Psy 421: Systems and Theories of Psychology
Cognitive Psychology and Cognitive Science
-
Starting in the late 1950's and the 1960's experimental psychologists
started focusing on problems of perception, attention, memory, and language.
Dealing with these seemed to require a different set of concepts and principles
than those that were used for most of the first part of the century.
-
Late 19th and early 20th Century:
Symbolic Logic (Frege, Russell,
Wittgenstein)--There are formal principles leading to valid conclusions
that are far more powerful than those of Aristotle.
-
Wiener--the concept of cybernetics and the
use of information in causal or explanatory systems.
-
Turing--the idea of a generalized Turing machine.
A machine that can (in principle) solve any well-specified problem that
can be solved. Church-Turing thesis--One can find different formalisms
which can solve these problems, and they can all solve exactly the same
set of problems.
-
Bartlett--One uses schemata to understand, learn and
remember. These schemata play an important role in our understanding of
the world.
-
Chomsky--critique of Skinner. Formal limitations of
associative processes. Invented Generative grammars. Proposed a
language organ in the brain. Are there innate grammars?
-
Broadbent--One can build flow-charts to show the movement
and transduction of information as a person receives information in the
senses, stores it and modifies it so that it is available for interpretation
of new information and the production of responses. The processes designed
to deal with the limited capacity of our processing mechanisms are called
Attention. These are factors that help focus and select information
for further processing, analysis or consciousness.
-
Minsky, and Simon and Newell--one can write
computer programs which can solve logic problems and other reasoning situations.
These receive information, store it and transform it so that it is available
to help interpret new information and for the production of responses.
Moreover, one can write programs that simulate the thought processes of
human problem solvers, even by getting the problems wrong in the same way.
This is done by the use of heuristics, and not only algorithms.
-
Piaget--children go through stages where they use
different kinds of processes to solve problems. After the first stage,
which might resemble behavioristic ideas, they require the use of memory
and reversible processes as different representations are manipulated.
These go far beyond the mechanisms allowed by associationistic principles.
-
G. Miller--Argued successfully that language couldn't
be explained without complex rules of generation. People must use higher
level representations in order to remember different things. The magic
number 7 ± 2 and the idea of "chunks" of information were illustrative
of the manipulation of symbols for memory and to solve problems.
-
Vygotsky--Children, and people with different cultural
backgrounds (particularly, lack of literacy) may use different principles
of organization and different heuristics.
-
Rosch--categorization: following up some of the ideas
of Wittgenstein, Rosch showed that categories are not clearly identifiable.
Aristotle’s idea that a category can be identified by singularly necessary
and jointly sufficient conditions is normally not the case. Certain basic
categories are the first to be learned, and most other categories are
learned using these as a basis. These categories tend to have perceptually
similar members that are responded to in a similar fashion. Categorization
is an important basis for our knowledge and understanding of language and
the world.
-
Tversky and Kahneman--People generally do not reason
according to sound logical principles. They are influenced by the framing
of a problem, and they use certain heuristics (such as the "representative"
heuristic which often lead them to the wrong conclusion.
-
Rumelhart, McClelland, Slominsky and others following
McCulloch and Pitts neural network theory have been generating "sub-symbolic"
models of parallel distributed processes to account for certain
cognitive activity. These "PDP" models have become of interest to many
cognitive and neural scientists.
-
Some issues are--What is innate in our cognition? How broadly
applicable is the computation model (the computer metaphor? Do children
think according to the same principles as adults or are do they use a different
kind of reasoning than adults? Are people basically rational? If so, how
do you explain their many errors in reasoning? If not how do they use rational
processes at all? Do all mental systems operate according to the same or
similar principles or are there large difference between mental systems?
What is the role of consciousness? How can consciousness be accounted for
in a physiological system? What is the relationship between symbolic and
sub-symbolic theoretical systems?