Approaches to modelling age of acquisition

1
Recognising written words witha split
fovea Andy Ellis University of York
2
DOG

dog
ltdoggt
DOG
3
CAT

cat
?
ltcatgt
CAT
4
Right Visual Field Advantage
CAT
DOG
lt
5
Question What happens when a word is presented
in a central position, crossing the midline?
6
Split fovea theory (Shillcock, Monaghan,Brysbaert
)
STIMULUS
ltstimulusgt
STIM-
-ULUS
7
Test 1 Based on interaction between word
length and visual field.
8
VHF x Length interaction
Ellis, Young, Anderson (1988) Lexical decision
9
VHF x Length interaction
Ellis, Young, Anderson (1988) Semantic decision
10
Lavidor, Ellis, Shillcock Bland (2001
Cognitive Brain Research)
For a centrally-presented word, variation in the
number of letters to the left of fixation (LVF)
should affect word recognition more than
variation in the number of letters to the right
of fixation (RVF).
11
Lavidor, Ellis, Shillcock Bland (2001
Cognitive Brain Research)
NAVEL MINSTREL
For a centrally-presented word, variation in the
number of letters to the left of fixation (LVF)
should affect word recognition more than
variation in the number of letters to the right
of fixation (RVF).
12
Lavidor, Ellis, Shillcock Bland (2001
Cognitive Brain Research)
EXCEL EXORCISE
For a centrally-presented word, variation in the
number of letters to the left of fixation (LVF)
should affect word recognition more than
variation in the number of letters to the right
of fixation (RVF).
13
Lavidor, Ellis, Shillcock Bland (2001
Cognitive Brain Research)
Lexical decision RT affected by thenumber of
letters to the left of fixation,not by the
number of letter to the right.
14
Test 2 Based on interaction between cAsE
aLtErNaTiOn and visual field.
15
VHF x cAsE interaction
Lavidor, Ellis Pansky (2002) Lexical decision
Also Fiset Arguin (1999), Lavidor Ellis (2001)
16
Ellis, Brooks, Lavidor (2005, Neuropsychologia)
For a centrally-presented word, case alternation
in letters to the right of fixation should affect
word recognition more than case alternation in
letters to the left of fixation.
17
Ellis, Brooks, Lavidor (2005, Neuropsychologia)
cAlEnDaR
Expt 1. Lexical decision in LVF / RVF 8-letter
words and nonwords presented in LVF or
RVF. lower case MiXeD cAsE
18
Ellis, Brooks, Lavidor (2005, Neuropsychologia)
Expt 1. Lexical decision in LVF / RVF
19
Ellis, Brooks, Lavidor (2005, Neuropsychologia)
caleNdAr
Expt 2. Lexical decision at fixation 8-letter
words / nonwords presented in lower case
left alternated right alternated
20
Ellis, Brooks, Lavidor (2005, Neuropsychologia)
calendar
caleNdAr
cAlEndar
Expt 2. Lexical decision at fixation Prediction
Relative to the lower case condition, case
alternation should be moredisruptive to the
right of fixationthan to the left.
21
Ellis, Brooks, Lavidor (2005, Neuropsychologia)
calendar
caleNdAr
cAlEndar
Expt 2. Lexical decision at fixation
22
STIMULUS
Fast, parallel Unaffected by length Unaffected by
N Requires standard format
Slower, less parallel Affected by length Affected
by N Less affected by format
ltstimulusgt
STIM-
-ULUS
23
Summary
Characteristics of early processing in the left
and righthemispheres apply to those parts of
centrally-fixated wordsthat fall to the right or
left of fixation. Suggests brain-based
constraints on models of early processing in
visual word recognition. Those constraints will
need to be incorporated into models of word
recognition.
24
Summary
Characteristics of early processing in the left
and righthemispheres apply to those parts of
centrally-fixated wordsthat fall to the right or
left of fixation. Suggests brain-based
constraints on models of early processing in
visual word recognition. Those constraints will
need to be incorporated into models of word
recognition. Next step Understanding the
neural basis of hemispheric differences in
visual word recognition. Intra- or
inter-hemispheric?
25
(No Transcript)