ArticlePDF Available

Cognitive perspectives of language switching: Factors of bilingualism and development

Authors:

Abstract and Figures

This study explores the use of inhibitory control during bilingual speech production. In particular, it investigates whether or not age of acquisition and proficiency of the second language affect whether or not bilinguals will rely on inhibitory control while speaking. English language learners of Spanish participated in a picture naming task in which they switched back and forth between their languages. The results suggest that proficiency, and not L2 age of acquisition, is a decisive factor which may determine whether inhibitory control or language-specific selection mechanisms underpin bilingual speech production. This supports a notion in which L2 proficiency leads to a shift away from inhibitory control. Theoretical implications and directions for future research are discussed.
Content may be subject to copyright.
Cognitive perspectives of language switching:
Factors of bilingualism and development
John W. Schwieter
Wilfrid Laurier University
408 Bricker Academic Building
75 University Avenue
Waterloo, Ontario
N2L 3C5
jschwieter@wlu.ca
Abstract
Abstract
This study explores the use of inhibitory control during bilingual speech production. In
particular, it investigates whether or not age of acquisition and proficiency of the second
language affect whether or not bilinguals will rely on inhibitory control while speaking.
English language learners of Spanish participated in a picture naming task in which they
switched back and forth between their languages. The results suggest that proficiency, and not
L2 age of acquisition, is a decisive factor which may determine whether inhibitory control or
language-specific selection mechanisms underpin bilingual speech production. This supports
a notion in which L2 proficiency leads to a shift away from inhibitory control. Theoretical
implications and directions for future research are discussed.
Keywords: bilingualism, speech production, language selection, inhibitory control.
1. Introduction
How is it that upon being presented with a picture of a couch, an English
monolingual knows to say ‘sofa’ or ‘couch’ when both are perfectly acceptable labels
for the object? This puzzling question regarding the competition between words in the
mental lexicon is of central interest to cognitive psychologists and psycholinguists.
These types of studies have traditionally focused on monolinguals; however, the last
decade has shown interest in speech production and lexical processing within
bilingualism. For monolinguals, it is rare that one concept can refer to two or more
words (e.g., couch-sofa). Indeed, most concepts only have one word to describe them.
27
However, when considering the bilingual case, most concepts will have two words
mapped on to them (i.e., due to translation equivalents). This, then, is a difficulty with
which bilinguals are constantly faced: they must resolve competition that exists between
the two possible candidates and select the correct language for production. Because
bilinguals are able to accurately speak in only one language and switch back and forth
between both languages automatically, there must be cognitive mechanisms that help
the bilingual mind restrict the lexicalization procedure to only one language.
In this paper, I explore issues of language control among bilinguals. Specifically,
I investigate whether or not certain factors of bilingualism (second language (L2) age of
acquisition, and L2 proficiency) affect the underlying cognitive mechanisms of
bilingual speech production. Green’s (1986, 1998) Inhibitory Control Model (ICM)
holds that the activation level of the non-relevant language is reduced at the lexical level
by mechanisms of suppression. Recently, it has been suggested that these underlying
mechanisms may be modulated by age of acquisition and/or proficiency in the L2
(Costa 2005, Costa & Santesteban 2004).
In the sections to follow, I first discuss the ICM and studies that have supported
its claims. Next, I introduce the present study and the results of the experimental
procedures. Finally, I discuss the theoretical implications of these results on models of
bilingual speech production.
2. Inhibitory Control Model
It may not be surprising that researchers have posited that while producing
speech in one language, the non-target language is suppressed or inhibited in order to
allow production of the former. Green (1986, 1998) has been the biggest advocate of
inhibitory processes in lexical selection. In 1986, Green proposed the ICM, a framework
that explained how bilinguals control both of their languages as to prevent massive
lexical intrusions from the language-not-in-use. The basic assumption of this model is
that when a bilingual wants to speak in one language only, it becomes selected and any
further production of the nonrelevant language is inhibited. Figure 1 is an illustration of
how the ICM explains bilingual speech production of a simple picture. In this example,
an English-Spanish bilingual is told to name a picture of a chair in English. Upon seeing
the picture, the semantic system sends activation to many lexical nodes of English and
Spanish (the target, its translation equivalent and a cohort of related words). At the
lexical level, each word contains a language tag and those belonging to Spanish become
inhibited. The bilingual is now able to select the word “chair” based on its activation
28
level and language tag.
Green (1986) hypothesizes that words in both languages must contain particular
language tags or certain labels that the speaker’s IC system makes use of to ensure that
the relevant output is in the target language. The use of language tags is a critical part of
Green’s model and also has been adopted by other researchers such as Costa, Miozzo,
and Caramazza (1999).
Figure 1. The Inhibitory Control Model (Green, 1986, 1998) (adapted from Finkbeiner,
Gollan, & Caramazza, 2006).
The language switching paradigm has shown that bilinguals have more difficulty
switching into their L1 than into their L2. This empirical finding has been said to
support the ICM because more inhibition must be applied to a larger system (L1) when
speaking in the L2 and therefore when a bilingual wants to switch back to the L1, it will
require more time to reactivate that larger system from being suppressed than it would
the L2 system. Such predictions have been supported in numeral naming tasks (Meuter
1994, Meuter & Allport 1999, Finkbeiner, Almeida, Janssen, & Caramazza 2006) and in
picture naming tasks (Costa & Santesteban 2004). It must be noted, however, that
although these findings have been reported in laboratory settings, mixed results in
nature speech situations have been found (Grosjean 1988, 1997, Grosjean & Miller
1994, Li 1996, 1998).
In their seminal article, Meuter and Allport (1999) conducted a battery of
switching experiments on “reasonably proficient” bilinguals who were presented with
lists of numerals (1-9). The order of the numerals and length of the lists were
randomized. Each list consisted of 5-14 numerals which were placed in either a blue or
29
yellow rectangle indicating that the participant was to name the numeral in his/her L1 or
L2, respectively. Within each list of numerals, Meuter and Allport randomly included
anywhere from 0-4 switches which could be either L1 or L2 switch trials. Their results
suggest many things: 1) that switch trials are slower than nonswitch trials; 2) that
nonswitch trials have faster RTs in L1; 3) that L2 switch trials have faster RTs than their
L1 counterparts (which suggests that a switch to L1 is more difficult); and 4) that the
RTs increase with each successive switch.
Meuter and Allport’s study has provided the most influential support for the IC
Model and its hypotheses have also been reported in Meuter (1994) and replicated in
Costa and Santesteban (2004). The latter is one of the most important studies to date
that puts both the two above theoretical frameworks within the same context. These
researchers questioned if highly proficient bilinguals would act differently than low
proficient bilinguals in terms of which cognitive mechanisms they would call upon
during speech production. They hypothesized that IC would not be observed in lexical
selection for highly proficient bilinguals but would be for those with lower proficiency
in the L2. As in Meuter and Allport (1999), Costa and Santesteban operationalized IC
by asymmetrical switching costs associated with L1-L2 and L2-L1 language switches.
To test the extent of the ICM in bilinguals of various proficiency levels, Costa
and Santesteban conducted a series of five experiments similar to the design used in
Meuter and Allport (1999). The results of their experiments suggested that inhibitory
control is sensitive to L2 proficiency level and can be summed up as follows:
(1).....................Language switching costs are present in all types of bilingual speakers
tested.
(2).....................Asymmetrical switching costs are present for L2 learners (those with
lower proficiency levels) but not for highly proficient bilinguals.
(3).....................The switching performance of a highly proficient bilingual is
independent of the difference in proficiency levels between the two languages involved
in the task.
(4).....................In a language switching task, highly proficient bilinguals are slower in
their dominant than in their non-dominant language both for switch and nonswitch trials.
Costa and Santesteban’s (2004) study sheds new light on how psycholinguistics
view bilingual speech production. Their study has posited that the control mechanisms
that are called upon when bilinguals speak is relative to their L2 proficiency level.
When bilinguals are less proficient in their L2, they will need to rely on inhibitory
mechanisms for them to select the right word in the correct language. However, with
gains in proficiency, bilinguals begin to make use of language-specific selection
30
mechanisms. Costa and Santesteban left open the possibility that this shift, however,
could be affected by L2 age of acquisition.
3. Present Study
The present study addresses the issue of whether proficiency or L2 age of
acquisition is responsible for the lack of inhibitory control among highly proficient
bilinguals reported in previous studies. It is hypothesized that how bilinguals select the
target language of production and control intrusions from the irrelevant language will
depend on their proficiency level. In accordance with Costa and Santesteban (2004),
low proficient bilinguals will exhibit evidence of inhibitory control and highly
proficient bilinguals will be able to rely upon language-specific selection mechanisms.
Moreover, regardless of proficiency level, I predict that age of acquisition will modulate
these mechanisms.
3.1. Participants and experimental groups
The subjects who participated in both experiments included 53 English native
speakers who were learning Spanish as their second language. All participants were
currently enrolled in at least one course conducted in the L2. These participants were
recruited from a large university in the United States and ranged in age from 18-44.
Two separate statistical procedures were conducted in the present study. For the
analysis of proficiency, the median proficiency score (110) determined two
experimental groups of bilinguals: a high proficiency group (N=29) and a low
proficiency group (N=24). For the analysis of L2 age of acquisition, the median age
reported in the language history questionnaire (12) determined two experimental groups
of bilinguals: early (N=28) and late (N=25).
3.2. Estimating L2 age of acquisition
A language history questionnaire was administered to gather information
regarding the participants’ experience with their second language. On this questionnaire,
all participants listed the age at which they began learning their second language. The
ranges of age of acquisition for all participants varied between < 1 and 40.
3.3. Measuring proficiency
31
A proficiency measure was conducted to estimate L2 proficiency level based on
spontaneously and rapidly generating as many words as possible related to a given
semantic category in the less dominant language. A total of ten categories (e.g., animals,
fruits, clothing, etc.) were taken from Gollan, Montoya, and Werner (2002) and were
individually verbalized to each participant. Upon hearing the category, each participant
was given 60 seconds to produce as many items within that category as s/he could.
These responses were taped, transcribed, and checked for accuracy by the researcher. A
total proficiency score was calculated by adding all responses from each of the ten
semantic categories described above. Each word was only counted once and therefore,
words that may have been repeated during the proficiency measure were not included in
the participant’s proficiency score. Essentially, this score represents the total number of
words produced in the L2 for all ten semantic categories during a total of ten minutes.
3.4. Stimuli
In accordance with Costa and Santesteban (2004), the present experiment
includes ten black and white line drawings from the Snodgrass and Vanderwart (1980)
picture list. The materials used for all participants were the same. As in Costa and
Santesteban, all ten pictures used in the present experiment were presented individually
on a computer screen and were classified as either a nonswitch trial or a switch trial. A
nonswitch trial is defined as one in which the previous trial is named in the same
language and a switch trial is one in which the previous trial is named in a different
language.
3.5. Design
A range of 5-14 pictures were randomly placed in 100 lists which contained
anywhere from 0-4 switch trials. The total number of trials in the experiment was 950
(665 nonswitch trials (70%) and 285 switch trials (30%)). All of the pictures were
individually presented in a colored box as a language cue: blue if the target was to be
produced in English or yellow if the target was to be produced in Spanish. There was
equal production of L1 and L2 (475 responses in L1 and 475 responses in L2 were
elicited). Each of the pictures was presented 95 times during the experiment. For lists
with 5-10 pictures in length, no pictures were duplicated. However, for lists of 11-14
pictures, the repeated pictures were placed at least three trials apart from their first
32
presentation.
All ten pictures used in the current experiment can be found in the illustrated
example of a list of 10 pictures in Figure 2. Note that due to space and color restrictions,
the letters “E” and “S” appear in the trials only to represent that the pictures are to be
named in English and Spanish, respectively. In the actual experiment, these language
cues were in the form of a color cue as mentioned above. In Figure 2, picture (1)
demonstrates to the participant that the first trial (2) will need to be named in English
trial. Trials (4) and (8) represent switch trials and trial (12) demonstrates that the list has
ended. Thus, this example includes a list of ten pictures with two language switches
(one L1 and one L2).
Figure 2. Experimental list example.
3.6. Procedure
There were six practice lists of pictures followed by 100 identically-structured
experimental lists. Each list first presented either the word “English” inside a blue box
or “Spanish” inside a yellow box for 2000 ms in the center of the computer screen to
establish a common fixation point for all pictures. This allowed participants to be able
to focus on the position and only name pictures throughout the experiment without
having to look in other places of the screen for the target. The words “English” and
“Spanish” also served to reinforce the association between color and language of
production (i.e., blue represented English and yellow indicated Spanish). In any given
list, the first picture was presented in the same box as the fixation box. This picture
English ESE S
S E E
S
E *
1 2 3 4 5 6
7 8 9 10 11 12
E
33
remained on the screen for 2000 ms or until the participant responded. The next picture
(either a switch or nonswitch trial) was shown and the cycle was repeated until the end
of the list, at which time an asterisk (*) was presented for 1000 ms to show that the list
had finished and that another one would begin in 1000 ms.
3.7. Data analyses
Only correct responses were included in the reaction time analyses; correct and
incorrect responses were both included in the accuracy analyses. Accuracy was coded
by the researcher as either correct (if the participant correctly named the target) or
incorrect (if the participant incorrectly named the target or if there was a technical
malfunction). Two ANOVAs were conducted to investigate the effects of L2 age of
acquisition and L2 proficiency on inhibitory control.
4. Effects of L2 Age of Acquisition
Mean scores across the four contexts were computed for each participant. An
ANOVA was conducted using participant means as random factors with age of
acquisition (early or late) as the between-group factor and response language (L1 or L2)
and trial type (nonswitch or switch) as within-group factors.
The descriptive statistics for reaction time, accuracy, and the switching costs
associated for L1 and L2 switches for the two groups tested appear in Table 1.
Table 1
Error Rate (in %) and Reaction Time (RT) and Switching costs (in ms) for Early
and Late Age of Acquisition
Early Age of Acquisition Late Age of Acquisition
L1 L2 L1 L2
RT Acc RT Acc RT Acc RT Acc
Switch 1013 94.4 931 96.8 957 94.1 887 96.4
Nonswitch 935 96.3 877 98.0 882 96.4 832 97.8
Switching
cost 78 54 75 55
34
4.1. Response latencies
At first glance, the data shown in Table 1 demonstrate similar switching costs
associated for those with early and late age of acquisition. In fact, the ANOVA revealed
a main effect for response language, F(1, 52) = 39.62, MSE = 230507.76, p < .0001.
This suggests that it took more time for participants to name pictures in the L1 than in
the L2 (947 ms and 881 ms, respectively). In this analysis, as in the previous ones,
picture naming was faster in the L2 (in accordance with previous findings). There was
also a main effect reported for trial type, F(1, 52) = 88.37, MSE = 232257.03, p < .0001.
This indicates that participants were slower at naming pictures in switch trials (947 ms)
than in nonswitch trials (881 ms).
There was one interaction from these analyses that is essential to discuss. As
expected, there was a significant interaction for response language and trial type, F(1,
52) = 6.20, MSE = 6181.59, p < .05. This interaction is important because it indicates
that the magnitude of the switching costs is different for L1 and L2 switches. It is very
important to note that this interaction, however, was not dependent upon whether the
participants were exposed to their less dominant language early or late in life. Thus,
unlike the results from the proficiency analyses, the three-way interaction for response
language, trial type, and age of acquisition was not significant, F(1, 52) = .09, MSE =
89.32, p = .77. In other words, bilinguals, regardless of the age at which they began
learning their L2, suffer similar asymmetrical switching costs and thus utilize IC to help
them select the target word for production.
4.2. Error analyses
In the analysis of accuracy, there were two main effects. First, there was a
reported main effect for response language, F(1, 52) = 25.68, MSE = .019, p < .0001.
This suggests that participants were more accurate naming pictures in a particular
language. Mean error percents showed that naming in L1 (95.3%) was less accurate
than in L2 (97.2%). Second, there was an observed main effect for trial type, F(1, 52) =
39.99, MSE = .015, p < .0001. This finding indicates that participants were more
accurate in nonswitch trials (97.1%) than in switch trials (95.4%). The ANOVA revealed
no significant interactions.
5. Effects of L2 Proficiency
35
Mean scores across the four contexts were computed for each participant. An
ANOVA was conducted using participant means as random factors with proficiency
(low or high) as the between-group factor and response language (English or Spanish)
and trial type (nonswitch or switch) as within-group factors.
The descriptive statistics for reaction time, accuracy, and switching costs
associated for switching between languages for the two proficiency groups tested appear
in Table 2.
Table 2
Error Rate (in %) and Reaction Time (RT) and Switching costs (in ms) for Low
and High Proficiency
Low Proficiency High Proficiency
L1 L2 L1 L2
RT Acc RT Acc RT Acc RT Acc
Switch 1006 93.9 917 96.9 974 94.7 908 96.3
Nonswitch 913 96.6 867 98.3 913 96.1 849 97.5
Switching
cost 93 50 61 59
5.1. Response latencies
Table 2 shows that the magnitude of the switching costs for both languages is
similar for the high proficiency group (L1: 61 ms; L2: 59 ms) but the opposite is
observed for those with lower proficiency (L1: 93 ms; L2: 50 ms). In other words, low
proficient bilinguals need more time to switch into their more dominant language.
The ANOVA revealed a main effect for response language, F(1, 52) = 39.62,
MSE = 230507.76, p < .0001. This suggests that naming pictures in one particular
language was slower than the other (L1: 947 ms; L2: 881 ms). In this analysis, as in the
previous ones reported in this dissertation, picture naming was faster in the L2 (in
accordance with previous findings). There was also a main effect reported for trial type,
F(1, 52) = 88.37, MSE = 232257.03, p < .0001. This indicates that participants were
slower at naming pictures on switch trials (947 ms) than on nonswitch trials (881 ms).
Logically, there were two interactions that are important to the response latencies
analyses. First of all, there was a significant interaction for response language and trial
type, F(1, 52) = 6.20, MSE = 6181.59, p = .016. This interaction is important because it
36
suggests that the magnitude of the switching costs is different for L1 and L2 switches.
A very important three-way interaction emerged from the speech production of
low and high proficiency groups for response language, trial type, and proficiency, F(1,
52) = 5.29, MSE = 5271.24, p = .025. Unlike the analyses for bilingual type that
suggested that the overall data were similar for all bilinguals, this three-way interaction
indicates that not only are there asymmetrical switching costs for L1 and L2 switch
trials, but that the magnitude of the switching costs are dependent upon the participants’
L2 proficiency. As shown in Table 2, the difference between L1 and L2 switch trials was
43 ms for those with low proficiency and was only 2 ms for those with high proficiency.
Figure 3 illustrates that participants with low proficiency take more of a “hit” when
switching to their more dominant language than those with high proficiency.
Figure 3. Differences in switching costs by proficiency.
5.2. Error analyses
In the analysis of accuracy, there were two main effects. First, there was a
750
850
950
1050
RT (in ms)
L1 L2
Low Proficiency
Nonswitch
Switch
750
850
950
1050
RT (in ms)
L1 L2
High Proficiency
Nonswitch
Switch
37
reported main effect for response language, F(1, 52) = 25.68, MSE = .019, p < .0001.
This suggests that participants were more accurate naming pictures in a particular
language. Mean error percents showed that overall picture naming in L1 (95.3%) was
less accurate than in L2 (97.2%). Second, there was an observed main effect for trial
type, F(1, 52) = 39.99, MSE = .015, p < .0001. This finding indicates that participants
were most accurate in nonswtich trials. Indeed, mean error percents showed that naming
pictures at nonswitch trials (97.1%) led to fewer errors than at switch trials (95.4%)
6. Discussion
The statistical analyses conducted revealed several interesting results. Recall that
Costa and Santesteban (2004) have suggested that during development in a second
language, bilinguals move away from using IC in their speech production. They
attributed this to either L2 proficiency or the age at which the L2 was acquired. The
present study investigated whether or not these two factors affected the reliance on
inhibitory control in the speech production of bilinguals. The L2 age of acquisition
analyses revealed no significant effects on inhibitory control. This is to say that overall,
participants with early and late ages of acquisition performed similarly. The proficiency
analyses in the present study, however, painted a very different picture. Indeed,
bilinguals with low proficiency in their less dominant language need to inhibit one
language for production of the other. However, as also revealed in Costa and
Santesteban (2004), the high proficiency group in the current study did not show
evidence supporting the claims put forth by the IC Model. Overall, the analyses that
explored the effects of L2 proficiency on inhibitory control have supported Costa and
Santesteban’s claim that with increases in proficiency in the L2, there is a shift away
from inhibitory control to reliance on language-specific selection mechanisms.
The results of the present study align well with Schwieter and Sunderman’s
(submitted) Selection by Proficiency (SbP) Model. As can be seen in Figure 4, this
model of bilingual speech production entertains the notion that inhibitory control is
modulated by L2 proficiency. According to these researchers, in this model, bilinguals
move along a proficiency continuum in a bi-directional manner. Figure 4 illustrates that
with second language use and practice—which in turn leads to increased proficiency—
bilinguals acquire the ability to achieve language selectivity and therefore, can rely on
language-specific selection mechanisms at the conceptual level. Until they can achieve
this, they will be forced to rely on inhibitory control to suppress any non-target
language words that may be activated at the lexical level.
38
Figure 4. The Selection by Proficiency Model (Schwieter & Sunderman, 2008;
sumbitted)
Schwieter and Sunderman’s (submitted) SbP Model is a visualization of
lexicalization in bilingual speech production for a less proficient and more proficient
bilingual. The model suggests that when a less proficient bilingual is asked to name a
picture of cat in his second language, several lexical candidates in both languages
receive activation. The words in the non-target language become suppressed so that it is
easy for the bilingual to decide which word in the target language matches the target
concept. The SbP Model argues that this procedure is very distinct for a highly
proficient bilingual. The latter has acquired a skill that allows him/her to specify the
target language from the beginning stages of speech production. When a highly
proficient English-Spanish bilingual is asked to name a picture of a cat in Spanish,
several lexical candidates in both languages receive activation from a preverbal message
which contains vital higher linguistic information about speech production including but
not limited to: 1) the target language; 2) the linguistic register; and 3) the concept to be
L2L1
L2
L1
Proficiency
Language: English
Register: Formal
gato
perro
dogcat
Concept: Cat
gato
perro
dogcat
L1
Language
39
lexicalized. These cues work together to ensure that the target lexical item has a higher
activation level than its competitors. Under this assumption, there is no need for
inhibitory control.
7. Conclusion
The present study examined the extent of inhibitory control during bilingual
speech production. In particular, it explored the effects that certain factors of
bilingualism (L2 age of acquisition and proficiency) have on determining whether or
not these control mechanisms will be relied on (versus reliance on language-specific
selection mechanisms). The statistical procedures revealed no effect for age of
acquisition. This suggests that learning a second language early or late in life does not
appear to impact the control mechanisms that are required to restrict the lexicalization
procedure to one language during bilingual speech production. This was not the same,
however, when analyses were conducted to explore the effects of proficiency. The
results of these analyses revealed striking patterns which were in accordance with Costa
and Santesteban’s (2004) hypothesis: it was suggested that reliance on inhibitory control
is essentially a continuum on which bilinguals move from processes involving
inhibitory control to rely on language-specific selection mechanisms. Essentially, then,
this suggests that with increases in L2 proficiency, bilinguals may acquire a skill that
allows them to establish the language of production and resolve competition at the
conceptual level. Until this skill has been acquired, (i.e., at low proficiencies) the
language of production is established at the lexical level where competition between
languages is also resolved.
The results of the present study articulate the marked differences in the cognitive
mechanisms that underpin bilingual speech production as explained by the SbP Model.
Future studies need to address what particular elements of proficiency (i.e., lexical
robustness, ability to conceptualize in the L2, etc.) lead to the reliance on language-
specific selection mechanisms and avoid having to suppress the irrelevant language.
Future studies also need to address the possibility that is demonstrated in the SbP
Model: highly proficient bilinguals may revert back to inhibitory control. If this is the
case, what contexts or factors dictate such a regression?
40
8. References
Costa, A. 2005. Lexical access in bilingual production. In J. F. Kroll & A. M. B. de
Groot (eds), Handbook of Bilingualism: Psycholinguistic Approaches. New York:
Oxford.
Costa, A., Miozzo, M. & Caramazza, A. 1999. Lexical selection in bilinguals: Do words
in the bilinguals’ two lexicons compete for selection? Journal of Memory and
Language 41. 365-397.
Costa, A. & Santesteban, M. 2004. Lexical access in bilingual speech production:
Evidence from language switching in highly proficient bilinguals and L2 learners.
Journal of Memory and Language, 50. 491-511.
Finkbeiner, M., Gollan, T. & Caramazza, A. 2006. Lexical access in bilingual speakers:
What’s the (hard) problem? Bilingualism: Language and Cognition 9(2). 153-166.
Finkbeiner, M., Almeida, J., Janssen, N. & Caramazza, A. 2006. Lexical selection in
bilingual speech production does not involve language suppression. Journal of
Experimental Psychology: Learning, Memory, and Cognition 32(5). 1075-1089.
Gollan, T. H., Montoya, R. I. & Werner, G. A. 2002. Semantic and letter fluency in
Spanish—English bilinguals. Neuropsychology 16(4). 562-576.
Green, D. W. 1986. Control, activation, and resource: A framework and a model for the
control of speech in bilinguals. Brain and Language 27. 210-223.
Green, D. 1998. Mental control of the bilingual lexico-semantic system. Bilingualism:
Language and Cognition 1. 67-81.
Grosjean, F. 1988. Exploring the recognition of guest words in bilingual speech.
Language and Cognitive Processes 3. 233–274.
Grosjean, F. 1997. Processing mixed languages: Issues, findings and models. In A. M. B.
de Groot & J. F. Kroll (eds), Tutorials in bilingualism: Psycholinguistic perspectives.
Mahwah, NJ: Erlbaum.
Grosjean, F. & Miller, J. L. 1994. Going in and out of languages: An example of
bilingual flexibility. Psychological Science 5. 201–206.
Li, P. 1996. Spoken word recognition of code-switched words by Chinese-English
bilinguals. Journal of Memory and Language 35. 757–774.
Li, P. 1998. Mental control, language tags, and language nodes in bilingual lexical
processing. Bilingualism: Language and Cognition 1. 92-93.
Meuter, R. 1994. Language switching in naming tasks. Unpublished doctoral
dissertation, Oxford University.
Meuter, R & Allport, A. 1999. Bilingual language switching in naming: Asymmetrical
41
costs of language selection. Journal of Memory and Language 40. 25-40.
Ratcliff, R. 1993. Methods for dealing with outliers. Psychological Review 114. 510-532.
Schwieter, J. W. & Sunderman, G. (2008). Language switching in bilingual speech
production: In search of the language-specific selection mechanism. The Mental
Lexicon, 3(2), 214-238.
Schwieter, J. W. & Sunderman, G. submitted. Concept selection in bilingual speech
production: The role of proficiency.
Snodgrass, J.G. & Vanderwart, M. 1980. A standardized set of 260 pictures: Norms for
name agreement, image agreement, familiarity, and visual complexity. Journal of
Experimental Psychology: Human Learning and Memory 6. 174-215.
42
ResearchGate has not been able to resolve any citations for this publication.
Article
Full-text available
Presents a standardized set of 260 pictures for use in experiments investigating differences and similarities in the processing of pictures and words. The pictures are black-and-white line drawings executed according to a set of rules that provide consistency of pictorial representation. They have been standardized on 4 variables of central relevance to memory and cognitive processing: name agreement, image agreement, familiarity, and visual complexity. The intercorrelations among the 4 measures were low, suggesting that they are indices of different attributes of the pictures. The concepts were selected to provide exemplars from several widely studied semantic categories. Sources of naming variance, and mean familiarity and complexity of the exemplars, differed significantly across the set of categories investigated. The potential significance of each of the normative variables to a number of semantic and episodic memory tasks is discussed. (34 ref) (PsycINFO Database Record (c) 2006 APA, all rights reserved).
Article
Full-text available
Recent research on language production suggests that bilinguals shift from using inhibitory control mechanisms to a language-specific selective mechanism during development (Costa, Santesteban, & Ivanova, 2006). Costa et al. argue that the robustness of the L2 lexical representations may be critical to the functionality of a language-specific selective mechanism. Accordingly, in the present study we measured the lexical robustness of a group of 54 English dominant learners of Spanish using a verbal fluency task and investigated its effect on their performance in a picture-naming task with language switches. The results suggest that L2 lexical robustness predicts the shift to a language-specific selective mechanism during speech production. Moreover, we demonstrate a specific threshold of lexical robustness necessary to engage the mechanism.
Article
Full-text available
Models of bilingual speech production generally assume that translation equivalent lexical nodes share a common semantic representation. Though this type of architecture is highly desirable on both theoretical and empirical grounds, it could create difficulty at the point of lexical selection. If two translation equivalent lexical nodes are activated to roughly equal levels every time that their shared semantic representation becomes activated, the lexical selection mechanism should find it difficult to “decide” between the two (the “hard problem”) – yet in some cases bilinguals benefit from the presence of a translation equivalent “competitor”. In this article, we review three models that have been proposed as solutions to the hard problem. Each of these models has difficulty accounting for the full range of findings in the literature but we suggest that these shortcomings stem from their acceptance of the assumption that lexical selection is competitive. We argue that without this assumption each proposal is able to provide a full account of the empirical findings. We conclude by suggesting that the simplest of these proposals should be rejected before more complicated models are considered.
Article
When bilinguals speak to one another, they choose a base language to interact in and then, depending on the need, code-switch to the other (guest) language for a word, a phrase, or a sentence During the perception of a code switch, there is a momentary dominance of base-language units at the onset of the switch, but it is unknown whether this base-language effect is also present in production, that is, whether the phonetics of the base language carry over into the guest language In this study, French-English bilinguals retold stories and read sentences monolingually in English and in French and bilingually in French with English code switches Both the stories and the sentences contained critical words that began with unvoiced stop consonants, whose voice onset times (VOT) were measured The results showed that the base language had no impact on the production of code switches The shift from one language to the other was total and immediate This manifestation of cross-linguistic flexibility is accounted for in terms of a bilingual production model
Article
Bilinguals, i.e. those who use two languages in their everyday lives, move in and out of various speech modes when speaking to different interlocutors. When conversing with monolinguals, they speak one language and reduce the activation level of the other language, but when conversing with other bilinguals they choose a base language of interaction and often bring in the other language by either code-switching or borrowing. The aim of the present study is to explore how “guest words” (code-switches and borrowings) are processed by bilingual listeners when interacting with other bilinguals. Different types of English words (varying in phonotactic configuration and lexicon membership) were embedded in French sentences and were produced either as code-switches or borrowings. The gating paradigm (Grosjean, 1980) was used to present these words to French-English bilingual listeners so as to determine the role played by word type and language phonetics in the lexical access of guest words, as well as to uncover the underlying operations involved in the recognition process. Results showed that the phonotactics of a guest word, the presence or absence of a base language homophone, the language phonetics of the word, as well as the language that precedes the word, all play a role in the recognition process. An interactive activation view of bilingual word recognition is proposed to account for the results found in the study.
Article
Two experiments with Chinese–English bilinguals were conducted to examine the recognition of code-switched words in speech. In Experiment 1, listeners were asked to identify a code-switched word in a sentence on the basis of increasing fragments of the word. In Experiment 2, listeners repeated the code-switched word following a predesignated point upon hearing the sentence. Converging evidence from these experiments shows that the successful recognition of code-switched words depends on the interaction among phonological, structural, and contextual information in the recognition process. The results also indicate that Chinese–English bilinguals can recognize code-switched words with the same amount of information as required by monolingual English listeners. These results are interpreted in terms of parallel activation and interactive processes in spoken word recognition.
Article
In an experimental study of language switching and selection, bilinguals named numerals in either their first or second language unpredictably. Response latencies (RTs) on switch trials (where the response language changed from the previous trial) were slower than on nonswitch trials. As predicted, the language-switching cost was consistently larger when switching to the dominant L₁ from the weaker L₂ than vice versa such that, on switch trials, L₁ responses were slower than in L₂. This "paradoxical" asymmetry in the cost of switching languages is explained in terms of differences in relative strength of the bilingual's two languages and the involuntary persistence of the previous language set across an intended switch of language. Naming in the weaker language, L₂, requires active inhibition or suppression of the stronger competitor language, L₁; the inhibition persists into the following (switch) trial in the form of "negative priming" of the L₁ lexicon as a whole. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Article
In this paper Green proposes an inhibitory control (IC) model of bilingual lexical processing. At the core of Green's arguments is the notion of “mental control,” formulated in terms of inhibition, control schemas, and a supervisory attentional system. The very notion of control, it seems, suggests some sort of intentional, exogenous force at work (e.g., the supervisory attentional system). Presumably mental control differs from automatic processes (Schneider & Shiffrin, 1977), yet in the IC model there is no precise computational specification of how the various parameters of the control system actually interact to determine automatic bilingual processes. In a computational view, the IC model has quite some symbolic AI flavor (e.g., with goal-oriented decision boxes and control schemas), but it also attempts to integrate activation-based accounts (e.g., interactive activation mechanisms). Again, because the model remains at a rather conceptual level as presented, it is difficult to determine how successful it will be in combining symbolic and connectionist approaches in understanding bilingual processing.