R. Port, Q500, Intro to Cognitive Science
It is the interdisciplinary study of human cognitive performance emphasizing the combined resources of experimental observation of behavior, computational and mathematical modelling, neuroscientific investigation and philosophical enquiry.
B. What is a mind? And what is a soul? The main difference seems to be that minds are generally assumed to be mortal like the body whereas souls are generally assumed to survive the body (and perhaps be immortal). Otherwise, both seem to be generally assumed to be nonphysical, unitary `beings' that embody our individuality, feelings, intellect, skills, etc. The notion seems commonplace, but, when you think about it, it is quite spooky.
Some notions associated with the mind:
SELF
Consciousness
Selfconsciousness
Subjectivity
SKILLS - INTENTIONALITY - INPUT/OUTPUT TO THE WORLD
Perception
Action (that is, purposive motor behavior)
Language
Thinking
Learning and memory
FEELINGS (affective properties)
Emotions
Personal Experience, Phenomenology
Moral-ethical weight
C. The Mind-Body Problem: ``How can mental phenomena in general, and intentional phenomena in particular (but also subjective and affective phenomena) arise in, and be compatible with, the evidently material world described by science?'' (quote from Brian Cantwell Smith)
Or:
D. Descartes' Dualism
Descartes formulated the modern view of human beings. He proposed that every human has two components:
1. A Mind (or Soul)- RES COGITANS. It thinks, experiences, reasons, perceives and controls actions, but takes up no space, has no weight (may be indestructible) and is possessed only by human beings, not other animals.
2. A Body - RES EXTENSIS. It takes up space, has mass and obeys the laws of physics.
Descartes proposed that the mind is `essential' to our being while the body is contingent (``I think, therefore I am''). And he proposed that mind-body interaction has a locus in the pineal gland (the only centerline organ as far as Descartes knew.)
This general view seems completely endemic in the civilized world. It seems difficult to find anyone (certainly not any Catholic, Protestant, Jew, Muslim, Hindu or Buddhist) who does not believe roughly this (though there are some `wicked atheists' who would deny it).
In the 20th century the development of mathematics and computers has led to a refinement of the notion of mind: Perhaps the mental is basically like mathematics, and lies in Plato's Heaven. Perhaps the mind is really a kind of formal system.
E. FORMAL SYSTEMS
Formal systems all involve:
1. TOKEN MANIPULATION
a. set of discrete FORMAL TYPES: pieces, positions, etc
- frequently in two kinds:
cells & pieces (checkers, go, computer bits)
cells & colors (cell'r automata)
positions & digits (decimal numbers)
positions & bits (digital computer)
b. one or more START POSITIONS
eg, calculator: 000.00
grammar: S (= sentence)
checkers: starting layout
cell'r aut'mton: initial configuration
c. list of explicit RULES about token arrangements
eg, move a token, replace, multiply, concatenate, etc.
2. DIGITALITY (`positive' = perfectly reliable)
a. positive technique for PRODUCING tokens (`writing')
b. positive technique for IDENTIFYING things (`reading')
Thus, the assumption is that the player can always determine what the current state of the system is without fail.
3. MEDIUM INDEPENDENCE. The physical form is not relevant (as long
as all tokens are kept distinct).
[ 4.DISCRETE TIME. Formal systems are always run in discrete time,]
[ one time-step per rule executed. ]
Rule execution is instantaneous - that is, they take zero time.Thus the system is always in a state BETWEEN the application of Rule N and Rule N+1. What happens DURING application of the rule is not accessible. Note that this property is overlooked by Haugeland.
Examples of formal systems: formal logic; arithmetic; abstract algebra; formal description of vending machine (though not the vending machine itself); a wordprocessing program; C++ programming language; any program in C++; your PC (when already plugged in), etc.
Examples of non-formal systems: pool and snooker (no discrete positions for balls, no discrete moves); game of basketball (shots through hoop may be digital since either in or out, but player positions and actions are not); weather; American history, etc.
Plato's Heaven. It seems like genuine formal systems exist only in `Plato's Heaven', the realm where mathematical structures exist and mere physical events cannot impact them.
Haugeland says that `automatic formal systems' are ``physical devices'' that have parts or states that can be interpreted as formal tokens and that automatically manipulate the tokens in accord with the rules of some (real) formal system. The best example, is, of course, a modern digital computer or calculator.
The problem for me (RFP and apparently some students in the class judging from the questions raised) is that it seems like no physical system could ever REALLY be a formal system - since anything physical will always be fallible and run in continuous time and therefore is nondigital and nonformal. So the phrase seems like an oxymoron - a self-contradiction (just like the notion of ``physical symbol system'' does). Of course, this doesn't mean that one couldn't implement devices that APPROXIMATE automatic formal systems pretty closely (like your PC and calculator do).
G. INTERPRETED FORMAL SYSTEM
Formal systems often have no semantics, that is, the tokens are just distinct from each other. They have no real `meaning' or intentional properties - eg, chess, checkers, cellular automata (usually). But sometimes the symbols mean something about (for example) the world-- eg, predicate logic, arithmetic (where numbers are not merely tokens but refer to meaningful quantities), a word-processing program, etc. Naturally interpreted formal systems are the most important and useful type. For interpreted formal systems we can ask whether the system `did the right thing or the wrong thing', whether it `makes sense', whether it is `useful', etc. Noninterpreted formal systems only generate meaningless patterns (even if it may matter to us which player wins the chess game).
And naturally, if human cognition is a formal system (or is usefully modelled by a formal system), then obviously it must be an interpreted formal system, since it clearly is intentional.