[alicebot-general] Personality models (was: Big Five personalitytraits)

Tue May 23 09:03:06 PDT 2006

One thing I looked into for a while was statistical distributions of
personality types.  MBTI for example has 16 possible types, but when
they do their studies, all 16 are not equally distributed in the
population.  Not surprisingly, there is a Zipf curve of types.  One or
two are most prominent, one or two are very rare.  Which leads me to
wonder, have they got the theory right?

Kiersey Temprements are a simplified version of Myers-Briggs.  I have
an interesting video called "Please Understand Me" that describes 4
basic personality types.  Kiersey also mentions that the types are not
equally distributed, two are more common, two more rare.  He also goes
into the typical problems in relationships when people marry between
types or within their own type.

In the Enneagram community, there are often claims that some types are
more dominated by females and others by males.   Once I signed up for
an Enneagram panel of 4's, and the organizer was so pleased to have me
because she said it was so rare to find a man who was a type Four.
There could be other factors at work, because I know several  homeless
men who could be Fours, but they don't show up on the radar because
they don't have jobs, careers, or the kind of lives that would lead
them to Enneagram panels.

I tend to agree with you that an experienced "reader" can get
someone's personality within a few minutes, even without a formal
personality test, maybe with only two or three key questions.  What
you don't want to do is tell someone their personality type, which is
a kind of personality-ism akin to racism or telling someone their
religion.  If you guess someone's type, keep it to yourself (even if
they ask, unless they've paid you for an analysis).    It's up to them
to figure it out and self-identify.  The best you can do is point them
in the right direction.

On 5/23/06, Chris Lofting <chrislofting at ozemail.com.au> wrote:
>
>
> > -----Original Message-----
> > From: alicebot-general-bounces at list.alicebot.org [mailto:alicebot-general-
> > bounces at list.alicebot.org] On Behalf Of Helio Perroni Filho
> > Sent: Tuesday, 23 May 2006 10:19 PM
> > To: Alicebot and AIML General Discussion
> > Subject: Re: [alicebot-general] Personality models (was: Big Five
> > personalitytraits)
> >
> > --- "Dr. Rich Wallace" <drwallace at gmail.com> escreveu:
> >
> > > For bot building, personality types have at least
> > > two applications: one is building a set of "stock"
> > > personalities that are easily customized into
> > > individuals when combined with biographical data.
> > > The other is building personality tests into bots
> > > (like CLAUDIO) so that the bot's personality can
> > > adapt to the client's type.
> >
> > I also think that, when properly described, a bot's
> > personality is a natural source for good query
> > responses. In this regard, I've found Anthony
> > Gregorc's Mind Styles [1] (used by Peter Plantec in
> > his book Virtual Humans [2]) to be a very useful tool,
> > as it lends itself well for the creation of
> > compreensive character profiles.
> >
> > [1] http://www.gregorc.com,
> > http://facultyweb.cortland.edu/andersmd/learning/Gregorc.htm
>
> These just work off two dichotomies and as such are weak. MBTI works off
> four dichotomies, big-5 off 5.
>
> The IDM work shows the ability to go way beyond any of this covering a range
> of 6 to 12 dichotomies (and so 64 to 4096 possible particulars that then get
> customised to be singular and so unique) and that leads into detecting
> 'purpose' in types, IOW each type has a goal. You cannot get this sort of
> detail from four or less dichotomies as self-referencing needs to develop a
> bit before one can get good details and so extract a type's "spectrum".
>
> What of interest is that it takes only three to six questions to get access
> to someone's spectrum and so identify a LOT of detail that is hidden, even
> from them (I have a system based on emotions that can often 'bypass'
> conscious censorship and extract some 'need' of an individual - this is
> based on understanding emotions and semi-autonomous and so operating on
> their own - and so the ability for our rational minds to see all is 'ok'
> whilst we feel 'uncomfortable' where the discomfort comes from our emotions
> reacting to the context)
>
> As such, the template derived in IDM can be used to flesh-out (1) way beyond
> its current form of presentation.
>
> The SCIENCE perspective on persona is dependent of statistical analysis and
> so a degree of sameness. True personality, one's unique nature, is outside
> of Science since it is incomparable. Thus our unique nature can act as a
> random element in interactions and so innovations come out of the singular
> more so than the particular.
>
> These type tests cover genetically-determined expressions that reflect what
> we find in other collectives - be they of primates or ants or bees -
> specialist types such as warriors, drones etc These types develop for the
> sake of the collective and has such have no 'individual' nature as such.
> They are like sperm where a lot try to fertilise one egg. Each sperm is
> 'meaningless' in this sort of functionality; they all have the same goal.
>
> With the development of our individual consciousness, about 24 months after
> birth, comes true uniqueness where sensory differentiation plus exposure to
> local context shapes the personality beyond that 'seeded' in the genetics.
>
> Thus Science-derived categories fit particular natures, but to get a
> singular means developing a personal history to contribute, to refine,
> expression.
>
> That said, with understanding of the structure of emotions we can get a lot
> of information but it is 'vague' to the individual.
>
> Chris.
>
>
> _______________________________________________
> This is the alicebot-general mailing list
> Reply to alicebot-general at list.alicebot.org
> Unsubscribe and change preferences at http://list.alicebot.org/mailman/listinfo/alicebot-general
> Learn netiquette at http://www.dtcc.edu/cs/rfc1855.html
> Learn to read at http://www.literacy.org/
>