Brain Science Podcast discussion
2013
>
Evolution of brain/mind software
date
newest »


Angst: Origins of Anxiety and Depression - Jeff

The "software" can be thought of as all the learning that happens after birth, while the "hardware" is indeed the body-embedded brain evolution has given us.
Evolution changes the "hardware" slowly and iteratively.
What changes the "software" is CULTURE. It's also an iterative process, and while much faster than evolution, still slow.
Finally, be careful not to get too caught up in the computer/software analogy. It's fun and useful, but has led theorists astray in the past. Varela & Thompson & Rausch's 1991 "The Embodied Mind" lay the groundwork for this increasingly accepted paradigm. It has a great synopsis of the evolving theories of mind, including the computational view.

CULTURE, yes, but also Nature: snakes/spiders are terror to some, yummy protein for others. Rats may've been okay until plagues reprogrammed certain afflicted tribes.
Getting out of the context of Darwinian evol ... Certainly the mind (software) can be dynamic while the brain (hardware) remains mostly static ... e.g., a person changes his mind about existence of God, etc.
Jeffrey: Not sure I would strongly agree that " 'mind hardware' varies considerably from one person to the next". Not w/o some more detail on what is meant by "considerably". Not sure what you mean by "mind hardware" (i.e., firmware??). IAC, and AFAIK, ... the human species is fairly young ..."Researchers believe that anatomically modern humans emerged between 150,000 and 200,000 years ago in eastern Africa. "
http://www.scientificamerican.com/art...
"Considerably" or "significantly" in my book may be like Pan troglodytes vs. Pan paniscus (common vs bonobo).

Modern conventional computers operate in a fundamentally different way than brains. There is a relatively clear distinction between hardware (CPU) and software which is loaded into memory. There is no clear distinction within the brain.
If we stretch the analogy and take the neural substrate of the brain (the neurons, axons, dendrites, etc) to be the hardware, there is no equivalent of RAM in the brain. Memory is stored as changes to the structure of the hardware itself.
Work is underway at IBM and some other organizations to develop neuromorphic chips which will operate in a massively parallel, massively interconnected manner more analogous to the brain. However, these computers will not be programmed as such, rather, they will be taught or trained.
Modern convention computers can programmed to simulate the way the brain works, but it requires massive resources to model even a small number of neurons. This is what the Blue Brain Project did. A new project was started this year (the Human Brain Project) which will embark on the mission to eventually simulate an entire human brain. Right now, there is simply no computer powerful enough to do that, but the hope is that over the next 10 years with advances in supercomputing, brain modelling and more refined mathematical models of neural processes this could be achievable.


Agree that there are MAJOR/fundamental diffs. between "Modern conventional computers" and human/mammal brains. (And it's good that you noted parallel processing, which organic brain does way better than artificial). HOWEVER, we STILL can't REALLY, TRULY compare the two because, RELATIVELY SPEAKING, we don't know that much about the human mind-brain system (e.g., the Hard Problem, integration of the "Second Brain" [ENS, digestive track*] etc.) whereas we excel at the former (no surprise).
* http://en.wikipedia.org/wiki/Enteric_...
First I have to admit that I haven't read everyone's posts carefully, but I want to make one point about thinking of brain evolution as being "software only." I do not think that that is an accurate representation of current thought in this area.
First, at present we have no way of knowing what kind of physical changes have occurred in the human brain in the last 10,000-100,000 years, but it is unlikely that the physical structure of our brains is "set in stone."
Next, on first reading it appears that the term software is being broadly applied. Using a computer simulation (with software) to emulate the apparent parallel processing of our brains is a far cry from understanding how our mind's "software" evolved. Certainly culture has evolved and developments like language and writing have had a huge impact on our brain/minds, but the key point I want to make is that these changes appear to impact our individual brains at a physical level. That is one of the points we tried to stress in BSP 97.
The more we appreciate how the world around us changes our brain/mind and even the expression of our genetic makeup, the more it seems to me that it might be time to give up the software analogy because it does not represent the dynamic nature of our minds.
First, at present we have no way of knowing what kind of physical changes have occurred in the human brain in the last 10,000-100,000 years, but it is unlikely that the physical structure of our brains is "set in stone."
Next, on first reading it appears that the term software is being broadly applied. Using a computer simulation (with software) to emulate the apparent parallel processing of our brains is a far cry from understanding how our mind's "software" evolved. Certainly culture has evolved and developments like language and writing have had a huge impact on our brain/minds, but the key point I want to make is that these changes appear to impact our individual brains at a physical level. That is one of the points we tried to stress in BSP 97.
The more we appreciate how the world around us changes our brain/mind and even the expression of our genetic makeup, the more it seems to me that it might be time to give up the software analogy because it does not represent the dynamic nature of our minds.

I think computers describe us. Software and the way we make software describes who we are historically. Software is not cut and dried. It is a set of rules or business logic. These are vague, ambiguous, result from meetings, and follow political persuasions.
The universe cannot be described in neat, tidy, indivisible packets of energy. Such packages are intellectual, only, expressing a desire to control. Neither can behavior be described in tidy little moral packages. That too is the desire to control, even the fantasy of control.
I have more faith in competition and scrutiny than I do in ethics.

Modern audio and video CODECs like flac or mp3 or h264 or divx may use lossy or lossless data compression. It took software engineers and programmers decades to work out the math and algorithms. In a way, it's like the game TETRIS.
This is only one dimension of the type of "software evolution" I was querying.
Also -- and as others have suggested -- language and culture probably have had a major impact on the evolution of intelligence. Fossil records of Homo s. s. -- which does not reveal much about soft tissue -- seems to suggest that cranium size has pretty much been static for ~200k yrs.
There's also the firmware issue -- like computer BIOS and drivers: Forgetting about JUST software, even given pretty much the same physical brain, wiring/interconnections (connectome, etc.) can also evolve pretty much in situ.

Dunno ... Nature may turn out to be TOO complex (given that it projects itself into quantum realm on one end and cosmology in the other).
Emergence aside, different ways to tackle a problem to come up with a common solution is something even Nature does ... e.g, flight in mammals, insects and birds
(sorry for hugely anthropomorphizing Nature!).
Maybe the Gods that created this universe did so because of the complexity of infinite regression.

Not sure what's meant here by "environment". In the large scheme of things, our solar system and galaxy are settling down after billions of years of chaotic formation "stages". And this decomplexion may very well lead to life being able to flourish.
Smaller "environments" -- ice ages, cultures, technological -- pose random and synergistic challenges. Some will lead to hormesis which may/may not lead to smarter minds.
An analogy may be computer OS, such as Apple, Windows, Linux. Programmers working for, say MS, may evolve code from Win 7 to Win 8 (mind/software) that operates on the SAME Intel CPU (brain/hardware); and that CPU may be far "older", e.g., Win XP-generation. So, mind/software may evolve independently (over time) or may be interdependent with hardware.
If the physical structure isn't evolving/adapting fast enough, then maybe Nature's "workaround" is more efficient/effective coding that uses the pathways. In busy cities, traffic lighting can be programmed to minimize jams -- all this w/o adding more roads/bridges/etc..
I know, I know!! ... a very broad/complex topic!