot be anticipated early in the technology's life.However, by the time undesirable consequences are discovered, the technology has often become so part of the entire economic and social fabric that controlling it is difficult.
In this sense, Chen Shaofeng calls for the establishment and improvement of technical security principles, brain privacy protection principles, independent decision-making principles, distributive justice principles and policy guarantee principles, which will help to objectively and rationally view the risks and risks that brain-computer interface technology brings to human beings. Profit and promote its comprehensive and healthy development in an ethical manner.
Should the brain be "transmitted"
Beyond legal ethics, as machines penetrate deeper into the human brain step by step, the development of brain-computer interfaces also spans the proposition of the relationship between humans and machines.
"The brain-computer interface needs to collect and read the brain's electrical signals through electrodes. These signals form an original data set. On the basis of labeling the meaning of relevant signals in some experiments, we can use machine learning algorithms to 'read' relevant Memory content." said Pan Enrong, a professor at the School of Marxism at Zhejiang University.
Some scholars believe that in the future, brain-computer interfaces may become smart terminals that surpass smartphones and lead mankind towards human-machine integration. Experts interviewed by
said that brain-computer interface and brain-computer intelligence will profoundly change the current situation in which artificial intelligence focuses too much on imitating the brain. It will produce intelligence that combines the advantages of brain- machine and change existing thinking, communication and production. The method will replace part of human production, life, learning and even thinking in the future, and may have a profound impact on technology, industry, economy and society.
In other words, when computers become part of the human mind, human self-concept may be challenged.
can foresee that with the development of brain-computer interface technology, discussions on human-machine relationships such as am I still me, is a machine still a machine, are humans still humans, etc., will continue.
For example, Deep Brain Stimulation (DBS) technology, which inserts electrodes into the patient's brain and suppresses abnormal local neural activity through current stimulation, is often used to treat obsessive-compulsive disorder, epilepsy, depression, etc. treat. But then a point of view was raised: By changing neural activity to make a severely depressed person happy again, is this happiness still real happiness? ■
