Class Notes (1,100,000)
CA (620,000)
UTSG (50,000)
PSY (4,000)
PSY210H1 (100)
Lecture 11

PSY210H1 Lecture Notes - Lecture 11: Metacognition, Implicature, Metamemory

Course Code

This preview shows pages 1-3. to view the full 10 pages of the document.
Problems with computer model
o Brain is plastic, computer is not
o At the software level, computational model is inadequate
Relies highly on propositional knowledge
Propositions are statements about facts 2+2=4,
Australia is a continent
Mind primarily as a machine for making beliefs
But, most of our cognition is not about acquisition of
beliefs, but acquisition of skills knowing HOW to do
things procedural kind of knowing
Computers suck at this stuff all their knowledge is
Computers can’t track patterns and follow them with
actions imagine playing a sports game you’re trying
to figure out patterns, anticipate what’s going to
happen, follow through with an action computer can’t
do this! Complex pattern tracking (core of procedural
knowledge and not captured well with propositions or
inferences) a lot of our cognition is not like a
computer program
o Computer metaphor tends to think of our cognition as very
sequential, but as we learn from Piaget, a lot of cognition
occurs in a self-organizing manner computer model has
difficulties explains dynamical, self-organizing cognitive
o Logic and inference are largely about trying to deal with truth
this is ONE important aspect of cognition
Computers are good at this kind of thinking - reliable
about not dropping truth if you put truth into them
BUT, you are not organized just to find “true”
information the “true” information out there is infinite
what we care about is RELEVANCE we only care
about RELEVANT truth. How you assimilate and
accommodate = attention to relevant truths
Logic is not designed for handling relevance, relevance
is not a property of a proposition logic doesn’t capture
Computational metaphor has had a perennial problem
with trying to account for relevance
o Two huge features of cognition procedural knowing and
relevance are not captured by computational metaphor
ability to know how to do things, and how to selectively attend

Only pages 1-3 are available for preview. Some parts have been intentionally blurred.

to information these are two of the things that make us most
Alternative model to logic machine proposed in 2009
o Newer model is logos multimachine
o Logistics how brain allocates cognitive resources
o Information is organized not in a logical sequential fashioned,
but in a self-organized, dynamical fashion
o Primary example of this is the relationship between
assimilation and accommodation trying to get most efficient
yet realistic use of your cognitive schemes
o Brain isn’t a single machine moving sequentially through states
it is a machine of machines that can make itself into a new
kind of machine!
o Constantly integrating and differentiating in order to
complexify itself, making itself into a new machine that has
new competencies
o Hardware affords qualitative change in the software - Running
hardware improves software, running software improves
o Cognition/brain, brain is constantly pushing toward qualitative
o Computational model was NOT a developmental model of
o Logos multi-machine (LMM) is a development-driven model
o Developmental psychology is not some poor cousin of
cognitive psychology
o LMM model says cognition is inherently developmental
o A lot of information processing models are becoming out of
o More and more, neuroscience is looking at things the LMM way
you are much more a factor in your development than the
computational model indicates
o Maybe you’re nothing but a self-organizing system running on
a self-organizing brain?
o This model emphasizes individual differences
Information Processing vs. Piagetian model
Process model (how/when, analyze, formalize, mechanize) over a
product model (Piaget; what, describe products of development)
concerned with studying five aspects of change, tracking change,
following the flow of information: 1. path/sequence/process of
change, 2. rate of change, 3. breadth of change (when one process
changes, how much of that is shared with the rest of cognition), 4.

Only pages 1-3 are available for preview. Some parts have been intentionally blurred.

variability in change (how much difference is there between children
in their pattern of change), 5. sources of change (what are the
mechanisms/rules/procedures that are resulting in the change
Differences: Piaget is product driven (what), Information process is
process driven (how)
Emphasis on analysis, formalization, and mechanization in
computational model
Information processing is much more specific and much more
complete in investigation of phenomenon
However, there are three important similarities that make them both
branches of the cognitive-developmental approach
Both are individualistic in nature no talk of cultures, communities,
distributed cognition - Both more concerned with mind-world
relations than mind-mind relations
Heavy emphasis on cognitive development, especially problem solving
Both Piaget and information processing model are very influenced by
natural science trying to make cognition studies as much like other
sciences as possible (Piaget -> biology, Information Processing ->
cognitive science)
Studying development from Information Processing perspective:
Important processing limitations
o Encoding
o Selective attention
o Transmission of information
Information bottle-neck studying flow of information, what to
selectively attend to this doesn’t even occur to Piaget – but a lot of
information processing is trying to deal with this informational bottle-
Maturation, brains grows and working memory does seem to go up,
but not a primary strategy
Working memory capacity is constant, working memory best
described as a relevance filter
Improving performance
o How you’re selecting information
o Get better at how you encode
Familiarity is poor indicator of recollection ability
Always trying to find a way to better transmit
information for your future self
Look at development through information processing design issues
What changes can you make to optimize information processing given
limitations on the design?
You're Reading a Preview

Unlock to view full version