00100	EVALUATION 
00200	
00300		The primary aim in constructing this model  was  to  explore,
00400	clarify, develop, test and improve -all with a model- a theory having
00500	explanatory value. To satisfy this aim, the model must meet norms  of
00600	internal    consistency   (systemicity)   and   norms   of   external
00700	correspondence with observation (testability). A secondary aim  would
00800	involve pragmatic norms of application.  These aims are not unrelated
00900	but the primary one is more  fundamental  since  useful  applications
01000	require some degree of consistency and correspondence to observation.
01100		As emphasized in Chapter  2,  a  model  in  the  form  of  an
01200	algorithm  consists  of  a structure of functions or procedures whose
01300	inner workings are  sufficient  to  reproduce  the  outward  symbolic
01400	behavior  under  consideration.   The theory embodied in the model is
01500	revealed  in  the  set of statements which illuminate the connections
01600	betweeen input and output, i.e.  which  describe  how  the  structure
01700	reacts under various circumstances.
01800		What constitutes a satisfactory explanation has been  treated
01900	in  Chapter  2. The "fit" or correspondence with facts of observation
02000	as indicated by  measurements  and  empirical  tests  indicating  the
02100	degree of faithfulness of the reproduction, were described in Chapter
02200	6. Given that the model has met the above criteria, what does  it  as
02300	an artifact tell us about naturally-occurring paranoid processes?
02400		First, the model attempts to revisualize  or  reconceptualize
02500	factors  underlying  the  phenomena  of  paranoid disorders. It draws
02600	attention to factors (such as the scan for malevolence as  protection
02700	against  humiliation)  which  might  not otherwise be attended to and
02800	which have therapeutic  implications.   Paranoid  disorders  are  not
02900	viewed  as first-order "diseases" but as a mode of processing symbols
03000	secondary to a primary disturbance.     The  patterns  of  linguistic
03100	paranoid  behavior  observed  in  an  interview  are  produced  by an
03200	underlying organized structure of rules  and  not  by  a  variety  of
03300	random  and unconnected mechanical failures.   Second, the underlying
03400	structure is posited to consist of an algorithm, an  organization  of
03500	symbol-processing  strategies or procedures.   Third, the model as an
03600	analogy indicates that to change this structure, its procedures  must
03700	be  accessible  to  reprogramming in the higher-level language of the
03800	algorithm. Finally, in the interests of a general conceptual  reform,
03900	the  model  suggests  that  other types of psychopathologies might be
04000	viewed from a symbol-processing standpoint.
04100		Decision procedures for consensus acceptability  of  a  model
04200	sometimes  depend  not  so  much  on  truth,  an elusive state, as on
04300	whether a majority of the  relevant  expert  community  believes  the
04400	theory  or  model to approximate truth to some unknown and unknowable
04500	degree and to be better than promising available  alternatives,  that
04600	is,  to  be the best we can do for the time being.   (Model builders,
04700	and perhaps all  scientists,  design  or  discover  structures  which
04800	delight  them,  not  because  they  are  true  but  because  they are
04900	beautiful). A model is tenable as long as it is worth working with by
05000	improving  it,  extending it, devising experiments and tests to probe
05100	it, and applying it in contexts of practical  action.  Validation  is
05200	ultimately a private experience of the individual. Empirical truth or
05300	falsity cannot be proven with certainty, but their  presence  can  be
05400	assayed by some sort of critical assessment and deliberation.  We can
05500	forgive models for being only nearly true.    A theory or  model  may
05600	bring  cognitive  or  pragmatic  comfort,  not because it is TRUE but
05700	because it is a workable and exciting possibility which represents an
05800	improvement over its contending rivals.
05900		Cognitive comfort is a  type  of  intellectual  satisfaction.
06000	Pragmatic comfort accrues from technically exploitable knowledge,i.e.
06100	applications which make things work the way humans want them to  work
06200	efficiently  in  practical  contexts of technological action. For the
06300	pragmatist, a model is a means to an end; for  the  theoretician,  an
06400	explanatory model is an end in itself. It is hoped that this paranoid
06500	model can contribute to understanding one of the mysteries  of  human
06600	conduct,  the paranoid mode. There remains the enigma of the paranoid
06700	"streak" which  renders  whole  nations  susceptible  to  ideological
06800	convictions  in  which  Elsewhereans  are  believed  to be malevolent
06900	oppressors.
07000		It  is  a  truism  of  methodology textbooks that an infinite
07100	number of theories or  models  can  account  for  the  same  data  of
07200	observation.        Without   questioning  whether  "infinite"  means
07300	indefinitely large or just more than one, we  must  allow  for  rival
07400	explanations.  For a rival to be a live and tenable option, it should
07500	be truly alternative (i.e., not just a family version saying the same
07600	thing  in  a  different way), and be confirmable or disconfirmable by
07700	tests.
07800		Prediction of new facts from a theory not only test a  theory
07900	but  provide  useful information. Not all acceptable theories predict
08000	new facts, e.g. Copernicus'. Although I would maintain that  faithful
08100	reproduction  (fidelity  as  measured  by  indistinguishability along
08200	specific dimensions) is a proper and major test for the  adequacy  of
08300	simulation models, it would be a bonus if our model could satisfy the
08400	function of making possible new knowledge through prediction.     The
08500	term  "prediction"  has a spectrum of meanings ranging from forecasts
08600	to prognoses to prophecies to precise point-predictions in time.   To
08700	predict  (and  to  postdict)  from a theory or model is to derive and
08800	announce  a  fact  prior  to  knowledge  of  its  actual  occurrence.
08900	However  one  needs  knowledge  of  the  kind  of  fact expected, the
09000	conditions which produce it and the circumstances under which it will
09100	occur.  The  interest  in  prediction  may  stem from a desire (1) to
09200	confirm or disconfirm a theory or  model  or  (2)  to  obtain  useful
09300	information  about  the future, as in weather forecasting.  Celestial
09400	mechanics provides the  ideal  of  accurate  long-range  predictions.
09500	But  even  astronomers,  with  the advantage of studying isolated and
09600	repetitive systems, have their troubles.    In  1759  Halley's  comet
09700	arrived  four  days later than predicted.    In spite of our advanced
09800	20th century knowledge, a prediction made in 1962 was  off  by  eight
09900	days, that is, twice as bad. (In fairness we must make allowances for
10000	the fact that great masses, distances and  velocities  are  involved.
10100	Also comets defy Newton's law of gravity).
10200		Predictions of individual human behavior are severely limited
10300	by  our  restrictions  of  knowledge.    For  example, (1) sufficient
10400	knowledge of initial conditions may require that we  know  the  whole
10500	past  history of an individual (something not yet achieved for even a
10600	single person), (2) individuals do not remain isolated over the  time
10700	stretch of the prediction; they interact with other individuals of an
10800	unknown nature, (3)  since  life  is  a  fortuitous  flux  of  chance
10900	intersections of independent causal chains, one would also have to be
11000	able to foresee events of the physical environment and  its  changes,
11100	(4)  the  process  of  observation  needed  to obtain information for
11200	predictions may have non-negligible and unforeseeable effects on  the
11300	observed.
11400		In  one  sense  our  paranoid  model  makes  moment-to-moment
11500	predictions and asserts  new  counterfactuals  about  behavior  in  a
11600	psychiatric  interview.  That  is,  if  an  interviewer  says X under
11700	conditions Y, then the model's  response  will  be  characterized  by
11800	z1...zn,   and   the   same   holds   true   for  paranoid  patients.
11900	Counterfactual  prediction  means  that  on  the  basis  of  observed
12000	behavior  we  are willing, with  an  inductive  risk,  to  assume the
12100	presence of unobserved behavior potentials in a model's or  patient's
12200	repertoire of capabilities.
12300		Predicting  new  kinds  of  events  or properties, instead of
12400	kinds we are already familiar with, would represent a genuine  bonus,
12500	indicating  the  model is more than ad hoc and has excess content. It
12600	would give both clinicians and investigators something to  look  for.
12700	This  novelty  could  arise  in  two  ways.    First, the model might
12800	demonstrate a property  of  the  paranoid  mode  hitherto  unobserved
12900	clinically.      In  principle  this could come about because the I-O
13000	behavior of  the  model  is  a  consequence  of  a  large  number  of
13100	interacting  hypotheses  and  assumptions chosen initially to explain
13200	frequently observed phenomena.    When the elements of such a complex
13300	conjunction  interact  with  highly  variable  inputs  they  generate
13400	consequences in addition to those  they  were  designed  to  explain.
13500	Whether  any  of these consequences are significant or characteristic
13600	of the paranoid mode remains a subject for future study.
13700		It  is  also  possible that a new property of paranoia may be
13800	discovered in the clinical  interview,  although  perhaps  everything
13900	that  can  be  said about paranoid dialogues has been said.  If a new
14000	property were found, a search  for  it  might  be  conducted  in  the
14100	model's  behavior. If successful, this again would add to the model's
14200	acceptability.
14300		A second novelty might arise in the behavior of the model  in
14400	some  new situation.   Since it is designed to simulate communicative
14500	behavior in an interview situation, the "new" circumstance would have
14600	to involve some new type of linguistic interaction to which the model
14700	is capable of responding. From its behavior one  might  then  predict
14800	how paranoid patients would behave under similar circumstances.   The
14900	requisite empirical tests and  measures  would  show  the  degree  of
15000	correspondence between patient and model behaviors.
15100		This possibility is of importance in considering emancipatory
15200	therapies for patients  entangled  in  the quandaries of the paranoid
15300	mode.    Since the model operates at a symbol processing level  using
15400	natural  language,  it  is      this  level  at  which linguistic and
15500	conceptual skills of clinicians can be  applied.   Language-based  or
15600	semantic  techniques  do not seem very effective in the psychoses but
15700	they are useful in states of lesser severity. A  wide  range  of  new
15800	semantic  techniques, including extremes, could be tried first on the
15900	model without subjecting patients to blind experimentation.
16000		While we have used the model principally to explore a  theory
16100	and  to study psychiatric judgements, its potential use as a training
16200	device has  not  escaped  our  reflections.    Medical  students  and
16300	psychiatric  residents  need  "disposable  patients"  to  practice on
16400	without jeopardy (to either). A version of  the  paranoid  model  can
16500	display  the  changes  in  its  inner  states  during  an  interview.
16600	Whether  the  optimal  goal  of  interviewing   (gathering   relevant
16700	information  without  upsetting  the patient)  has been achieved, can
16800	thus be estimated.     A  beginning  interviewer  could  practice  in
16900	private or with a supervisor present. Many interviewers have reported
17000	that the model has a definite effect on them.    The student can  get
17100	the  feel  of  the  paranoid mode long before he interviews an actual
17200	patient.     The effect  of  various  interviewing  styles  might  be
17300	studied and compared.
17400	
17500		Although this simulation of  paranoia  covers  a  variety  of
17600	phenomena,  it  is  circumscribed in what it attempts to explain. The
17700	proffered explanation is local and restricted in that it accounts for
17800	only  one  type  of  symbol-processing mode.  Past attempts at grand-
17900	scale explanations of all  mental  processes  in  all  contexts  have
18000	failed.   A  preferable strategy, successful in other sciences, is to
18100	build one circumscribed and tested theory or model at a time so  that
18200	the  field  can  gradually  move  forward a step at a time, each step
18300	gaining consensus before attempting the next.