[ExI] Point of emotion

Stathis Papaioannou stathisp at gmail.com
Tue Apr 22 12:03:43 UTC 2008

2008/4/22 The Avantguardian <avantguardian2020 at yahoo.com>:

>  Are you suggesting that emotions are necessary prerequisite for goal-
>  oriented cognition? Are not certain psychopathologies like major
>  depression or schizoid presonality disorder characterized by a lack of
>  emotion? While depressed people may not be operating well in a social
>  sense, I don't think one could say that they no longer are capable of
>  complex thought.

A better example is schizophrenia with predominantly negative
symptoms. These patients have what is called "blunted affect", a
general flattening of the emotional landscape which can be likened to
reducing the dynamic range of an audio recording. They are cognitively
intact insofar as their memory and ability to reason logically is not
impaired. However, they invariably lack motivation, and they sometimes
show poor judgement when they have to make decisions (although more
commonly impaired judgement is due to positive psychotic symptoms).
This is because there isn't such a gradient of desirability separating
different choices, including the choice of sitting around doing
nothing. Now one can imagine improving the situation by increasing the
dynamic range without increasing the intensity of emotion. For
example, a person can easily play a game in which token values are
applied to different outcomes, with the aim of maximising the total
score: 25 is clearly a better score than 24, even though the absolute
and differential emotional impact of these numbers is minimal. An AI
could therefore be given pseudo-emotions using some similar system.
But if that's the case, why didn't nature give us such
pseudo-emotions? Could it be that a rich enough system of
pseudo-emotions would ipso facto *be* regular emotions, in the same
way as a rich enough system of environmental interaction is thought
(by some) to *be* consciousness, without the need for any extra
special sauce?

Stathis Papaioannou

More information about the extropy-chat mailing list