emotions

CSGnet Archive CSG1993
1

[From: Bruce Nevin (Tue 930928 09:30:33 EDT)]

( David Goldstein 09/25/93 (Sun, 26 Sep 1993 21:25:13 EDT) )

I read _Shogun_ a while back and remembered Todayama citing a list of
seven emotions. I assume the author got these from some Japanese source.
The list:

    The seven emotions, hate, adoration, joy, anxiety, anger, grief, fear.
    (Book III, Chapter 38 (p. 593))

    There are seven emotions, neh? Joy, anger, anxiety, adoration,
    grief, fear, and hate. If a man doesn't give way to these, he's patient.
    I'm not very strong, but I'm patient. (Book VI, Chapter 61 (p. 1081))

V. What is the difference between remembering what I felt in a
given situation and actually re-experiencing the feelings?

I suspect that we would see the physiological
activity indicative of a stress response in the later case but
not the former case.

I suspect that "remembering what I felt" involves recitation of a verbal
description of the experience.

It has been suggested (I forget where) that we do not retain memories of
early childhood because they are pre-verbal, and the memory of
experiences is organized around memory of verbal descriptions. "Reliving
an experience" is itself an experience that one can describe as it is
happening, sort of, but is not itself organized around a pre-existing
story.

Either mode of remembering can evoke the other.

    Bruce Nevin
    bn@bbn.com

2

[From: Bruce Nevin (Thu 930930 09:02:48 EDT)]

( David Goldstein 09/29/93 (Thu, 30 Sep 1993 06:26:14 EDT) ) --

Gracious, David, I had no intention of suggesting anything about the
origin of Plutchik's octochotomy of emotions. The Japanese example
(which took me a while to locate) was a follow-up to my request to Greg.

I put Plutchik's labels on my whiteboard to think about. I made some
changes to make them more parallel to one another:

                 Happiness

     Expectation Acceptance
                     >
                  \ | /
                     >
  Anger -----+----- Fear
                     >
                  / | \
     Rejection | Surprise

                 Sadness

(You could also change them to happy/sad, accepting/rejecting,
expectant/surprised or expecting/surprised, angry/fearful. These are
less detached terms, describing an active, ongoing experience of the
feeling. The ones above describe states or conditions abstracted from
experience. Mixing the two kinds of terms muddles points of view.)

Adoration, joy, grief, and anxiety are four in the Shogun set that don't
appear here. Manfred Clines' work with what he calls sentics includes
reverence, probably others that I don't recall (couldn't find the book).
You could argue, I suppose, that joy and grief are extremes of happiness
and sadness, and that the terms in this wheel are primitive elements of
which complex feelings such as reverence, adoration, and anxiety are
constituted. Testable evidence? Does the evidence bear on feelings or
on word associations, and in what ways and in what degree do those
differ?

Emotions opposite to each other were opposite in
meaning. Emotions close to each other on the circle were similar
in meaning. Emotions at right angles to each other were
independent.

This is of course a statistical result reflecting something about a given
population. Not all of us associate surprise with fear and sadness, or
at least not all the time.

Joy I would find associated with acceptance and surprise, which is where
fear is associated in the above wheel. It appears to me that a charting
of such values beyond their relations as pairs of opposites requires more
than two dimensions.

It strikes me that there are at least four negative emotions (anger/fear,
rejection, sadness) but only two unequivocally positive ones (happiness,
acceptance). The other two are equivocal (expectation/surprise): some
people like surprises, some not (and I there are different sorts of
surprises); and one may have expectations of different sorts.

Those on the left side and top (happy, expectant, angry, rejecting) are
more active, and their opposites more passive, the old yang/yin. They
seem to emphasize a self perception of doing vs. one of being done to,
agent vs. patient.

I apologize, David, I don't remember just how you wanted to apply this in
your work, and I have deleted your original query. Perhaps this wheel
fits your purposes and my observations are not relevant to what you are
doing.

    Bruce
    bn@bbn.com

3

[Hans Blom, 961014]

(Bill Powers (961012.1830 MDT)), (Bruce Abbott (961012.1050 EST))

Hans Blom (961010) noted that it is similar to a view
proposed by Nico Frijda, in which emotions are "action tendencies,"
a perception (e.g., of danger) coupled to "an innate response or
action pattern (of flight, maybe)," but one in which "the coupling
between perception and action can be modified or overridden."

In a control system model, an "action tendency" (which I interpret
to mean a tendency to produce particular actions) has little
meaning; actions must adjust to suit the demands of environmental
disturbances as well as the set of operative reference levels. What
psychologists often label as "actions" turn out, in control theory,
to be controlled perceptual consequences of actions. Anyway, a
"tendency" is not a mechanism, so such a term does little to help us
visualized the underlying model.

Frijda uses the notion "action tendency" in opposition to fixed
stimulus-response behavior, where stimuli always results in the
responses that "belong to them". "Action tendencies" are, in his
view, much like control systems that lie dormant until they need to
be used. Thus a certain high-level perception, such as threat,
invokes a control system to become active in order to eliminate the
threat. If there is no threat, there is nothing that indicates that
such a control system even exists. Thus a "tendency". Such a
"tendency" IS a mechanism, akin to a PCT one-way control system being
activated by a reference level that suddenly causes action where
there was no such action before.

Such a control system is conceptual rather than physically separate
from other control systems; it may, to a very large extent, "reuse"
general system components such as those of locomotion, but for a
different high-level reason. Thus I consider Frijda's views to be
largely compatible to the PCT views, coupled to one-way control, i.e.
a zero reference level for threat. If there is threat, there is the
appropriate action; if there is no threat, the action apparatus can
be employed for different objectives.

I think Warren McCulloch proposed a model which might be of some use
here.
He was interested in the reticular formation (RF) and decided that
it was a modal control system. He hypothesized that animals are, at
any given moment, in some particular behavioral mode, such as
eating, exploring, nesting, fighting, defecating, and so on (I think
he had a list of 15 or so, but the exact number and contents of the
list aren't important here).
Each of these modes corresponds to a particular global pattern of
brain activation which best serves that particular mode. How does
the animal decide which mode to be in? McCulloch figured that the
RF monitored the entire state of the brain, including, of course,
the state of the intrinsic variables (to use BP's term). It would do
some computation over that state and commit the animal to the mode
which best served the current highest priority need. [MuCulloch's
particular interest was in the nature of this computation.]

I don't know whether McCulloch included behaviors that go together
with intense emotions. Intense emotions arise, however, when these
very normal behaviors are disallowed. Just observe that little boy
who isn't able to get to a bathroom ...

Anyway, it might be that an emotion "forces" a context switch and is
somewhat like an (unconscious) analog of what we consciously do when
we focus our attention: it focusses our attention FOR us in such an
inescapable way that it may save our lives.

Also, the idea of a "coupling between perception and action" strikes
me as a straightforward stimulus-response concept, implying that
there are certain perceptions that stimulate certain responses. A
control-system analysis of the same apparent "coupling" would, as
you know, involve structural relationships that are quite different
from a simple input-output model.

Not if one looks at observable behavior only. Given that the
controller has a fixed reference level, which must be a pretty good
approximation in highly emotional contexts, the "disturbances" of the
world pretty much govern the organism's ensuing actions. Thus an
external threat causes threat-avoiding actions. Simple stimulus-
response observations. A control systems analysis tells us, in
addition, _why_ this is so, of course ;-).

Anyway, one point that I raised is still unanswered. Why do
biologists say that instincts have no feeling component, whereas
emotions do? What's the difference? Why?

Greetings,

Hans

···

This ties in with what Bill Benzon <bbenzon@GLOBAL2000.NET> wrote:

4

Hans Blom, 961014 sez:

I think Warren McCulloch proposed a model which might be of some use
here.
He was interested in the reticular formation (RF) and decided that
it was a modal control system. He hypothesized that animals are, at
any given moment, in some particular behavioral mode, such as
eating, exploring, nesting, fighting, defecating, and so on (I think
he had a list of 15 or so, but the exact number and contents of the
list aren't important here).
Each of these modes corresponds to a particular global pattern of
brain activation which best serves that particular mode. How does
the animal decide which mode to be in? McCulloch figured that the
RF monitored the entire state of the brain, including, of course,
the state of the intrinsic variables (to use BP's term). It would do
some computation over that state and commit the animal to the mode
which best served the current highest priority need. [MuCulloch's
particular interest was in the nature of this computation.]

I don't know whether McCulloch included behaviors that go together
with intense emotions. Intense emotions arise, however, when these
very normal behaviors are disallowed. Just observe that little boy
who isn't able to get to a bathroom ...

Here's McCulloch's list ( on pp. 279-280 of Kilmer, McCulloch, and Blum, "A
Model of the Vertebrate Central Command System," Int. J. of Man-Machine
Studies (1969) 1, 279-309):

(1) sleep
(2) eat
(3) drink
(4) fight
(5) flee
(6) hunt (for prey of fodder)
(7) search (or explore)
(8) urinate
(9) defecate
(10) groom
(11) mate (or sex)
(12) give birth (or lay egg)
(13) mother the young (including suckling or hatching, retrieval, perineal
licking, and so on
(14) build or locate a nest
(15) and special innate forms of behavior such as migrate, hibernate, gnaw,
and hoard, depending on the species.

Anyway, it might be that an emotion "forces" a context switch and is
somewhat like an (unconscious) analog of what we consciously do when
we focus our attention: it focusses our attention FOR us in such an
inescapable way that it may save our lives.

McCulloch wasn't interested in emotion. He was interested in a technical
account of a computational procedure which would "force" the animal into
one or another of these modes. Given the nature of these behaviors (the
list above) one could reasonably expect such a mechanism to be strongly
associated with emotion. Also, those behaviors are closely linked to PCT's
innate variables.

How does PCT model the process by which one chooses to pursue one or
another of those behaviours? Given that one is already pursuing one of
them, that is, seeking perceptions which satisfy the need, PCT says alot
about how that pursuit is conducted. But what does it say about choosing,
e.g. to eat, hunt, urinate, or groom?

Anyway, one point that I raised is still unanswered. Why do
biologists say that instincts have no feeling component, whereas
emotions do? What's the difference? Why?

This seems to me to be a semantic quibble & I wouldn't expect much from an
attempt to answer it. We need a technical theory in which to pose such
issues. Is there any technical account of instinct which is as explicit as
PCT? And if PCT can get the animal through its daily rounds without
talking about instincts, why bother with them?

Now, I don't think PCT can do that. I think it needs to add something like
McCulloch's mechanism to its account of the relationship between the
perceptual stack and the reorganizing system, which is what Hays and I did
in our very speculative account of the brain (it's good intellectual
entertainment (I think) but not, alas, science).

···

This ties in with what Bill Benzon <bbenzon@GLOBAL2000.NET> wrote:

********************************************************
William L. Benzon 518.272.4733
161 2nd Street bbenzon@global2000.net
Troy, NY 12180 Account Suspended
USA
********************************************************
What color would you be if you didn't know what you was?
That's what color I am.
********************************************************

5

[Hans Blom, 961017d]

(Bill Benzon <bbenzon@GLOBAL2000.NET>)

Here's McCulloch's list ( on pp. 279-280 of Kilmer, McCulloch, and
Blum, "A Model of the Vertebrate Central Command System," Int. J. of
Man-Machine Studies (1969) 1, 279-309):

(1) sleep
(2) eat
(3) drink
(4) fight
(5) flee
(6) hunt (for prey of fodder)
(7) search (or explore)
(8) urinate
(9) defecate
(10) groom
(11) mate (or sex)
(12) give birth (or lay egg)
(13) mother the young (including suckling or hatching, retrieval,
perineal licking, and so on
(14) build or locate a nest
(15) and special innate forms of behavior such as migrate, hibernate,
gnaw, and hoard, depending on the species.

Thanks for the list. These resemble the descriptions that ethologists
use to categorize behaviors (ethograms), although the latter are much
more refined.

Thanks again.

Hans