Re: Personal Constructs and Memes

Joe E. Dees (
Tue, 1 Jun 1999 13:27:25 -0500

Message-Id: <>
From: "Joe E. Dees" <>
Date: Tue, 1 Jun 1999 13:27:25 -0500
Subject: Re: Personal Constructs and Memes

From: "William Chambers" <>
To: <>
Subject: Personal Constructs and Memes
Date sent: Tue, 1 Jun 1999 11:32:21 -0500
Send reply to:

> Hans-Cees,
> Thanks for your prompt and friendly response, I will try to answer your questions.Your responses were very helpful.
> A repertory grid is a matrix of ratings or rankings in which some set of elements are elaborated along some construct dimension, For example, we might use a list of ten memes as the elements and rate them on a list of eight constructs, such as promotes civility, historically relevant,
ambiguous, and so forth. The elements are generally listed as the column headings and the constructs as the row headings,
> Another type of grid is called the coordinate grid (my invention). In this grid the row and column headings are the same and are called figures, The person ranks the figures to one another according to some criterion, such as "general similarity." The coordinate grid can be analyzed for
logical consistency and integrative complexity. Logical consistency concerns the transitivity of the rankings, If a person looses track of her definitions or is psychologically motivated to ignore or fabricate some aspects of the "general similarity" space in some comparisons (for example, the
similarity of wife to mother), then inconsistencies emerge in the analysis, My studies have shown that the coordiate grid measure of logical inconsistency is sensitive to simulated contradictions and lies, Random grids are the most inconsistent, followed by those of schizophrenics, neurotics
and young child
> Since we can choose what to use as figures, we can objectively assess the logical consistency of just about any person's views of anything, making the assessment based on the person's own premises. Since it is possible for those premises to be factually incorrect though consistent with one
another, consistency does not necessarily imply truth. Many bigoted people are logical so long as they restrict their constructions to certain premises and information. This makes them very difficult to reason with. Inconsistency, on the other hand, does imply untruth.
> Integrative complexity concerns the extent to which each figure contributes an equal amount of information in defining "general similarity." This is especially important when the figures are people, Poor integrative complexity is usually a type of stereotyping, People high in integrative
complexity tend to be more flexible and better at integrating information, Older kids also tend to be more integratively complex. Perfect integrative complexity produces a self similar pattern of elaboration:
> 1 2 3 4
> 2 1 4 3
> 3 4 1 2
> 4 3 2 1
There are 24 possible distinct juxtapositions of four numbers, and
six possible of three (and 120 possible of five, 720 possible of six,
etc. n! (n factorial)(1 x 2 x ...x (n - 1) x n) determines the number
of possible combinations for any quantity n of distinct numbers).
Just keeping the math straight so that your proposed grid will
circumscribe all possibilities.
> This can be expanded in size by doubling to infinity, The above matrix, which I call the mandala grid, is also perfectly logical and serves as a model of healthy construction of persons, The Platonic solids and other forms, which some people refer to as "sacred geometry" are also perfectly
integratively complex.
> My measure of causation allows us to infer causation from correlation, Do not laugh too hard, I have presented the method to thousands of statisticians and not one has been able to disprove it. Several eminent methodologists have admitted that it appears to work, Most people just dismiss it
without being able to argue their case with logic or data, Ad hominems are still alive and well in academe. Anyway, corresponding regressions/ correlations allows to determine is some variable (whole) is made up of other variables (parts). This is what Aristotle called formal causation. I have
a paper that can be downloaded on this the following sight:
> Additional papers are now under review and it looks like they are going to be accepted.
> Corresponding regressions/correlations can be used to trace the evolution of variables.
> The distinction between loose and tight constructs is found in George Kelly's personal construct theory, In essence, a loose construct is one that retains its identity but changes its element relations with the context, The personal construct literature has had a great deal of trouble
distinguishing loose from illogical construction but according to Kelly they are not the same (the coordinate grid measure does not confound the two). An example of loose but logical construction might be getting different factor structures for the same constructs, across male versus female
elements. There is nothing illogical about this. Friendly may have a common abstract meaning for both sets of elements but differ in details for the two sets of elements.
> Well that is my ten minute version of 25 years, Are there any parallels between personal constructs and memes?
> Bill

This was distributed via the memetics list associated with the
Journal of Memetics - Evolutionary Models of Information Transmission
For information about the journal and the list (e.g. unsubscribing)