We chat with Randal Koene, aforementioned neuroscientist about his involvement with Eternal Trust, the dream platform, but first we checked out what a neuroscientist is exactly:
RK: The Wikipedia entry on ‘neuroscientist’ is a pretty good detailed description.
In short, a neuroscientist is a scientist with a focus on the workings of the brain, studying human and other animal brains to understand what gives rise to cognition (thought, memories, etc).
Where this differs from a psychologist is largely in the approach. While a psychologist deals with human cognition through phenomenal and behavioural traits, the neuroscientist looks into the internal mechanisms, opening up the box, you might say. We study the components of brains, such as neurons, and their interactions through connections between them. We study the way these ‘circuits’ of the brain process information.
In my case, I’m a computational neuroscientist, which means that I am particularly interested in the formal representation of those neural mechanisms and brain structures. How can the information processing carried out by a specific part of the brain be expressed through mathematical data operations?
JG: What are substrate minds and how do you map them?
I think you probably meant to ask about ‘substrate-independent minds’. This terminology refers to the notion that mental processing can be expressed in mathematical terms, and therefore, any processing substrate that can implement those mathematical processes can carry out the same mental functions. If this is true, which is the de-facto standard model of neuroscience (even if it is not often explicitly verbalized), then it means that you could build an equivalent mind using any processing hardware that can implement the necessary mathematical operations. Concretely, you could run a mind in a computer made of silicon just as well as you could run it in one made of squishy carbon materials. To be clear, the ‘independent’ in ‘substrate-independent mind’ does not mean that there is no processing platform, but rather that any formally equivalent processing platform should be sufficient.
As to how you map them: That is still an open question! There are some initial assumptions that today’s neural models work with. For example, we assume that neural computation is carried out when neurons receive and respond to electrical signals. If this fundamental assumption continues to hold as it has for the past 70 years of study, then we should be able to map the processing circuitry of a mind by identifying the neurons in the system, tracing the connections between them, and establishing the effect that a pulse of activity from one neuron has on its connected neighbors. This is the essence of what is known today as the study of the ‘connectome’ and neural modeling.
JG: If you are future gazing – how will your current work play out in the future – say in 10 years, 50 years, 100 years , more? Could you give examples of possible advances that might be made in those periods
RK: I think in 10 years, my work can contribute to clinical neural prosthesis: the design of electronic replacement parts that can heal specific brain dysfunctions. In 50 years, there will probably be example cases of patients where a large proportion (or all) of the brain was gradually replaced with neural prosthetic devices, so that the patient has what could nearly be called a whole brain emulation. In 100 years, I think that the pace of neuroscience and technology development means that it’s likely we will have the ability to transfer or convert entire human minds to run on whole brain emulations, in effect enabling what is know in SciFi stories as ‘mind uploading’.
At this point, it will be possible to merge human and artificial intelligence, to augment and improve the brain in a multitude of directions, and to use ‘backups’ to ensure reliable survival and further development for those people who choose to avail themselves of the option of whole brain emulation and mind uploading.
JG: When did you come to hear about the Eternal Trusts project?
RK: I learned about Eternal Trusts a few months ago, possibly in February or March of 2018.
JG: How do you see medicine and ethics working into the future? Do you see any issues?
RK: I think that medicine and ethics always have issues, because they deal with such very personal matters. We will always care greatly what sort of treatments we receive, and why, how our best interests are safe-guarded, and what our rights as patients are. Similarly, we will care a lot about access to our medical data, ownership of our medical data, and fairness in universal access to medical options.
JG: What would you put in your trust if you were to write one?
RK: When I write mine, I will take care to specify that my aim is whole brain emulation and mind uploading. Activity recordings, as well as carefully preserved brain tissue should be stored, along with macroscopic validation data for my personality (e.g. audio/video recordings, Q&A, etc). All of this should be used in the most accurate and reliable processes for creating a whole brain emulation from my stored brain data. Then, I will wish to experience revival with an option to switch between a sophisticated artificial body (think: robotics as detailed or more detailed in their sensation and actuation than even a human body) and a body in virtual reality. There is probably a lot more, but to be honest, it will take a while to carefully think through what a precise set of wishes, as encoded in a Trust, should be.
Thank you for your time – that was a fascinating view into the mind of a computational neuroscientist.
Randal Koene is a Dutch neuroscientist and neuroengineer, co-founder of carboncopies.org, the outreach and roadmapping organization for advancing Substrate-Independent Minds. Former Professor at the Center for Memory and Former of Boston University, co-founder of the Neural Engineering Corporation of Massachusetts.
For more information on Eternal Trusts please visit here