CONTACTS 
0201 - University of Colorado             303 492 2875 fax or 800...
020101 - Mr. Thomas K. Landauer
0202 - University of Colorado             303 492 2875 fax or 800...
020201 - Mr. Walter Kintsch

SUBJECTS
Calculate Cost Benefit Communication
Errors in Communication
Meaning, Build and Maintain Shared, Over
Cognitive Science, Meetings
Fragility of Knowledge, Fades
Maintain Shared Meaning/Knowledge

1808 -    ..
1809 - Summary/Objective
1810 -
181001 - Follow up ref SDS 4 0000, ref SDS 2 0000.
181002 -
181003 - Received via Internet, ref DRT 1 0000 from Tom responding to my
181004 - letter, ref DIP 1 0000 issued on 970725, ref SDS 4 7326, asking for
181005 - help in finding a study that supports the hypothesis in his LSA paper
181006 - that word meanings are fluid and are in constant flux, from the review
181007 - of Tom's paper on 960518. ref SDS 1 4488, and at ref SDS 1 3734.
181008 -
181009 - Tom says the closest thing is an article by Ferstl and Kintsch on
181010 - changes in word associations as a function of reading. it's
181011 - unpublished. if you email wkintsch@clipr.colorado.edu and give him a
181012 - paper-mail address he might be able to send you a copy.
181013 -
181014 - Also sent an email to Tom's colleague, Walter Kintsch, who seems to be
181015 - head of the psychology department at the University of Colorado.  It
181016 - was similar to the letter to Tom, so I did not enter it separately.
181017 -
181018 -  ..
181019 -
181020 - Received reply from Walter saying the study I propose is interesting,
181021 - but he does not see a relationship with LSA, ref DRT 2 7499.
181022 -
181023 -    I wonder if the model would predict the rate of meaning drift as a
181024 -    function of exposure rate to new words, described in the analysis
181025 -    on 960518, ref SDS 1 3734, as described in Tom's paper on page 41,
181026 -    ref OF 1 line 1552.
181027 -
181028 -       In contrast, the LSA simulations computed an increment in
181029 -       probability correct for every word in the text (as well as every
181030 -       other word in the potential vocabulary).
181031 -
181032 -       Thus, it [LSA] implicitly expresses the hypothesis that word
181033 -       meanings grow continuously and that correct performance on a
181034 -       multiple choice vocabulary test is a stochastic event governed
181035 -       by individual differences in experience, by sampling of
181036 -       alternatives in the test items and by fluctuations, perhaps
181037 -       contextually determined, in momentary knowledge states. As a
181038 -       result, word meanings are constantly in flux, and no word is
181039 -       ever known perfectly. So, for the most extreme example, the
181040 -       simulation
181041 -
181042 - Since Tom's response says to contact Walter, ref DRT 1 8477, and
181043 - Walter says their work does not relate to my analysis, we seem to be
181044 - at a dead end.
181045 -
181046 -    [On 970731 follow up sending letter to Walter using above quote.
181047 -    ref SDS 5 0070]
181048 -
181049 -    [On 980813 notified Tom about web site that is linked to analysis
181050 -    of his book and the paper on Plato's Problem. ref SDS 7 0000]
181051 -
181052 -
181053 -
181054 -
1811 -