Brain Science Podcast discussion

My fascination of the brain science started 6 months ago when I read Time magazine on Singularity (my degree is in EE with a background in Computer Science). I read books written by computer scientists and neuro/bio scientist and listen to the many podcasts offered by Ginger. There is not one definition of intelligence. Wikepedia is the closest that I can associate (understanding). Seek opinion. Some books of interest that I recommend: "On Intelligence by Jeff Hawkins", "Making Up the Mind by Chris Frith", "Brain the Complete Mind", "The Singularity is Near by Ray Kurzweil" and "Emotional Machine by Marvin Minsky". All these books have different view of intelligence/Brain.

Henry wrote: "My fascination of the brain science started 6 months ago when I read Time magazine on Singularity (my degree is in EE with a background in Computer Science). I read books written by computer scient..."

Thank you for contributing to this group. Intelligence certainly is an example of something that is not easy to define even though everyone thinks they know what it is! I tend to agree with those who argue that human beings have many different types of intelligence.

Also, the more we learn about animals the more we come to appreciate that most are "smarter" than we ever imagined.

Ginger thank you for your reply. I do agree with your comment.

I think of intelligence as the ability to effectively deal with our environment and survive. There are lots of ideas out there on intelligence, like Howard Gardner's ideas on Multiple Intelligence. The theories of intelligence I have seen tend to deal chiefly with the rational and symbolic aspects, and with declarative memory. The emotional brain and animal intelligence operates without language (with exceptions), and so is less accessible to introspection. Jaak Panksepp's work, featured on the Brain Science Podcast (Episode 65) and Evolutionary Psychology are shedding light on what was formerly lumped into "the unconscious." The silent intelligence of the emotional brain seems to dominate the actions of humans and animals.

I would love to get feedback on a model that deals with both executives (logical/symbolic and emotional/intuitive) and sense-centered intelligence. Another key aspect stems from competing demands for limited mental resources: http://web.me.com/rcmsite/RCMsite

Intelligence: I think the first thing to do when trying to describe intelligence is to separate it from the expression of intelligence. We tend to mix together the perception, evaluation, and comprehension of experiences, be they common or novel, with the response that’s chose and execute. In my own connectionist view, the foundation of intelligence begins with the senses and is largely dependent upon the number, resolution, and bandwidth of the senses - perception is very dependent upon the data sources. Beyond sensory input comes cognitive prowess which is essentially made up the of excess gray matter left over after satisfying hierarchical processing demands that allows us to establish memories and networks of associations between them to give them depth and meaning so that one can draw upon past experiences to select successful responses. At this point, the number and variety of experiences stored and associated in that excess gray matter are every bit as important to intelligence as the amount of excess. Next, the output is dependent upon the ability to access templated response patterns established under the influence of previous experiences and employ dynamic control systems for a well coordinated response. Finally, sensory feed back together with an emotional response provides a mechanism for refining this response based on its context with either positive or negative reinforcement. Intelligence is dependent upon the process from beginning to end. The problem of defining intelligence is the same as the problem defining consciousness, life, circumstances, universe, country, family, and God. Some words are simple, embodying only one concept. Others, like those examples, embody a multiplicity of concepts both directly and implied by extension.

Now, if you want to have some real fun, ponder what the definition of a super-intelligence might be.

This page has the most extensive collection of intelligence definitions I have seen. Might be of interest.

I spend all day writing computer programs.
Modern program development environments enhance your intelligence in a number of ways:
1) Fast search is provided so you can quickly find variables and functions, when as usual you forget their exact positions in the 200 or so files that make up your program.
2) Programs are organised into a taxonomy with classes at the top, and functions below those, and the DE provides you with a tree display to locate things within the taxonomy. They will display alphabetically the function names in any selected class, supporting more efficient Recognition compared to Recall.
3) Whenever you click on a function or a variable, in 2 more mouse clicks you can go directly to its definition, or to a list of the references to it.
4) The windows onto programs that allow us to press buttons and read lists are built from thousands of lines of code, but the DE will build this code semi-automatically for you, so it is not unusual to produce 2000 lines of code a day, which you then tailor to your own requirements, easily done once you learnt the structure of the code used to build a window.

Research at Microsoft has looked at putting cameras on people's heads, and microphones on their lapels, to record day-to-day memories allowing them to be searched as in the DE.

Working in a DE has funny effects on the mind. A week after writing code you forget completely why you chose the particular design, and it tends to take half a day to re-familiarise yourself.
You wake up in the night with code swimming before your eyes, that you try to analyse but can't quite make it (without all the support of the DE).
You tend to remember short phrases that typify the code in each function.
For example, "I can't remember the name of the function, but I know it contains de.Value". The DE will search and list all references, with the file names (it ought to be the class names), and that is usually enough to trigger Recognition within a minute or so.

I am constantly up against my limitations in memory and in analysing code I wrote a year ago. And I find it very hard to generalise about the structure of my code spead over so many files.

I gather that long-term memory theories in Cognitive Psychology say that we create similar taxonomies, and frames to describe frequent events, like a meal out. The frame describes a typical meal, for example, and an individual meal has memory links to the frame, and any unusual features that distinguish the memory from the frame. Three-level taxonomies are the most common.
(See Murphy, "The Big Book of Concepts")

So I suppose having more intelligence would mean remembering deeper taxonomies, and having the ability to generalise a frame from a smaller number of repeated memories (but we tend to do that anyway when we become paranoid). That means doing the statistics more efficiently, and developing the ability to filter out noise.

We do seem to have the sort of search facilities that the DE has, but they are very slow. Often I can remember a person but not their name, but by repeatedly retrieving memories of seeing them, the name will eventually (sometimes over 2 days) pop into my consciousness.

I don't feel I need a better memory, just better analytic abilities, and I would be very happy to rely on a portable DE to enhance memory.

Query: I wonder about the impact of childhood myopia, prior to correction, on perceptual, cognitive,and learning functioning and style,the development of the brain and other senses. Think of how much different a myopic child's view of the world is misses and how confusing it must be to not be able to see what others can clearly see (clearly). Is there a greater susceptibility to ADHD? Has this been investigated and if not we might want to. If there are benefits to correcting vision earlier, we should know it.