L2 Language, Visual World Recognition 2

Question 1

What did James Cattell find in 1886?

Answer 1

Frequency of usage of words in a language influences tasks involving printed words
Common words are recognised fastest

Question 2

What did Monsell find in 1989?

Answer 2

Investigated word frequency effect in lexical decisions
High, Medium and Low frequency words
Much longer reaction time for low frequency words

Question 3

What is the best predictor when explaining lexical decision making ?

Answer 3

Spoken word frequency

Question 4

What is some research into language exposure?

Answer 4

Men and women speak about 16,000 words a day (Mehl 2007)
Li 2014 - 60 year old: 1.64 billion words

Question 5

What did Adelman 2006 find about contextual diversity ?

Answer 5

Lexical decision times are dependant on how many contexts a word has been seen
Jones 2012 suggests that semantic diversity is more important that CD because it ignored information redundancy

Question 6

What did Jones find about semantic diversity (2012)?

Answer 6

Reaction times and naming latencies obtained from the English Lexicon Project
Semantic diversity is more important

Question 7

What is print exposure in adults ?

Answer 7

Chateau and Jared 2000 found impact of the exposure to print on word recognition
Exposure to print was measured using the author recognition task
Massive difference in how fast they respond to words

Question 8

What did Mol and Bus (2011) find about print exposure in children ?

Answer 8

Meta-analysis of studies investigating the impact of print exposure from infancy to early adulthood
Readers aged 3-5 associated oral language with print exposure
Reading improves academic success and oral growth

Question 9

What did Arnon and Snyder (2010) find about frequency effect beyond single words ?

Answer 9

Used 4 word sequences
High frequency phrases are faster than low frequency phrases
Also exists in combinations of words

Question 10

What are multiword sequences ?

Answer 10

Siyanov-Chanturia 2011 focused on binominal expressions e.g. bride and groom
Read sentences with the binominals in reverse (groom and bride)
Readers - native and non-native speakers - looked at the reverse for longer, native speakers took longer than non native

Question 11

What is the IA model for word frequency ?

Answer 11

Created by McClelland and Rumelhart, 1981
Resting level activation of words varies between -0.046 and 0
Varying RLA of word nodes to stimulate word frequencies effects is also used in other models

Question 12

What is lexical similarity ?

Answer 12

Coltheart 1977
Orthographic neighbours - number of words created by changing one letter of a word
Andrews 1992 found a significant interaction between frequency and neighbourhood density
Slower reaction for high frequency words, faster reaction for low frequency words

Question 13

What are orthographic neighbours ?

Answer 13

Words that only differ by one letter e.g. Sleet, Fleet, Sleep
When there are more neighbours it takes a longer period of time to reach activation

Question 14

What is the multiple read out model ?

Answer 14

IA model with decision criteria
Predicts inhibitory effects of neighbours due to competition between word nodes (lateral inhibition)
Grainger and Jacobs 1996

Question 15

What s Levenshtein distance ?

Answer 15

Minimum number of edits between two words
This predicted naming and lexical decision latencies better than orthographic neighbour size
Facilitatory

Question 16

What is the decompositional theory of words ?

Answer 16

Word meanings best described in terms of sets of bivalent semantic features or semantic markers
Defining features - features an object must have
Characteristic features - features possessed by most of the objects (not necessary)

Question 17

What are semantic features ?

Answer 17

Intercorrelated features - occur together, many members of a natural category will share intercorrelated features
Distinguishing features - enable to distinguish, exclusive to single items in a category, artefacts are represented by many distinguishing factors

Question 18

What is the semantic network model ?

Answer 18

Collins and Quillian 1969
Concepts represented by nodes
Nodes joined together by links representing relations between concepts
Meaning is determined by node placement in the network
Hollan showed that semantic networks and feature theories were formally equivalent
Rips et al showed formal equivalence doesnt imply psychological equivalent

Question 19

What is prototype theory ?

Answer 19

Concepts are centred around a representation of a prototypical member of the class e.g. robin vs ostrich for bird
Boundaries can often be fuzzy
Explains why the correct classification of objects can be in doubt when features are not and why some concepts are more typical

Question 20

What is the sentence verification task ?

Answer 20

Ps asked to respond true or false to sentences
Set inclusion - a robin is a bird, a whale is a fruit
Property attribution - a robin has feathers, a whale has seeds
Use of simple sentences control different syntactic processing
Reaction times infer how meaning of words is stored and used

Question 21

What is the principle of cognitive economy ?

Answer 21

Information about concepts is stored at appropriate level in hierarchy e.g. has feathers is stored with bird, not robin
Sentence verification takes longer as the distance in the hierarchy increased
Found a clear linear pattern
Superordinate were quicker than properties

Question 22

What is a problem with cognitive economy ?

Answer 22

Typicality effect (Rips 1973) - robin and ostrich are both one level away from bird but robin is identified quicker

Question 23

What is the spreading activation model ?

Answer 23

Non - hierarchical
Semantic similarity (number of properties it shares)
Semantic priming - spreading activation explains why words prime related words in lexical verification
Between-category typicality - move from organisation e.g. a dog is an animal vs a dog is a mammal is not a problem

Question 24

What are distributional semantics ?

Answer 24

Words with similar meanings are used in similar context
Count models and Predict models

Question 25

What are count models ?

Answer 25

Using large corpora to obtain relatedness valued between pairs of words
LSA - count words in passages
HAL - co-occurrence of word pairs

Question 26

What are predict models ?

Answer 26

Predict target words based on context words, methods uses a neural network that learns from large corpora
Predict meaning

Question 27

What are the problems with semantic theories ?

Answer 27

Meaning = relations to other words, how do we know the meaning of the words ?
Grounding problem -what is the relation between abstract symbols.
Grounding conceptual knowledge in modality specific systems - use vision, sound etc to make the model more powerful