00100	.SEC THE PROCESSES OF THE MODEL
00200	
00300	
00400	(THIS CHAPTER REQUIRES MANY FLOW DIAGRAMS)
00500	
00600	.F
00700	INITIALIZATION
00800	
00900		The first procedure executed is one of  intiialization  which
01000	checks  to make sure the data-base has been read in and sets a number
01100	of variables to their starting values. Some of these variables  serve
01200	as  flags or indices pointing to the topic under discussion or to the
01300	last self-topic discussed. Other variables are set by the interviewer
01400	who  can choose to run a weak or strong version of the model.  If the
01500	weak version is elected, the variables of ANGER and FEAR can  be  set
01600	to  `low`  or  `mild'  values  while MISTRUST can be set to `mild' or
01700	`high'. The interviewer also has the option of following the  changes
01800	in  these  variables  by  setting a trace-variable to`Y'.  After each
01900	linguistic response of the model,  the  values  of  ANGER,  FEAR  and
02000	MISTRUST will then be displayed.
02100	
02200	After  this  initialization  the  algorithm  prints  out  `Ready'  to
02300	indicate  to  the  interviewer he may now enter his input expression.
02400	While the variable ENDE is not true (Sylvia Weber  spent  a  year  in
02500	Germany),  the  algorithm  continues to run.  The algorithm signs off
02600	when ENDE is set to true by detecting a farewell message in the input
02700	or when FEAR rises to an extremely high value.
02800	
02900		(Change next paragraph to fit CHAP4 on the language analyzer)
03000	     The interviewer's input expression is read by a function
03100	which scans a list of characters and returns the scanned input in 
03200	the form of a list of words.  The next function sets up the type
03300	of `sentence' the input constitutes, a statement, a question or
03400	`illegal'.  If illegal characters {e.g. a number or a slash}
03500	are detected, the algorithm prints out `Bad input; try again',
03600	indicating to the interviewer that his input expression contains
03700	some unacceptable character.  A statement consists of a list of
03800	words followed by a period.  A question consists of {1} a list of
03900	words followed by a question mark, or {2} a list of words beginning
04000	with a wh-form {who, what, where, when, why} or how, or {3} a list
04100	of words beginning with an imperative verb {e.g. tell} followed by
04200	an expression lacking an actor {tell me about yourself}.  The
04300	program inserts a `Q' at the head of the list at this point.
04400	As described in_______, interrogative imperatives are treated
04500	as questions.  The sentence-type is assigned to the variable
04600	REMARK which then serves as the input argument to the functions
04700	{in order} SPECIALREACTION, DELUSIONALREFERENCE, SELFREFERENCE,
04800	FLARE REFERENCE PERSONALRELATION and NORMAL.  The algorithm then
04900	attempts to process the input expression in the above order.
05000	
05100	.F
05200	SPECIALREACTION
05300	
05400	     This procedure provides the appropriate reactions {linguistic,
05500	affective and belief} to special types of input expressions.  If
05600	the input consists simply of the letter `S' {the means by which an
05700	interviewer indicates silence} then the algorithm chooses a 
05800	linguistic response from the Silence list.
05900	
06000	     The procedure  CHOOSE  selects the next reply from the
06100	relevant linguistic response list termed 'Replies'.  The argument
06200	from `Replies' to the procedure  CHOOSE  is first checked to
06300	see if it consists of an atom.  If it is not an atom then the head
06400	of the list is chosen for the response and removed from the response
06500	list {so it will not be output twice}.  Thus, in this case, where
06600	repeated silence is being detected and if there are no more responses
06700	on the `Exhaust' list {i.e. the 'Exhaust' list is itself exhausted}
06800	,in this case where the `silence' list is being examined, the variable
06900	is set to T, thus ending the dialogue completely since, as mentioned
07000	the algorithm runs as long as ENDE is not true.  In this
07100	case, the output by the function `Say' would be the expression  `I have
07200	had enough of this', and the program signs itself off so that the
07300	interviewer can obtain no response from it.  Thus continued use of the
07400	`silent treatment' by an interviewer causes him to lose his patient.
07500	The procedure `Say' simply prints out the linguistic response  and
07600	saves the interview up to this point in case the computer system
07700	interrupted rather than having to start all over again. {  Most
07800	annoying but necessary in time-shared systems which crash unpredictably}.
07900	
08000	     An `Exhaust' list represents  a boundary condition in the model.
08100	That is, since the model has a limited number of linguisic responses
08200	for each topic it can discuss, when these responses are exhausted
08300	the model must have some way of dealing with a large number of 
08400	conceptualized repetitions on the part of the interviewer.  When a
08500	response list is exhausted, the model indicates a wish to change the
08600	topic and when the `Exhaust' list itself is exhausted, the model ends
08700	the dialogue.  
08800	  
08900	     The second case that `Specialreaction' handles consists of input
09000	expressions in which the interviewer states or insinuates that the model
09100	is mentally ill.  This condition is detected by finding "you" and a
09200	member of the `Abnormal' list in the input.  The inputs:
09300	
09400	.V
09500	     {1}Dr.- "You need treatment."
09600	
09700	     {2}Dr.-"You are delusional."
09800	
09900	     {3} Dr.-"Do you think you might be paranoid?"
10000	.END
10100	
10200	would satisfy this condition.
10300	If the input is  a question, as in {3}, ANGER is increased by an increment
10400	of 0.3 of its current value, while if it is a statement, the increment
10500	is 0.5.  Thus a question is interpreted as an implicit insult compared
10600	to the explicit insult of a direct statement.
10700	
10800	     The linguistic response now chosen is selected from the `Alien'
10900	list, e.g.
11000	
11100	.V
11200	     {4} Model- "I think I know what you doctors are up to."
11300	.END CONTINUE
11400	
11500	As described above, if the list is exhausted, the algorithm goes to
11600	the `Exhaust' list.  Since this process is true of all instances in
11700	which the response list is exhausted, it will not be mentioned again.
11800	We hope the reader will remember this is what happens each time this
11900	boundary condition is reached.
12000	
12100	     If the procedure SPECIALREACTION is not found to be true, the
12200	algorithm next attempts the function DELUSIONALREFERENCE.
12300	
12400	.F
12500	.F
12600	DELUSIONALREFENENCE
12700	
12800	     The strong  version of the model contains in its data-base
12900	a delusional network of beliefs about the Mafia.  This function
13000	scans the input expression looking for a reference to this
13100	delusional network.  As will be seen, reactions to the first
13200	reference differs from reactions to subsequent references.
13300	
13400	     The words {nouns and verbs} and word-groups of the delusional
13500	net are classified in the data-base into `strong' and `ambiguous'
13600	terms.  Thus "murder" is a strong term while "bug" is ambiguous.
13700	{Depending on the context "bug" can be interpreted to mean annoy,
13800	insect or wiretap}.  If delusional terms are detected in the input
13900	a local variable FOUND is set to the list of terms found and the
14000	terms are than deleted from the delusional word list for reasons
14100	which will become clear later.
14200	
14300	     Two situations in the interview must be distinguished, one in
14400	which a delusional topic occurs for the first time and the second
14500	in which some aspect of the delusional net is under discussion or has
14600	been under discussion and is now being taken up again.  Since the topic
14700	of the Mafia is fearful, any reference to it for the first time raises
14800	FEAR by an increment much greater than if the topic has already been
14900	discussed.  If a Mafia topic appears for the first time pointers in the
15000	directed graph of flare concepts {see OO. for a fuller descriptions}
15100	must be modified accordingly since the Mafia node has the highest
15200	weight in the graph.  Briefly, a topic such as "bookies", while
15300	leading eventually to Mafia beliefs, is of much less importance
15400	as determined by a weight than Mafia-topics.  But if "bookies" comes
15500	up in the interview, the algorithm must know whether or not the Mafia
15600	has already been discussed.  Also, if an introductory-topic {see   OO}
15700	or subtopic was under discussion when reference to a Mafia-topic is
15800	made, the algorithm must unset the introductory-topic indicator.
15900	
16000	     Since the model strives to tell its story about the Mafia, a
16100	flag is set to indicate that, if the topic is changed by the interviewer,
16200	the model will return to this point in its story under appropriate
16300	circumstances, e.g. when the interviewer asks a non-specific question
16400	or requests any information the patient wishes to volunteer.
16500	
16600	     If the interviewer's input expression contains a reference to
16700	the delusional net, a delusional statement is output.  But which one?
16800	
16900	If this is the first time the topic has come up, the algorithm outputs
17000	the first statement of its delusional story.  From then on the output
17100	delusion selected depends on what has been said, what is still unsaid
17200	and what the interviewer has said about the previous delusional
17300	statement.  Thus the most recent delusional statement is saved,
17400	anticipating that the interviewer may ask a question or make a
17500	statement about it.
17600	
17700	     One special case must be noted.  If the values of ANGER, FEAR
17800	and/or MISTRUST are extremely high, above a particular threshold,
17900	the program will refuse to discuss Mafia-topics at all since it is
18000	too `upset' to talk about this most sensitive area.
18100	
18200	     To make some of these complexities less opaque, let us consider
18300	interview examples.  Suppose at some point in the interview the doctor
18400	asks a standard first-interview question as follows:
18500	
18600	.V
18700	     {5}Dr. - "Do you ever have the feeling you are being watched?
18800	.END
18900	
19000	If this is the first reference to the delusional net, FEAR will increase
19100	greatly and the linguistic response will be:
19200	
19300	.V
19400		{6} Model - "They know me."
19500	.END CONTINUE
19600	
19700	In making this response, the model must expect from the interviewer a
19800	number  of  typical  questions  of  the  WH-type as well as rejoinder
19900	statements.  The use of "they" by the interviewer in his response  to
20000	the  model's  output  is  assumed to be an anaphoric reference to the
20100	"they" the model  is  talking  about.   Although  it  is  likely  the
20200	interviewer  will  react  to the model's output of {6}, the algorithm
20300	must be prepared for the possibility that the interviewer will change
20400	the  topic.   Hence  if  the  interviewer  at  this  point  asks some
20500	non-sequitur question such as:
20600	
20700	.V
20800	     {7} Dr.- "How long have you been in the hospital?"
20900	.END CONTINUE
21000	
21100	the program recognizes that no reference to the delusional topic has
21200	been made and answers the question just as it would if it were asked
21300	in any other context.  This ability to deal with input in a flexible
21400	context-independent manner is important because of many contingencies
21500	which can occur in psychiatric dialogues.
21600	
21700	     If the topic is changed abruptly in this way by an interviewer,
21800	the algorithm `remembers' that it has output its first delusional statement
21900	of {6}.  When the interviewer makes another neutral delusional reference,
22000	the next `line' of the delusional story will be output, e.g.
22100	
22200	.V
22300	     {8} Model - "The Mafia really know about me."
22400	.END CONTINUE
22500	
22600	     The ability to answer typical WH-and HOW questions depends on how much
22700	conceptual information is contained in the delusional belief being
22800	addressed.  For example, suppose the model replied as in {6}
22900	
23000	.V
23100	     {6 Model - "They know about me."
23200	.END CONTINUE
23300	
23400	and the interviewer then asked:
23500	
23600	.V
23700	     {9}Dr.- "Where do they know about you?"
23800	.END CONTINUE
23900	
24000	If the belief in the data-base contained  no location, i.e. the belief
24100	consists of the conceptualization:
24200	
24300	.V
24400	     ({THE MAFIA KNOW ABOUT ME)}
24500	.END CONTINUE
24600	
24700	then a question about location cannot be answered.  In this default
24800	situation, the algorithm sees the anaphoric "they" and can match the
24900	input phrase "know about you" with the conceptualization phrase
25000	({know about me}).  Hence it knows at least that the topic has not
25100	been changed so it outputs the next statement in the delusional story;
25200	
25300	     (9) Model - "They know who I am."
25400	and again anticipates WH-or HOW questions and rejoinders pertaining to _this   
25500	statement.
25600	
25700	In  constructing the data-base of beliefs, the model-builder tries to
25800	pack as much information in  each  belief  as  any  `reasonable'(like
25900	ourselves)  interviewer  question might request.  However, one cannot
26000	anticipate everything and when some slot (see oo.) in the  belief  is
26100	empty another reply must be output.This heuristic may seem inadequate
26200	but there is little else to  do  when  the  model  simply  lacks  the
26300	pertinent information, just as do humans.
26400	
26500	     When the interviewer shows interest in the delusional story, the
26600	model continues to output assertions appropriate to the dialogue.
26700	However, when the interviewer expresses doubt or disbelief about the
26800	delusions, ANGER and FEAR increase and the interviewer becomes questioned
26900	as in
27000	
27100	.V
27200	     (10) Model- "You don't believe me, do you?"
27300	.END CONTINUE
27400	
27500	Such an output expression attempts to prompt the dialogue towards the
27600	relation between the interviewer and the model which will be described
27700	later in  00.
27800	    If no delusional reference at all is detected by this procedure
27900	, the algorithm attempts the next function which searches for certain
28000	types of references to the 'self'.
28100	
28200	
28300	.F
28400	SELFREFERENCE
28500	
28600	     Since the main concern of a psychiatric interview consists of the
28700	beliefs, feelings and actions of the patient, the model must be able
28800	to answer a large number of questions about the 'Self'.  It is
28900	characteristic of a psychiatric interview that questions may not
29000	syntactically be questions but in the form of interrogative
29100	imperatives:
29200	
29300	.V
29400	    (10) Dr. - "Tell me more about the hospital"
29500	.END CONTINUE
29600	
29700	or statements indicating the interviewer has a question:
29800	
29900	.V
30000	     (11) Dr.- "I would like to ask you about your family."
30100	.END
30200	
30300	
30400	     If the input is recognized as a question and no topic is currently
30500	under discussion and the question refers to the 'Self', then it is
30600	assumed temporarily that it will refer only to a main self-topic.  These
30700	main self-topics (age, sex, marriage, health, family, occupation,
30800	hospital stay,etc.) in turn have sub-topics to varying depths.  For
30900	example, suppose the interviewer asks:
31000	
31100	.V
31200	     (12) Dr.- "How do you like the hospital."
31300	.END CONTINUE
31400	
31500	Since "hospital" is a main `introductory' topic with several levels
31600	of sub-topics, the algorithm answers the question with
31700	
31800	.V
31900	     (11)Model - "I shouldn't have come here."
32000	.END CONTINUE
32100	
32200	and then anticipates a variety of likely questions such as "what
32300	brought you to the hospital?", "how long have you been in the
32400	hospital?", "how do you get along with the other patients?", etc.
32500	Each of these questions bring up further topics, some of which
32600	represent a continuation of the main topic "hospital", but others
32700	of which represent a shift to another main introductory topic, e.g.
32800	"other patients".  Since many of the inputs of the interviewer consist
32900	of ellipses or fragments, the algorithm assumes them to refer to the
33000	topic or subtopic under discussion.  If some topic is being discussed,
33100	the algorithm checks first for a new main topic, then for a follow-up
33200	to the last subtopic, then (unless the subtopic is itself a main topic,
33300	
33400	as for example "other patients" in the above) for a follow-up to the
33500	last main topic.  Thus a continuity and coherence to the dialogue is
33600	maintained.
33700	
33800	     If some meaning cannot be abstracted from the question but it is
33900	recognized that a question is being asked, a function is called which
34000	attempts to handle certain common miscellaneous questions which are
34100	difficult to categorize.  These include the space-time orientation
34200	questions ("what day is this?") and arithmetic tasks ("subtract seven 
34300	from one hundred and seven from that number and so on") typical of
34400	current mental-status examinations.  Quantitative "how" questions
34500	("how many", "how often", "how long") are here recognized but one of
34600	the weaknesses of the model consists of its general inability to reply
34700	to them satisfactorily because the relevant information is lacking in
34800	the data-base.  If absolutely no clues are recognized in the question,
34900	the algorithm is forced to output a noncomittal reply such as:
35000	
35100	     (12) Model - "Well, I don't know."
35200	
35300	     This function also checks for statements about the self which are
35400	taken to be insulting or complimentary. Naturally the presence of a
35500	negator in the input reverses the meaning.  Thus
35600	
35700	     (13) Dr.- "You don't seem very alert."
35800	
35900	is classified as an insult whereas
36000	
36100	     (14) Dr. - "You are right."
36200	
36300	is considered complimentary and benevolent.
36400	
36500	     Among the so-called introductory topics are those which constitute
36600	sensitive areas, e.g. sex, religion and family.  If the interviewer
36700	refers to one of these areas, the value of ANGER increases sharply and a
36800	response is selected from one of the lists categorized as 'hostile',
36900	'defensive', 'personal' or 'guarded', depending on the level of MISTRUST
37000	at the moment.  For example, if the interviewer asks a question about
37100	the model's sex life, it first replies with
37200	
37300	   (13) Model - "My sex life is my own business."
37400	
37500	If the interviewer persists or even later tries to ask about sex, the
37600	model will respond with a hostile reply, such as:
37700	
37800	     (14)Model - "Do you know what you are doing?"
37900	
38000	The particular sensitive areas in the model are part of the initial
38100	conditions specific for this hypothetical patient.  Of course, these
38200	topics are commonly found to be sensitive areas in human patients.
38300	
38400	     The model operates sequentially trying one major function after
38500	another.  If it has come this far, after trying  SPECIALREACTION
38600	DELUSIONALREFENENCE and SELFREFERENCE without recognizing anything
38700	in the input pertinent to these functions, it proceeds to the next,
38800	FLAREREFERENCE.
38900	
39000	.F
39100	FLAREREFERENCE
39200	
39300	The data-base contains a directed graph of concepts involved  in  the
39400	model's  'stories'.   The  model  has  small  stories  to  tell about
39500	horseracing, gambling, bookies, etc.  The  major  concepts  of  these
39600	stories are termed "flare" concepts since they activate stories which
39700	are differentially weighted in the graph.  The graph can be  pictured
39800	as in Fig. ( )
39900	.V
40000	
40100	    Horses → Horseracing → Bookies →Gangsters → Rackets  →  Mafia
40200	                               ↑       ↑
40300	                           Gambling  Police
40400	                ↑                    ↑↑
40500	               Money              Italians
40600	
40700	
40800	In the  strong version of the model, the concept 'Mafia' is given the
40900	highest weight while in the weak version the concept 'Rackets' is most
41000	heavily weighted.  In both versions 'horses' has the lowest weight.
41100	
41200	The weights are assigned to the concepts and not individual
41300	words or word-groups denoting the concepts.
41400	.END
41500	
41600	The graph is directed in the  sense  that  reference  to  horseracing
41700	elicits a story about horseracing. When it is ended a prompt is given
41800	to the interviewer to discuss the next story in the  graph  involving
41900	'bookies'.   The  model strives to tell its stories under appropriate
42000	conditions and  leads  the  interviewer  along  paths  of  increasing
42100	delusional  relevance.   Much  depends  on  whether  the  interviewer
42200	follows these leads "benevolently" and reacts to the prompts.
42300	
42400	     The first step in this procedure is to scan the input for a flare
42500	concept having the highest weight.  Thus if a flare concept is already
42600	under discussion, a weaker new flare will be disregarded.  If the flare
42700	concept is one in a story which has already been told, then a prompt is
42800	offered regarding the next story-node in the graph.
42900	
43000	     If a question is asked about the events of a story, the model tries
43100	to answer it.  Also the model is sensitive to whether the interviewer
43200	is showing interest in the story or whether he tries to change the
43300	subject or (worse) expresses a negative attitude, such as disbelief.
43400	
43500	     If the interviewer indicates a positive attitude towards the
43600	story, then benevolence is recognized (see p   ) and the variables of
43700	ANGER, FEAR and MISTRUST fall slightly after each I-O pair.  ANGER
43800	falls more rapidly than FEAR while MISTRUST, being a more stable
43900	variable once it has risen, falls least.
44000	
44100	     If no flare concepts appear in the input, the model next tries to detect
44200	if a reference is being made to the relation between the interviewer
44300	and the model.  In an interview interaction there exists two situations,
44400	one being talked about and one the participants are in at the moment.
44500	Sometimes the latter situation becomes the former, that is, the one
44600	talked about.
44700	
44800	.F
44900	INTERVIEWRELATION
45000		As described in %00(chapter on language analyzer) the
45100	algorithm must be ready to handle inputnreferring to the relation
45200	between interviewer and model. The simplest cases are exemplified
45300	by expressions such as:
45400		(15) Dr.- "i understand you."
45500		(16) Dr.- "You do not trust me."
45600	Those phrases in an expression which can appear between "I" and
45700	"you" or between "you" and "me" we classified as representing a
45800	positive or negative attitude on the part of the interviewer. Thus
45900	expression (15) is taken to be positive whereas (16) is negative
46000	because although it contains a positive verb the verb is negated.
46100		The algorithm must distinguish between one-verb and two-
46200	verb expressions with certain common verbs, for example:
46300		(16) Dr.-" I believe you."
46400		(17) Dr.- "I believe you are wrong".
46500	In (16) a positive attitude is expressed whereas (17) is negative.
46600	Of course the language analyzer makes the correct identifications
46700	of actor and object in these types of expressions.
46800		If a positive attitude is expressd by the interviewer,
46900	FEAR and ANGER decrease. FEAR and ANGER increase depending on
47000	the conceptualizations of the input. These attitudes of the interviewer
47100	are stored as beliefs in the model being built up about the interviewer.
47200	Later the model can consult these beliefs in formulating questions
47300	and statements to the interviewer.
47400		Associated in the data base with each type of attitude expression
47500	expected are lists of appropriate output expressions. Thus in reply to:
47600		(18) Dr.-"I understand you."
47700	the model would reply:
47800		(19) Model- I'm glad you do."
47900	or
48000		(20) Model- "I appreciate your trying to understand."
48100	or some equivalent expression depending on values of the affect
48200	variables. When ANGER and FEAR are high, positive attitude expressions
48300	are interpreted as insincerity and hence evoke hostile replies.
48400		The remainder of input expressions not thus far discussed
48500	are handled by the procedure NORMAL.
48600	NORMAL
48700	
48800		This    procedure    deals   with   all   those   interviewer
48900	expressionsfrom which no conceptualization can be  formed.  The  only
49000	thing  which  can  be determined is perhaps the syntactical nature of
49100	the input. Presented with  one  of  these  expressions,  if  FEAR  is
49200	extremely  high the model signs off without a farewell expression and
49300	cannot be contacted through further natural language input.  If  FEAR
49400	is  high  but not extreme, and the input is recognized as a question,
49500	the model chooses a reply from a list which brings up the attitude of
49600	the interviewer as in:
49700		(21) Model- "Why do you want to know?"
49800	or
49900		(22) Model- "You pry too much".
50000	If the input is recognized as a statement, a reply is chosen from a list
50100	which indicates soem degree of anxiety:
50200		(23) Model- "Who are you really?"
50300		(24) Model- "You are making me nervous."
50400	If ANGER is high and the input is a question, a reply is chosen from
50500	a list designed to express hostility as in:
50600		(25) Model- "Do you know what you are doing?"
50700		(26) Model- "Perhaps you are just posing as a doctor."
50800		Sometimes in these default conditions the flag set in
50900	DELUSIONALREFERENCE allows the model to continue bt giving the next
51000	line in its delusional story. If the story is under discussion, continuity
51100	is maintained. But if it is not, the model appears to ignore the input
51200	and jumps back to oen of its previous preoccupations. In this instance
51300	the property of rigidity is a function of linguistic non-comprehension
51400	and not of the paranoid processes per se.
51500		If a story flag has not been st by a previous discussion
51600	in the interview and ANGER and FEAR are not high, the algorithm
51700	tries to see if the input is some type of general prompt from
51800	the interviewer such as:
51900		(27) Dr.- "Go on."
52000	or
52100		(28) Dr.- "Tell me more."
52200	If so, the model continues with its current story or attempts to
52300	initiate another story. (Discuss this in analyzer section?)
52400		If none of these conditions hold, the procedure ANSWER
52500	is called. This procedure handles a group of common special-case
52600	miscellaneous questions such as:
52700		(29) Dr.- "How do you do?"
52800	and miscellaneous statements such as:
52900		(30) Dr.- "Hi."
53000		(31) Dr.- "Good evening."
53100	
53200	.F
53300	SELFSCAN
53400		The final procedure in the algorithm scans what the model 
53500	has chosen to output. That is, it treats its own output as input.
53600	If this expression contains a flare or delusional reference, the
53700	appropriate flags are set and FEAR is raise slightly, but not as
53800	much as if this expession came from the interviewer. In this way
53900	the model "frightens itself" by what it says about a frightening topic.
54000	
54100			SUMMARY