Summon the AI4U Textbook Mind into your presence with MSIE.

The Conjoin Mind-Module of the AI Breakthrough
by Mentifex

1. Theory of AI4U Algorithm Steps: Code the Conjoin Mind-Module

   /^^^^^^^^^\   Thinking "...because I love you"  /^^^^^^^^^\
  / visual    \     | | |   ____________          / auditory  \
 /  memory     \    | | |  / Conjoined  \        /  memory     \
|   channel     |   | | | (  Sentence    )      |   channel     |
|   _______     |   | | |  \____________/-------|-------\       |
|  /image  \    |   | | |    |     |   \        |       |       |
| / percept \---|-----+ |    |     |    \____   |       |       |
| \ engram  /   |  a| | |    |     |    /Verb\  |       |       |
|  \_______/    |  b|c| |    |     |   (Phrase) |       |       |
|               |  s|o|f|    |     |   /\____/  |       |       |
|               |  t|n|i|    |    _V__/   |     |       |       |
|               |  r|c|b|    |   /Noun\   |     |       |       |
|               |  a|e|e|    |  (Phrase)  |     |    ___|___    |
|               |  c|p|r|   _V_  \____/   |     |   /       \   |
|               |  t|t|s|  /   \   |      |     |  /         \  |
|               |   | | | (Conj.)  |      |     |  |"because"|  |
|               |   | | |  \___/   |      |     |  |         |  |
|               |   | | |          |      |     |  |"I"      |  |
|   _______     |   | | |  ________V_   __V__   |  |         |  |
|  /fresh  \    |   |_|_| ( English  ) / En  \--|--|"love"   |  |
| / image   \   |  /     \ \ Nouns  / / Verbs \ |  |         |  |
| \ engram  /---|--\ Psi /--\      /  \_______/ |  \"you"    /  |
|  \_______/    |   \___/    \____/-------------|---\_______/   |
View the complete set of original, non-derivative brain-mind AI theory diagrams.

Just as the Joint Chiefs of Staff run the military side of the
American military-industrial complex (Eisenhower, 1961) likewise
it does not require military intelligence to draw the conclusion
that the Conjoin Module allows a vaunted stream of consciousness
to think one idea and link it to another idea while considering
not how but why, because the unexamined life is not worth living.
In other words, conjunctions are the not-so-missing link between
thoughts expressed in a sentence joined semantically with others.

2. Semantically Different but Logically Equivalent

The naive and sentimental student of logic (if you are a student
of AI, you are a student of logic) may be perplexed or surprised
to find that symbolic logic makes no fine distinction in meaning
between conjunctions that we humans consider loaded with meaning.

For instance consider the following hypothetical and/but possible
dialogue between a human being and a robot outfitted with a Mind.

Human: You are my friend but you will do what I say.
Robot: I am your slave and I must do what you say.

Logic does not dictate and logic does not care whether Earthlings
use "and" or "but" to link conceptual aggregates into a sentence.
The Vulcan ambassador regards "alien but nice" or "alien and nice"
to be essentially one and the same statement, while we humanoids
immediately assume certain unstated meanings when we hear someone
use the conjunction "but" instead of "and." We realize that the
mindset or attitude of the speaker determines the choice of "and"
or "but" in a sentence, but we don't worry overly much about the
issue and we are more concerned with getting our AI Mind software
to select any conjunction at all than with the conjoined nuances.

3. Why Code AI? Because It Is There

In the Conjoin Module of the Mind-1.1 source code, the detection
of the adverb "why" within a sentence of input triggers "because"
as the conjunction likely to be used in a statement of response.
We have no a priori proof that the primitive AI Mind will answer
a "why?" question with a logically satisfactory "because" clause.
We simply assume (and hope) that the associative processes within
the Mind will cause the AI to state the logically most compelling
explanation in a "because" response to the input question of "why."

As AI Minds advance or evolve into greater complexity and obvious
consciousness, we expect that the background associative thoughts
will be more on-target with "because" explanations and that the
conscious AI will become aware of the need to explain something.
Meanwhile we consider the merely mechanical linking of "why" with
"because" in the Mind-1.1 software to be a triumphant achievement
of a goal that we once thought lay untold decades in the future.
The suddenly early achievement of the "why-because" AI milestone
has been one trigger for the release of this generic AI textbook. is an overview of Mind.

4. JavaScript Seed AI source code with User Manual
// Conjoin() selects a hopefully appropriate conjunction and
// allows the AI to answer a "why" question with a "because"
// statement, under the assumption here that the thinking of
// the AI will tend to display a modicum of explanatory logic.
function Conjoin() {  // ATM 12aug2002; or your ID & date.
  if (question == 16) {    // If the question is "why" ...
    conj = 18;             // use the conjunction "because";
  } else conj = 17;        // otherwise use "and".
  for (i = t; i>midway; i--) {  // Look backwards for "conj".
    enLexicon[i].enExam();  // Inspect the English nodes.
      if (en0 == conj) {    // If the conjunction is found...
        motjuste = conj;    // "nen" concept for conjunction;
        aud = en5;  // Auditory recall-vector for conjunction.
        break;  // Only find one instance of the conjunction. 
      }  // End of search for conjunction.
    }  // End of loop finding the appropriate conjunction.
  Speech();      // Speak or display the chosen conjunction.
  question = 0;  // Reset "question" after any use.
}  // End of Conjoin(); return to the SVO() module.

5. Mind.Forth free artificial general intelligence with User Manual
\ Conjoin selects a hopefully appropriate conjunction and
\ allows the AI to answer a "why" question with a "because"
\ statement, under the assumption here that the thinking of
\ the AI will tend to display a modicum of explanatory logic.
:  Conjoin  \ ATM 16aug2002; or your ID & date.
  questype @  16 =  IF  \ If the question-type is "why" ...
    18     conj !       \ use the conjunction "because";
  ELSE  17 conj !       \ otherwise use "and".
  THEN                  \ "questype" from oldConcept.
  \ Code adapted "mutatis mutandis" from auxVerb "do" search:
  midway @  t @  DO  \ Look backwards for "conj".
    I       0 en{ @  conj @ = IF  \ If conjunction is found,
      conj @  motjuste !  \ "nen" concept for conjunction;
      I     5 en{ @  aud !  \ Recall-vector for conjunction.
      LEAVE  \ Only find one instance of the conjunction.
    THEN  \ End of search for conjunction.
  -1 +LOOP  \ End of loop finding the appropriate conjunction.
  SPEECH    \ Speak or display the chosen conjunction. 
  0 questype !  \ Reset "questype" after any use.
;  \ End of Conjoin; return to the SVO module. is the Mind.Forth User Manual. explains the Seed AI variables.

6. Analysis of the Modus Operandi

The primitive Conjoin Module is only a rough approximation of how
conjunctions really ought to be selected within a Robot AI Mind.
We start with a default selection of "and" as the quasi-universal
conjunction, and then we permit special cases to trigger a choice
of special conjunctions. If we had not resorted to such makeshift
methods, we might not have achieved the basic Mind-1.1 functionality.

After the AI breakthrough of 7 June 2006, on 14 September 2006 it
became clear that we could computationalize the use of conjunctions.
For the tutorial mode of the JavaScript Mind.html software,
we had developed code that would count how many direct objects
were logically valid for a particular combination of subject-noun
and transitive verb. It occurred to us that if the AI Mind had two
direct objects available, it could transform the set of two sentences

into a single sentence formed with the "and" conjunction.
We would no longer blindly be conjoining two sentences,
and instead the AI would computationally be selecting a
particular conjunction because the use of it made sense.
It occurred further to us that our software was probably
capable of detecting negational direct objects, so that
could be transformed into
or even Of course, AI programmers would still be hand-coding the
various algorithms for the selection of conjunctions, and
the AI Minds would not yet be learning the syntax and rules
of intricate logic for the proper use of conjunctions.
Programmers would have to hand-code the learning spiral
of adding or substracting syntactic units in sentence
structures that slowly mutate over time as child minds
try to communicate with both their peers and their elders.

7. Troubleshooting and Robotic Psychosurgery

Try in advance not to introduce any evolutionary bugs.

The AI Debugger program may shed some light in general on how to debug
and troubleshoot programs in artificial intelligence.

8. Conjoin Module Resources for Seed AI Germination and Evolution

The Conjoin mind-module is the subject of Chapter 13
in your POD (print-on-demand) AI4U textbook,
as worthily reviewed and intellectually evaluated
by Mr. Christopher Doyon of the on-line Turing Store; and
by Prof. Robert W. Jones of Emporia State University.
A search on eBay may reveal offerings of AI4U and
a catalog search for hardbound and paperback copies
may reveal libraries beyond the following where students
of artificial intelligence may borrow the AI4U textbook:
  • Hong Kong University Call Number: 006.3 M981
  • North Carolina State University (NCSU) Call Number: Q335 .M87 2002
  • Texas A&M University
    Consider as a way to
    circulate and track your wandering copy of AI4U.
    At your own library you may submit a request for
    the acquisition of AI4U with ISBN 0595654371.

    Return to top; or to sitemap; or to
    [For the above link to work, copy to your local hard disk
    and name it Mind.html in the C:\Windows\Desktop\ directory.]

    SourceForge Logo