Operating in a Mixedup World

1
Operating in a Mixed-up World

• Contextual factors and their Impact on Bilingual
  Language Production

2
The Language Hierarchy
Conceptual
Meaning
Lemmas
Abstract rep
dog
Lexemes
Sound patterns
/dawg/
Phonology
Sound information
Orthography
Visual information
O
3
Conceptual Level
Allows the bilingual to perform in one language
without interference from another.
Phonology
Two Monolinguals in One
Orthography
4
Conceptual Level
Bilingual Language Comprehension
Language Nonspecific Access
Evidence for the influence of the nontarget
languageslowing of lexical decisions.
L1 Lemmas
house
peln
book
gato
L1 Lexemes
The BIA model of recognition (Dijkstra Van
Heuven, 1998).
Phonology
Influence of Contextdegree of fluency, mix of
input, language of surrounding, previous use.
Orthography
5
Bilingual Language Production?
Language Specific Selection
Conceptual Level Express leg
L1 leg
L2 pierna
/laeg/
/pierna/
6
Examining Nontarget Language Processing
Identically Related Picture-Word Pairs
Same-Language
Identity Asymmetry
(leche)
(late)
7
Importance of Context Grosjean
Language Mode Hypothesis
Monolingual
Bilingual
Monolingual Modebilingual chooses a particular
language to use (base) and inhibits the other
language (guest).
Bilingual Modea bilingual chooses to keep both
languages active.
8
Importance of Context Grosjean
Language Mode Hypothesis
Monolingual
Bilingual
Response Language
Stimulus Language
9
Variables Predictions
Contextual Continuum
Blocked/Blocked
Conceptual Level Express leg
L1
L2
Significant identity asymmetry.
L1 leg
L2 pierna
Significant cross language identity facilitation?
No through translation facilitation.
/laeg/
/pierna/
10
Variables Predictions
Contextual Continuum
Conceptual Level Express leg
Blocked/Mixed
L1
L2
Significant identity asymmetry.
L1 leg
L2 pierna
Significant cross language identity facilitation.
Through translation facilitation?
/laeg/
/pierna/
11
Variables Predictions
Contextual Continuum
Conceptual Level Express leg
Mixed/Blocked
L1
L2
Reduced identity asymmetry.
L1 leg
L2 pierna
Greater cross language identity facilitation.
Through translation facilitation.
/laeg/
/pierna/
12
Variables Predictions
Contextual Continuum
Conceptual Level Express leg
L1
L2
Mixed/Mixed
Reduced identity asymmetry.
L1 leg
L2 pierna
Strong cross language identity facilitation.
Through translation facilitation.
/laeg/
/pierna/
13
Task
Speak English

2000 ms
1000 ms
14
Experiment 1 Identically Related Pairs
English/English
Spanish/English
Tree
English/Spanish
Spanish/Spanish
15
Identity Exp Contextual Continum
BB
885 ms
BM
939 ms
MB
1111 ms
MM
1167 ms
16
Identity Exp Context Effects
Facilitation (ms)
Blocked/Blocked
Identity
201
EE
Unrelated
(same)
18
ES
(cross)
RT (in ms)
65
SE
(cross)
SS
219
(same)
17
Identity Exp Context Effects
Facilitation (ms)
Blocked/Mixed
Identity
163
EE
Unrelated
(same)
38
ES
(cross)
RT (in ms)
108
SE
(cross)
SS
208
(same)
18
Identity Exp Context Effects
Facilitation (ms)
Mixed/Blocked
Identity
184
EE
Unrelated
(same)
111
ES
(cross)
RT (in ms)
96
SE
(cross)
SS
181
(same)
19
Identity Exp Context Effects
Facilitation (ms)
Mixed/Mixed
Identity
191
EE
Unrelated
(same)
90
ES
(cross)
RT (in ms)
99
SE
(cross)
SS
194
(same)
20
Identity Exp Context Effects, Summary
3. Significant identity asymmetry was found in
blocked response contexts, but weakened when
both languages were used.
21
Experiment 2 Through Translation Related Pairs
English/English
Spanish/English
sueño
Swan
English/Spanish
Spanish/Spanish
22
Thru Trans Exp Contextual Continum
BB
925 ms
BM
1013 ms
MB
1161 ms
MM
1273 ms
23
Thru Trans Exp Context Effects
Facilitation (ms)
Blocked/Blocked
Thru Trans
7
EE
Unrelated
(same)
15
ES
(cross)
RT (in ms)
21
SE
(cross)
SS
8
(same)
24
Thru Trans Exp Context Effects
Facilitation (ms)
Blocked/Mixed
Thru Trans
-9
EE
Unrelated
(same)
18
ES
(cross)
RT (in ms)
77
SE
(cross)
SS
97
(same)
25
Thru Trans Exp Context Effects
Facilitation (ms)
Mixed/Blocked
Thru Trans
72
EE
Unrelated
(same)
79
ES
(cross)
RT (in ms)
121
SE
(cross)
SS
65
(same)
26
Thru Trans Exp Context Effects
Facilitation (ms)
Mixed/Mixed
Thru Trans
118
EE
Unrelated
(same)
87
ES
(cross)
RT (in ms)
97
SE
(cross)
SS
152
(same)
27
Thru Trans Exp Context Effects, Summary
1. TT Words do not help with naming pictures
when only one language is used and seen.
2. TT Words help some if both languages are seen
and help more if both languages are used.
28
Conclusions Contextual Continuum
Reflection of a bilingual contextual continuum in
which production times are sensitive to the
processing environment.
Language being used in the environment as well as
visual language cues in the environment can
impact the degree of interaction.
29
Conclusions Other Factors that could Impact
Interaction?
Learning Environmentunconstrained bilinguals
will activate both languages more than
constrained bilingual.
Grammatical knowledgeif incomplete in one
language, the more complete language will stay
active.
Time in a monolingual environmentthe longer
bilinguals are in a monolingual setting, the
less likely they are to activate both languages.
Stress or difficultywhen confronted with a
difficult or stressful setting, more likely to
activate both languages (fewer control resources).
Working Memorythose with stronger working memory
may exhibit more control over the nontarget
language.
30
Conclusions Implications for Classroom?
Using materials with L1 words makes it more
likely that L1 will stay active during the
learning process.
Speaking L1 in the learning environment makes it
more likely that L1 will stay active during the
learning process.
Students who are more anxious or less willing to
use L2 are more likely to have interference from
L1.
L2 acquisition in classroom less successful than
immersion programs in which L1 is less likely to
be active.