week 10 - language Flashcards
semantic memory and word meaning
“CAT”
APPROACH 1: definitions
a carnivrous mammal long domesticated and kept by humans as a pet or for catching rats
BUT: what does mammal, pet or rat mean?
APPROACH 2: its one of those (a picture)
BUT: not everything can be pointed at (eg. trust, democracy) and pointing cant relate cat to other things (breathes, has hair)
approaches to meaning:
semantic networks
network of unitary nodes (no internal structure) and labelled links between them (the nodes)
good for heirarchies
properties are inheritied (eg. everything below animal breathes)
approaches to meaning:
semantic networks
what does it predict?
sentence verification times
more links to cross = more time
approaches to meaning:
semantic networks
diagram
see notes
approaches to meaning:
semantic networks
evidence?
sentence verification data
evidence for semantic networks:
sentence verification data
task: present sentence, participants must respond true / false
- a robin is a robin (fastest - no links)
- a robin is a bird (1 link)
- a robin is an animal (2 links)
- a robin has wings (at bird level)
- a robin has lungs (lungs at animal level)
slower if you have to cross more information
BUT:
a cow is a mammal slower than a cow is an animal
- because of familiarity
a robin is a bird faster than a penguin is a bird
- typicality
semantic memory organised on …
… basis relatedness or semantic distance
(instead of hierarchy)
problems with semantic networks
definition problem
influence of world knowledge
problems with semantic networks
defintion problem
are definitions really as sharp as a network implies?
necessary and sufficient conditions are the usual aproach to sharp definitions
- well defined set of attributes
- if all of them are present we have X
eg. bachelor
- never married
- male
problems with semantic networks
definition problem
problem with necessary and sufficient conditions:
necessary and sufficient conditions can rarely be found for an adequate definition
- so it cant define something in a way that captures the true meaning of what we think we understand
problems with semantic networks
definition problem
problem with necessary and sufficient conditions:
example
define “game” using necessary and sufficient conditions
the prototypical example was stated by Wittgestein
—> “we see a complicated network of similarities overlapping and cross-crossing”
so clear that it really hard to define
problems with semantic networks
definition problem
wittgensetins example shows the 2 problems
problem for learning
problem for meaning
problems with semantic networks
definition problem
problem for learning
how do children figure out what a game is if there is no definition, but the set is also not arbitrary
ie not everything is a game
—> how do children figure this out?
problems with semantic networks
definition problem
problem for meaning
how do i know what you mean by “game”
what if you pick one way and i pick a different way
problems with semantic networks
definition problem
concepts are…
fuzzy!
problems with semantic networks
influence of world knowledge
affects?
influence of world knowledge stored in semantic memory
can affect sentence comprehension
problems with semantic networks
influence of world knowledge
word knowledge and word meaning at same time?
integrate word meaning and word knowledge at same time
problems with semantic networks
influence of word knowledge
world knowledge and word meaning at same time?
evidence
2004
(world knowledge - in netherlands the train will always be yellow)
in the experiment would read these sentences
1) dutch trains are yellow and very crowded (true)
2) dutch trains are sour and very crowede (false - meaning of sour)
3) dutch trains are white and very crowded (false - world knowledge)
do people detect a mismatch in 2 as fast as 3
(EEG measurements reading N400 amplitude)
people mismatch just as fast
