Motivation and Intentionality in a Computer Simulation Model of Paranoia

115 Pages·1978·4.54 MB·English

by  William S. Faught (auth.)

Editor / Herausgeber:
Prof. Salomon Klaczko·Ryndziun, Frankfurt a. M.
Co-Editors / Mitherausgeber:
Prof. Ranan Banerji, Temple University, Philadelphia
Prof. Jerome A. Feldman, University of Rochester, Rochester
Prof. Mohamed Abdelrahman Mansour, ETH. Zürich
Prof. Ernst Billeter. Universität Fribourg. Fribourg
Prof. Christof Burckhardt, EPF, Lausanne
Prof. Ivar Ugi, Technische Universität München
Prof. King·Sun Fu, Purdue University, West Lafayette
Prof. Gerhard Fehl, R. W TH., Aachen
Dr. -Ing. Ekkehard Brunn, Universität, Dortmund
Interdisciplinary Systems Research Interdisziplinäre Systemforschung
AnalysIs - Modeling - Simulation Analyse - Formallslerung - Simulation
The system science has been developed from several Die System wissenschaft hat sich aus der Verbindung
scientific fields: control and communication theory, mehrerer Wissenschaftszweige entwickelt: der Rege
model theory and computer science. Nowadays it ful lungs-und Steuerungstheorie, der Kommunikationswis
fills the requirements which Norbert Wiener formulated senschaft, der Modelltheorie und der Informatik. Sie
originally for cybernetics; and were not feasible at his erfüllt heute das Programm, das Norbert Wiener mit
time, because of insufficient development of computer seiner Definition von Kybernetik ursprünglich vorgelegt
science in the past. hat und dessen Durchführung zu seiner Zeit durch die
Research and practical application of system science noch ungenügend entwickelte Computerwissenschaft
involve wOl'ks of specialists of system science as weil stark eingeschränkt war.
as of those from various fields of application. Up to Die Forschung und die praktische Anwendung der Sy
now, the efficiency of this co-operation has been proved stemwissenschaft bezieht heute sowohl die Fachleute
in many theoretical and practical works. der System wissenschaft als auch die Spezialisten der
The series 'Interdisciplinary Systems Research' is in Anwendungsgebiete ein. In vielen Bereichen hat sich
tended to be a source of information for university diese Zusammenarbeit mittlerweile bewährt.
students and scientists involved in theoretical and ap Die Reihe «Interdisziplinäre Systemforschung» setzt
plied systems research. The reader shall be informed sich zum Ziel, dem Studenten, dem Theoretiker und
about the most advanced state of the art in research, dem Praktiker über den neuesten Stand aus Lehre und
application, lecturing and metatheoretical criticism in Forschung, aus der Anwendung und der metatheore
this area. It is also intended to enlarge this area by in tischen Kritik dieser Wissenschaft zu berichten.
cluding diverse mathematical modeling procedures Dieser Rahmen soll noch insofern erweitert werden, als
developed in many decades for the description and op die Reihe in ihren Publikationen die mathematischen
timization of systems. Modellierungsverfahren mit einbezieht, die in verschie
In contrast to the former tradition, which restricted the densten Wissenschaften in vielen Jahrzehnten zur
theoretical control and computer science to mathemati Beschreibung und Optimierung von Systemen erarbeitet
cians, physicists and engineers, the present series em wurden.
.phasizes the interdisciplinarity which system science Entgegen der früheren Tradition, in der die theoretische
has reached until now, and which tends to expand. City Regelungs-und Computerwissenschaft auf den Kreis
and regional planners, psychologists, physiologists, der Mathematiker, Physiker und Ingenieure beschränkt
economists, ecologists, food scientists, sociologists. war, liegt die Betonung dieser Reihe auf der Interdiszi
political scientists, lawyers, pedagogues, philologists, plinarität, die die System wissenschaft mittlerweile er
managers, diplomats, military scientists and other spe reicht hat und weiter anstrebt. Stadt-und Regionalpla
cialists are increasingly confronted or even charged ner, Psychologen, Physiologen, Betriebswirte, Volks
with problems of system science. wirtschafter, Okologen, Ernährungswissenschafter,
The ISR series will contain research reports - including Soziologen, Politologen, Juristen, Pädagogen, Mana
PhD-theses -Iecture notes, readers for lectures and ger, Diplomaten, Militärwissenschafter und andere Fach
proceedings of scientific symposia. The use of less ex leute sehen sich zunehmend mit Aufgaben der System
pensive printing methods is provided to assure that the forschung konfrontiert oder sogar beauftragt.
authors' results may be offered for discussion in the Die ISR-Reihe wird Forschungsberichte - einschliess
shortest time to a broad, interested community. In or lich Dissertationen -, Vorlesungsskripten, Readers zu
der to assure the reproducibility of the published results Vorlesungen und Tagungsberichte enthalten. Die Ver
the coding lists of the used programs should be in wendung wenig aufwendiger Herstellungsverfahren soll
cluded in reports about computer simulation. dazu dienen, die Ergebnisse der Autoren in kürzester
The international character of this series is intended to Frist einer möglichst breiten, interessierten Öffentlich
be accomplished by including reports in German, Eng keit zur Diskussion zu stellen. Um auch die Reprodu
lish and French, both from universities and research zierbarkeit der Ergebnisse zu gewährleisten, werden in
centers in the whole world. To assure this goal, the edi Berichten über Arbeiten mit dem Computer wenn im
tors' board will be composed of representatives of the mer möglich auch die Befehlslisten im Anhang mitge
different countries and areas of interest. druckt.
Der internationale Charakter der Reihe soll durch die
Aufnahme von Arbeiten in Deutsch, Englisch und Franzö
sisch aus Hochschulen und Forschungszentren aus aller
Welt verwirklicht werden. Dafür soll eine entspre
chende Zusammensetzung des Herausgebergremiums
sorgen.
ISR58
Interdisciplinary Systems Research
Interdisziplinäre Systemforschung
William S. Faught
Motivation and Intentionality
in a Computer Simulation Model
ofParanoia
1978 Springer Basel AG
William S. Faught
CIP-Kurztitelaufnahme der Deutschen Bibliothek
Faught, William S.:
Motivation and intentionality in a computer simu
lation model of paranoia / William S. Faught. -
Basel. Stuttgart: Birkhäuser, 1978.
(Interdisciplinary systems research; 58)
All rights reserved. No part of this publication may be
reproduced, stored in a retrieval system, or transmitted,
in any form or by any means, electronic, mechanical.
photocopying, recording or otherwise, without the prior
permission of the copyright owner.
© Springer Basel AG, 1978
Originally published by Birkhäuser Verlag Basel in 1978.
ISBN 978-3-7643-1034-9 ISBN 978-3-0348-6547-0 (eBook)
DOI 10.1007/978-3-0348-6547-0
v
Abstract
This book describes a computer simulation model of paranoia. The model mimics
the behavior of a patient participating in a psychiatrie interview by answerlng questions,
introducing its own topics, and responding to negatively-valued (e.g., threatening or shame
producing) situations.
The focus of this work Is on the motivation al mechaniS"ms required to Instigate and
direct the modelIed behavlor.
The major components of the model are:
(I) A production system (PS) formalism accounting for the instigation and guidance
of behavior as a function of internal (affective) and external (real-world) environmental factors.
Each rule in the PS Is either an action pattern (AP) or an Interpretation pattern (IP). Both
may have either affect (emotion) conditions, external variables, or outputs of other patterns as
their initial condltions (Jeft-hand sides). The PS activates all rules whose left-hand sides are
true, selects the one with the highest affect, and performs the action specified by the right-hand
side.
(2) A model of affeets (emotions) as an anticipation meehanism based on a small
number of basic pain-pleasure faetors. Primary activation (raising an affect's strength) occurs
when the particular condition for the affect is anticipated (e.g., anticipation of pain for the fear
affect). Secondary activation occurs when an internal construct (AP, IP, belief) is used and its
associated affect is processed.
(3) A formalism for intensional behavior (directed by internal models) requiring a
dual representation of symbol and concept. An intensional object (belief) can be accessed either
by sensing it in the environment (concept) or by Its name (token). Similarly, an intensional .
action (intention) can be specified either by its conditions in the immediate environment
(concept) or by its name (token).
Issues of intelligence, psychopathological modelling, and artificial intelligence
programming are discussed. The paranoid phenomenon is found to be explainable as an
extremely skewed use of normal processes. Applications of these constructs are found to be
useful in AI programs dealing with error recovery, incompletely specified input data, and
natural language specifieation of tasks to perform.
VI
Preface
The work presented in this book is part of an attempt to develop a simulation model
of a particular theory of paranoia. The original goal was to simulate the behavior of a
psychiatrie patient in a first diagnostic interview. For this purpose it was necessary to develop,
in addition to the specific embodiment of the theory of paranoia, a natural language interface
and a number of behavioral responses to the myriad inputs possible. It soon became obvious
that extensive normal mental processing would also have to be modelled. Also, as the theory of
paranoia was elaborated and extended, it reHed more and more on an underlying base of
normal mental processes.
From the exploration of these normal processes, a second goal emerged. This goal is
to develop a model of human motivation al processes. It stems from the author's wish to
understand what the top lev~1 execution loop in humans is, and therefore what the top level
execution loop in autonomous intelligent computers ought to be. The focal point of this book is
the set of these motivation al processes.
This work attempts to:
(I) show the necessity for modelling motivational and intensional constructs in models
of human behavior (intensional constructs being characterized by internal representation);
(2) outline a simulation model of motivational processes as a base for a model of
paranoia; and
(3) discuss the implications of cognitive direction of conative processes for intelligent
computer programs.
Most of this research was performed at the Stanford Artificial Intelligence Laboratory
and at the Department of Psychiatry at VCLA. I am indebted to their principal sponsors, the
National Institute of Mental Health and the Advanced Research Projects Agency, for their
support.
A great many people helped me, directly and indirectly, to complete this work. I wish
to thank Dr. Kenneth Colby for his questions and suggestions, and for his untiring efforts to
make sense of the human mind. I am grateful to Roger Parkison for his analysis and
suggestions while my ideas were in the formative stage. Finally, I would Iike to thank alt of
those who came into contact with my work and helped in the completion of the ideas and
manuscript, induding David Shaw, Carole Parkison, Erin Colby, Cathy Villa, and Bob Filman.
Table of Contents VII
CONTENTS
Chapter 1. Introduction. ........................................................................................................................................................ 1
1.1 Overview. ..................................................................................................................................................................... 1
1.2 The humiliation theory of paranOia ................................................................................................................. 3
1.3 Psychopathology ...................................................................................................................................................... 4
1.4 Intensionality. .............................................................................................................................................................5
Chapter 2. Overview of the Simulation Mode 1.. .......................................................................................................... 7
2.1 T ask specification .................................................................................................................................................... 7
. 2.2 Actions .......................................................................................................................................................................... 9
2.3 Motivation. ................................................................................................................................................................ 12
2.4 Intensionality. .......................................................................................................................................................... 14
2.5 Implement at ion ........................................................................................................................................................ 17
Chapter 3. Review of related work. ............................................................................................................................. 21
3.1 Problem solving ......................................................................................................................................................2 1
3.2 Recognizing plans and actions ........................................................................................................................ 22
3.3 Simulations of motivation and belief systems .......................................................................................... 24
Chapter 4. Action and Interpretation Patterns ........................................................................................................ 27
4.1 Action Patterns ...................................................................................................................................................... 29
4.2 Pattern Activation ................................................................................................................................................. 30
4.3 Pattern Interpretation. ........................................................................................................................................ 32
4.4 Model .......................................................................................................................................................................... 33
4.5 An example of the interpreter cycle ........................................................................................................... 37
Chapter 5. Aflect ................................................... _. ........................................................................................................... 39
5.1 Theory of aflect ...................................................................................................................................................3 9
5.2 Components 01 the theory ................................................................................................................................ 40
5.3 Individual Affects ................................................................................................................................................... 41
5.4 Aflect activation in humans ........................................................................................ _ ................................... 43
5.5 Allects in the simulation mode! ...................................................................................................................... 43
5.6 Affect response ..................................................................................................................................................... 45
5.7 Issues and Implications ........................................................................................................................................4 7
Chapter 6. Intensional Constructs ..................................................................................................................................5 0
6.1 Background .............................................................................................................................................................. 51
6.2 Problems 01 Intensionality .................................................................................................................................5 2
6.3 Tokens vs Concepts .............................................................................................................................................5 4
6.4 Representation 01 Concepts, TOkens, and Links ..................................................................................... 55
6.5 Uses 01 Intensional constructs ........................................................................................................................ 57
6.6 Global Issues ............................................................................................................................................................6 0
Chapter 7. Meta"systems, Self-motivating systems, and Paranoia ................................................................. 62
7.1 Meta-systems ................................................................. _ ...................................................................................... 62
7.2 Self-Motivating systems ..................... _ .............................................................................................................6 5
7.3 Paranoia .....................................................................................................................................................................6 7
Chapter 8. Summary and Critique ............................................... _ .. _ ........................................................................... 72
8.1 Summary .........................................................._ ......................................... __ ......................................................... 72
8.2 Meta issues ................................................ __. _ _ .. _ ...... _ ...... _ ...... _ .... _ ................. _._. ... _ ..................... 73
8.3 Improvements to the model ..................... _ .... _ ...................................... _ .............. _ ..................................... 73
8.4 Application to Artificial Intelligence ......... _ ..................................................................................................7 5
VI I I Table of Contents
Appendix A. Sampie interview with the mode!. ....................................................................................................... 76
Appendix B. Patterns ........................................................................................................................................................... 85
Appendix C. Actions ............................................................................................................................................................. 91
Appendix D. Beliefs ................................................................ _. ..... _. ................................................................................ 92
Appendix E. Rules of Inference ....................................................................................................................................... 94
Bibliography. ................... _. ..... _. ................................ _. .............................................................................................................. 97
I. Introduction
Chapter 1. Introduction
The basic question 01 p.ycholog,. i.: Wh,. 110 men anti animal.
behalHl tlI 'he,. 1101
--Boden
Section 1.1 Overview
When a person acts, his behavior is guided by several motivation al factors. Among
these factors are his intern al needs (affects or emotions), sensory input from the external
environment, and internal mental constructs that represent the environment. To model the
motivation processes in a computer, we need a program which combines these motivational
factors in such a way that they operate as the sole motivation of the system and they guide
processing decisions at the most detailed level.
This book presents one way to model motivation al processes in a simple and
consistent structure. We implement the internal processes in a special production system (PS).
The PS incorporates rules for two types of processing: sequentially-performed actions and
parallel interpretation of situations. The factors that determine rule selection are afrects,
environmental conditions, and intensional constructs.
We believe that implementing complex theoretical models of human behavior in
computer simulations is a reasonable method of exploring those models. The computer
program in our work (PARRY3) simulates a paranoid patient. A psychiatrist sits at a teletype
and interviews the program as if it were a patient. The psychiatrist's task is to conduct a first
diagnostic interview with the model-patient. The program responds by answenng questions,
introducing its own topics, and dealing with negatively-valued (e.g., threatening or shame-
producing) situa.tions. .
We had three goals in constructing the simulation model. The first was to develop a
model of human motivation mechanisms that could account for the various factors that
influence human behavior. Although the task domain is Iimited to the behavior of a human
over a time span of about thirty minutes and the subject matter is Iimited to a psychiatrie
interview, the range of potential situations is large enough to provide a wide val iety of
behavior to explain.
The second goal was to explore a theory of paranoia. Earlier model vers ions of thf'
same theory contained special processes for the paranoid phenomena. In the current model, we
implemented a mechanism capable of generating paranoid behavior as an extension of the
"normal" motivation al processes. This implementation provides a consistency test for both
normal and abnormal processes.
The third goal was to construct a performance model of the theory of paranoia to
discover the consequences of the theory. A performance model is useful when cognitive models
become so complex that a person cannot determine all of the interactions by hand-simulating
the theory. A performance model also helps to insure the theory's completeness by requiring
the theory builder to account for all of the behavior within the task to be performed.
The book is organized as folIows.