The Naked Scientists

The Naked Scientists Forum

Author Topic: Quantifying information and experts' knowledge  (Read 2826 times)

Offline vrba

  • First timers
  • *
  • Posts: 4
    • View Profile
Quantifying information and experts' knowledge
« on: 19/01/2009 10:19:38 »
Hello all,

This is my first post here.

I would need several hard-to-find information for my project and would be grateful to anyone that could point me to the direction of papers that could contain such information or provide those info directly.

First, speaking about long-established science fields like mythematics, chemistry, sociology, etc. (or their subdisciplines), what is the approximate quantity of knowledge information currently reached in those fields (in bits or bytes)?

Second, how big is the quantity of information that an expert in a specific field has stored in his/her brain?

I'm fully aware that the precise quantification is impossible. What I look for are approximations, like that of Georges Anderla, who took all the human knowledge in the year 1 AD as a unit and then made a study of the estimated decrease in the time necessary for it doubling, but I need similar data specified across particular scientific disciplines.

Now, I must presume that the quantity of facts stored in an individual's memory doesn't change much with time and progress (Thomas Landauer made a nice study about the estimate of "bits in the brain" used for storing learned facts and came up with 10**9 bits) so, as the scope of the knowledge broadens, an expert will hold in his memory the ever diminishing percentage of the whole quantity of knowledge in his specialized field.

I would like to know some estimated figures (not necessarily the newest ones, in fact, the historic ones would be very useful for comparison), I'm sure they must exist somewhere.

Thank you very much in advance,
Hrvoje


 

Offline Chemistry4me

  • Neilep Level Member
  • ******
  • Posts: 7709
    • View Profile
Quantifying information and experts' knowledge
« Reply #1 on: 19/01/2009 11:01:16 »
From one of the experts:

It's almost impossible to put a figure on it in computer memory terms. In a linear (conventional) computer, each byte can hold only 1 of 2 values (0 or 1) & each byte is discrete insofar as its content and action is not dependent on the value of any other byte. Memory in the brain doesn't work like that.

There are approximately 100 billion neurons (brain cells) in the adult human brain and they are intricately linked. Some are associated with memory, others with autonomic functions such as breathing, yet others serve different purposes again. Those that are associated with memory are intricately linked to many others (up to 10,000) and each 1 can participate in many memories.

Even if as few as 1% of neurons are associated with memory, that means an effective 1 billion x 10,000 connections (10,000,000,000,000). If each connection of each neuron = 1 part of a memory, that equates to 10 thousand terabytes. However, it's not quite as simple as that.

It is thought that neurons can participate in different memories as a result of the levels of neurotransmitters they produce at any given time. So, not only can each neuron connect to 10,000 others, but they can affect those others differently depending on the level of neurotransmitter they produce. It is almost beyond imagination how complicated such a network is, and how many possible memories it could store.

I'm not a neuroscientist, so I have given a very simplistic explanation which others here may take exception to. However, I think it gives a rough idea of the difficulty of answering your question.

If you want the whole story: http://www.thenakedscientists.com/forum/index.php?topic=13660.0 

 

Offline Chemistry4me

  • Neilep Level Member
  • ******
  • Posts: 7709
    • View Profile
 

Offline vrba

  • First timers
  • *
  • Posts: 4
    • View Profile
Quantifying information and experts' knowledge
« Reply #3 on: 19/01/2009 12:26:58 »
Thank you very much but, in fact, I don't need the human brain capacity info.

I need something along the lines of the following example (the numbers are fully arbitrary for the sake of this example):

"10 most significant world historic institutions have the libraries with 10**5 different volumes (different translations are not important and two works in different languages are considered one and the same work) containing in total 10**15 bits of information.

Those knowledge represents the vast majority of the facts that historiography in the world (bar the special projects that could be deemed sub-disciplines) uses and most historians would rarely have to venture far outside those information.

Now, a good historian had, during his lifetime, memorized 10**8 out of those 10**15."

So I would need a study that would put the "real" numbers instead of those I arbitrary inserted.

Thanks.
Hrvoje
« Last Edit: 19/01/2009 12:51:39 by vrba »
 

Offline rosy

  • Neilep Level Member
  • ******
  • Posts: 1018
  • Chemistry
    • View Profile
Quantifying information and experts' knowledge
« Reply #4 on: 19/01/2009 15:43:32 »
If there is such a study, it will be nonsense. Sorry.

The amount of information an "expert" needs to hold in his/her brain will vary dramatically with subject, a mathematician, physicist, or chemist might need to know fewer "facts" than a historian, because in all those fields a large proportion of the knowledge is tacit, based on experience (and inspiration, and sheer bloody hard work) in fitting together known facts not on the facts themselves... and to be "expert" in a given niche of one of those fields one might only need to know about quite a small proportion of what is known about the behaviour of physical matter.
 

Offline vrba

  • First timers
  • *
  • Posts: 4
    • View Profile
Quantifying information and experts' knowledge
« Reply #5 on: 19/01/2009 17:12:38 »
No, it would not be a nonsense. The point of such a study would not be exact quantification or comparing people by the bytes they know. The idea of collecting such a data is to show the need for hyperspecialization and why is ever harder to be a qualified polymath/polyhistor these days. There are sciences that doesn't require too much insight into physical matter but is...well, admit it, bordering on non-being a science (like history). The quantification could be possible with enough accuracy there - there are limited amount of bytes in general history that an average historian knows, etc.

Do you see my point?

Regards,
Hrvoje

 

Offline rosy

  • Neilep Level Member
  • ******
  • Posts: 1018
  • Chemistry
    • View Profile
Quantifying information and experts' knowledge
« Reply #6 on: 19/01/2009 18:09:33 »
I see your point. But I think you are fundamentally wrong.

This may be because I think the idea of an individual being a polymath in any sense beyond "having a greater familiarity with a greater number of subjects than most people" has been laughable for centuries, and the phrase "qualified polymath" doesn't seem to me to have any meaning, unless you mean in the sense of a "qualifying statement".

Quote
There are sciences that doesn't require too much insight into physical matter but is...well, admit it, bordering on non-being a science (like history).
I would agree with that statement, but all the same, it's a matter of degree. Would you consider that a zoologist requires the same degree of physical intuition as a physicist? Because I wouldn't, and yet I still consider zoology to be a science. (Actually, if it comes to that, I regard some forms of the study of history to be pretty close to science.)

« Last Edit: 19/01/2009 18:16:32 by rosy »
 

Offline vrba

  • First timers
  • *
  • Posts: 4
    • View Profile
Quantifying information and experts' knowledge
« Reply #7 on: 19/01/2009 19:29:45 »
Thank you rosy, I appreciate your reply.

Maybe the best thing to do would be for me to explain why do I need such information: I would like to write a report that would explain the need for hyperspecialization, teamwork and interdisciplinary work and the extinction of true polymaths (I use these terms in order to distinguish the polymaths from trivia specialists that would score big in a quiz game but have no depth of knowledge) in today's era due to the enormous increase of acquired and accumulated knowledge and would like to back it up with some quantified estimations.

For example, in the era of Thales or Pythagoras or their unremembered ancestors, it was possible for them to store in their brains a very significant portion of the cumulative knowledge of all scientific fields and probably the large majority of knowledge in certain specialized fields. Today it is not possible, unless we narrow the fields of knowledge into very limited sub-disciplines. Now, all that is logical but I would like to support it with some numbers....

Thanks,
Hrvoje
« Last Edit: 19/01/2009 19:36:28 by vrba »
 

The Naked Scientists Forum

Quantifying information and experts' knowledge
« Reply #7 on: 19/01/2009 19:29:45 »

 

SMF 2.0.10 | SMF © 2015, Simple Machines
SMFAds for Free Forums
 
Login
Login with username, password and session length