Bilingualism in 11 pages, Kosz z czasopismami

[ Pobierz całość w formacie PDF ]
Journal of Deaf Studies and Deaf Education
Bilingualism and Attention: A Study of Balanced and
Unbalanced Bilingual Deaf Users of American Sign Language
and English
Poorna Kushalnagar*
1
, H. Julia Hannay
2
, Arturo E. Hernandez
2
1
University of Washington
2
University of Houston
Received October 26, 2009; revisions received March 9, 2010; accepted March 17, 2010
Early deafness is thought to affect low-level sensorimotor pro-
cessing such as selective attention, whereas bilingualism is
thought to be strongly associated with higher order cognitive
processing such as attention switching under cognitive load.
This study explores the effects of bimodal-bilingualism (in
American Sign Language and written English) on attention
switching, in order to contrast the roles of bilingual proficiency
and age of acquisition in relation to cognitive flexibility among
deaf adults. Results indicated a strong high-proficiency bilin-
gual advantage in the higher order attention task. The level of
proficiency in 2 languages appears to be the driving force for
cognitive flexibility. However, additional data are needed to
reach conclusive interpretation for the influence of age of sec-
ond language acquisition on higher order attention-switching
ability and associated cognitive flexibility.
visual attention to the peripheral field among deaf indi-
viduals as compared to hearing individuals (Bavelier
et al., 2001; Bosworth & Dobkins, 2002; Loke & Song,
1991; Neville & Lawson, 1987a, 1987b; Parasnis &
Samar, 1985; Proksch & Bavelier, 2002), but not the
mechanisms of attention efficiency for alerting and ori-
enting (Dye, Baril, & Bavelier, 2007) or simultaneous
visual processing of multiple objects (Hauser, Dye,
Boutla, Green, & Bavelier, 2007). When the individual
is required to use endogenous attention to rapidly switch
from one visual field to another visual field, higher order
cognitive processing is activated. An important part of
higher order cognitive processing is that it reduces bur-
den on the frontal lobe and increases efficiency for at-
tentional flexibility. Additionally, higher order cognitive
ability appears to be enhanced by bilingual experience
(Bialystok, Craik, Klein, & Viswanathan, 2004; Bialystok,
Craik, & Ryan, 2006).
According to a code-switching theory (Peal &
Lambert, 1962), switching between languages provides
the bilingual individual with a higher degree of mental
flexibility and concept formation. This cognitive ad-
vantage may be based on use of inhibitory functions of
the frontal lobe such that interference from another
language is inhibited and one can selectively attend to
the language that is currently in use (Green, 1998).
Data from a case study on bilingual patient with a fron-
tal lesion as well as functional neuroimaging study of
normal bilinguals suggest that the anterior cingulate
and dorsolateral prefrontal regions are intimately
Visual selective attention and attention switching differ
in their representational nature. Visual selective attention
is thought to be more reliant on low-level sensorimotor
processing, whereas attention switching is strongly asso-
ciated with higher order cognitive processing. As low-
level sensorimotor and visual functions mature, neural
networks for higher order cognitive functions are built
upon these foundations and subsequently emerge later in
life (Gogtay et al., 2004). However, early deafness may
affect the cortical organization for low-level percep-
tual processes such as visuospatial attention or the
associated neural network that facilitates newborn
neural development. Specifically, differences in early
sensory experience appear to influence allocation of
*Correspondence should be sent to: Poorna Kushalnagar, Department
of Health Services, University of Washington, Seattle, WA 98195
(e-mail: poornak@u.washington.edu).
The Author 2010. Published by Oxford University Press. All rights reserved.
For Permissions, please email: journals.permissions@oxfordjournals.org
doi:10.1093/deafed/enq011
Advance Access publication on April 18, 2010
264 Journal of Deaf Studies and Deaf Education 15:3 Summer 2010
involved in switching between two languages (Fabbro,
Skrap, & Aglioti, 2000; Hernandez, Dapretto, Mazziotta,
& Bookheimer, 2001; Hernandez, Martinez, & Kohnert,
2000). These regions were also identified consistently as
important for shifting attention function in a meta-anal-
ysis study across 31 positron emission tomography and
functional magnetic resonance imaging (fMRI) studies
(Wager, Jonides, & Reading, 2004). This analysis also
concluded that neural activation during tasks requiring
location shifting (switching attention to targets from one
location to another location) was associated with signif-
icantlymoreactivityintherightpremotorandintra-
parietal sulcus regions.
Behavioral studies of bilingualism and attention
have provided additional support for neuropsycholog-
ical and neuroimaging findings on the beneficial effects
of bilingualism on frontal lobe functioning. For exam-
ple, Bialystok, Craik, Klein, and Viswanathan (2004)
found that complex attention performance under
a cognitive load among bilingual adults exceeded that
of the same-age monolingual adults. The cognitive
benefits associated with bilingual proficiency persist
through old age (Bialystok, Craik, & Ryan, 2006). In
a study with children, Yang (2007) reported higher
performance among bilinguals compared to monolin-
guals on an attention network test that involves several
aspects of executive control and attention shifting
skills. The continuous involvement of several neural
areas in the frontal regions for language and attention
switching affords an opportunity to strengthen higher
order neurocircuitry within the frontal cortex, provid-
ing a bilingual individual with an attention network
system advantage. Interestingly, this bilingual advan-
tage appears to extend to nonlinguistic domains. As
such, we propose that (a) low-level sensorimotor de-
velopment associated with selective attention is con-
stant across early deafness and (b) higher order
cognitive development for attention switching is
driven by distinct bilingual experience. We expect that
deaf people will perform similarly on selective attention
tasks, but may show variation on attention-switching
task due to bilingual proficiency level.
Empirical studies on bilingualism and cognition thus
far have been limited to participants whose hearing ability
falls within normal limits. Recent research suggests that
that the cognitive advantage associated with bilingualism
is specific to languages that share output modalities (e.g.,
spoken English and Spanish). When a bimodal-bilingual
user simultaneously uses two languages with different
output modalities (e.g., spoken English and American
Sign Language [ASL]), neither language is actively sup-
pressed, as compared to bilingual speakers who can only
use one spoken language at a time. In a recent study by
Emmorey, Luk, Pyers, and Bailystok (2008), 15 hearing
adults who are accustomed to speaking English while
signing in ASL at the same time (bimodal-bilingual
users) did not differ from 15 monolinguals (one spoken
language) in performing a task requiring inhibition and
mental flexibility. The authors argued that this is a con-
sequence of simultaneous use of two different language
output modalities such as spoken and signed language.
This appears to promote development of a distinct neu-
ral system that permits simultaneous use or merging of
these two languages, unlike using two spoken languages
consecutively, which requires suppression or inhibition
of one language during utilization of the other language.
Emmorey et al. (2008) proposed that deaf people
who are users of two same-modality signed languages
(e.g., ASL and British Sign Language) should demon-
strate similar cognitive benefits to those observed in dual
spoken-language bilinguals. However, the majority of
deaf Americans use ASL as the only signed language.
They are able to read and write English, ranging from
low to high fluency. Some of these individuals may not
produce clear speech production that may mislead hear-
ing people to perceive that they are also not fluent in
reading and writing English. If an individual is highly
fluent in reading and writing English, then this individ-
ual is considered proficient in this language even if his/
her speech production is unclear. Such an individual
may be considered to be bimodal-bilingual in that he/
sheisabletoreadEnglishanduseASL.However,itis
unlikely that this individual will simultaneously speak in
English and use sign language at the same level as hear-
ing bimodal-bilingual users in the study by Emmorey
et al. Additionally, it is not possible to simultaneously
communicate in both modalities (sign and write). The
effect of these differences on the cognition of deaf Amer-
icans who are proficient English readers and proficient
ASL signers has not been explored. It is essentially ‘‘un-
charted territory.’’ We attempted to determine whether
high proficiency in ASL and written English provide
Attention Processing and Bilingualism 265
deaf individuals with the advantage that is frequently
observed in hearing bilinguals of two spoken languages.
Given the above points, it may be logical to expect
that deaf Americans who are proficient in ‘‘both’’ signed
and written languages would demonstrate better cogni-
tive control and mental flexibility abilities than deaf
Americans who demonstrate high proficiency in one
language but lower proficiency in another language.
English and ASL were found to utilize the classical
language regions in the left-hemisphere brains of hear-
ing speakers of English and deaf people who are fluent
users of ASL (Neville et al., 1998). Deaf people who are
highly proficient in written English and ASL should
also display the cognitive advantages associated with
bilingualism that was observed in studies of hearing
bilinguals reviewed in this study. Thus, the level of
bilingual proficiency is likely to play a significant role
in higher order cognitive processing such as attention
switching and executive control rather than low-level
perceptual processing such as selective attention.
This article was designed to examine the effects of
bilingualism on switching attention between central
and peripheral visual stimuli among deaf adults. If
bilingualism provides the individual with an advantage
on tasks that require involvement of the cognitive
control network, then we expect to find that deaf bilin-
guals who have high proficiency in both languages
(balanced bilinguals) will significantly outperform
low-proficient deaf bilinguals (unbalanced bilinguals)
on a higher order attention-switching task. However,
group differences are not expected on low-level selec-
tive attention tasks that do ‘‘not’’ require shifting at-
tention from one part of the visual field (i.e., central)
to another visual field (periphery).
central or peripheral vision. A separate repeated meas-
ures ANOVA was conducted with reaction time (RT)
as the dependent measure of attention.
Participants
Institutional Review Board approval was obtained
from the University of Houston’s Committee for the
Protection of Human Subjects. A total of 59 deaf par-
ticipants were recruited in Texas, New Mexico, and
Washington, DC. Nine participants were excluded
from data analyses due to videogame experience
(N
5
3), difficulty with visual perception (N
5
1),
inability to complete testing (N
5
1), reported neu-
rological issues known to affect attention performance
(N
5
2), or late onset of hearing loss beyond 36
months (N
5
2). All participants had normal or
corrected-to-normal vision.
Participants were given objective proficiency testing
in English and ASL. Two participants who obtained
scores lower than 1 standard deviation (SD)fromthe
group mean on both language tests were removed.
Grouping of the remaining 48 eligible participants was
based on the following criteria: (a) High proficiency on
‘‘both’’ English and ASL measures as determined by
cut-off scores resulted in assignment to the balanced
bilingual group (N
5
24) and (b) high proficiency in
one language and low proficiency in another language
resulted in assignment to the unbalanced bilingual
group (N
5
24). Participants in both groups had normal
or corrected-to-normal vision. There were no age group
differences (t
52
0.238; p
5
.81). For nonverbal in-
telligence, the difference (t
5
1.81; p
5
.08) was con-
sidered to lack significance statistically by conventional
criteria. Nevertheless, this variable was entered as a cova-
riate in analyses to control for its potential effect on the
outcomes. The two bilingual groups differed on the age
of first language acquisition (t
5
2.6; p
,
.01) and age of
second language acquisition (t
5
2.8; p
,
.01). These two
covariates were also entered in the analyses. A summary
of demographics and language characteristics for each
groupisprovidedinTable1.
Methods
Design Overview
The design involved a mixed analysis of variance
(ANOVA) with bilingual proficiency (balanced vs un-
balanced) as the between-groups factor. Type of atten-
tion (central, peripheral, switch) was the repeated
measures factor. The attention conditions were coun-
terbalanced in terms of order of presentation. From
the theory of signal detection (TSD; Swets, 1964), d#
provided a measure of sensitivity to visual stimuli in
Apparatus and Stimuli
Stimulus presentation and recording of responses were
controlled by PsyScope software (Cohen, MacWhinney,
266 Journal of Deaf Studies and Deaf Education 15:3 Summer 2010
Ta b l e 1 Demographics and language characteristics for balanced and unbalanced bilingual groups (N
5
48)
Balanced bilinguals
Unbalanced bilinguals
Age of first language acquisition in months
a
English as first language
N
5
13 (20 months; 4)
N
5
19 (25 months; 4)
ASL as first language
N
5
11 (4 months; 3)
N
5
5 (22 months; 9)
Age of second language acquisition in years
a
English as second language
N
5
11 (3 years; 2)
N
5
5 (4.5 years; 1)
ASL as second language
N
5
13 (13 years; 2)
N
5
19 (16 years; 1)
Age
a
34 (5)
34 (7)
IQ
a
105 (12)
99 (11)
Ethnicity
18 (75%) Caucasian
16 (67%) Caucasian
Gender
14 (58%) females
14 (58%) females
Etiology
Genetic
12 (50%)
5 (21%)
Nongenetic causes
4 (17%)
8 (33%)
Unknown
8 (33%)
11 (46%)
Higher proficiency in
English
Proficiency in English and ASL is
balanced for this group
14 (58%)
ASL
10 (42%)
a
Mean (SD) listed for each variable; N (percentage) is listed for all other variables.
Flatt, & Provost, 1993). The stimuli were presented on
a 17-inch monitor connected to a PowerBook or iMac,
and responses were collected via any key pressed on the
keyboard. Each of the attention conditions involved
four cross-symbols (
1
) arranged in a square format
in the center and corresponding four cross-symbols in
the periphery forming a larger square, with different
instructions for each condition. The predetermined
number of the four cross-symbol (
1
) stimuli in the
periphery was obtained from the findings of Proksch
and Bavelier (2002) regarding deaf participants’ greater
allocation of their attention resource to the peripheral
visual field at a maintenance load of four compared to
six stimuli. Four additional stimuli with an identical
cross-symbol (
1
) shape were also displayed in the fo-
veal region of attention. The participant sat 36 inches
from the monitor. The cross symbols for the central
attention condition were 1.25 inches from the center,
yielding an eccentricity of 2 degrees. The peripheral
attention crosses were 3.25 inches from the center,
yielding an eccentricity of 5 degrees. The stimulus loca-
tions were at angular positions of 60,120 ,240 ,or
300 from the center of the screen. The cross symbol
was either white or black. The background was gray.
A Go-No/Go paradigm was used to develop an
attention-switching computer experiment. The Go-
No/Go paradigm requires a simple motor response
to a target while withholding this response when a tar-
get is not present. In its classic form, the Go-No/Go
paradigm provides a clean measure of inhibition with-
out additional cognitive processing such as that in-
volved in executive function demands or response
selection for two-button (yes/no) choice task (Proctor
& Vu, 2003). The one-button response in the Go-No/
Go paradigm promotes more accurate responding and
reduces additional cognitive (e.g., decision making as-
sociated with button pressing) and motor processing
that typically emerge in two-button Yes/No paradigm
(Perea, Rosa, & Gomez, 2002). The attention condi-
tions in this study were designed to tap two functions:
selectively attending to targets among distracters (Go)
and inhibiting responses to nontargets (No/Go).
Materials and Procedure
Two measures were used to assign deaf participants to
language groups and to obtain other demographic and
descriptive information. We developed a Demographic
Questionnaire to gather self-reported information about
the participant’s background that included the etiology of
participants’ deafness, early intervention period, language
experience, self-rating of parent–child communication
fluency level during childhood, handedness, and educa-
tion history. The Raven’s Standard Progressive Matrices
Attention Processing and Bilingualism 267
Test (Raven, Raven, & Court, 1998/2003) is a culturally
nonspecific, well-validated instrument that provided
a valid nonverbal measure of intellectual ability in chil-
dren, adolescents, and adults (Lezak, 2004).
Two language measures were administered and
used for group assignment based on language profi-
ciency in English and ASL. High proficiency for En-
glish was defined as a score equivalent to or above
Grade 10.5 on the Passage Comprehension subtest
(Woodcock & Johnson, 1989). With consultation from
an ASL linguist, we developed 30 items that tapped
linguistic aspects of ASL syntax. These include nega-
tion, wh- and rhetorical questions, yes–no, verb agree-
ment, and number agreement. On this ASL syntax
test, the participant was asked to judge if the ASL
syntax on video was correct or incorrect. Because
there is no objective measure of ASL that has been
validated for public use, a self-rating of ASL profi-
ciency developed by the investigator was also admin-
istered to all participants and examined for its
correlation with objective proficiency test of ASL syn-
tax (r
5
.52, p
,
.001). A composite score was calcu-
lated by averaging scores on the objective (ASL test)
and the subjective (ASL self-rating) measures. ASL
composite scores at the group mean or higher were
labeled as highly proficient for ASL fluency. High-
proficiency scores on both English and ASL tests were
required for a participant to be assigned to the bal-
anced bilingual group. Participants who had ASL
composite score that fell below group mean ‘‘or’’ En-
glish proficiency score of 10.4 or below were assigned
to the unbalanced bilingual group.
Participants were tested individually and typically
completed all tasks within 2 hr. The computer atten-
tion experiment comprised 600 test trials divided into
three blocks (conditions) of 200 trials each with a rest
break between each pair of blocks. After instructions
were presented on the screen and by the researcher,
the participant completed 25 practice trials with feed-
back and repetition to become familiar with each con-
dition. RT was recorded in milliseconds on all trials,
but no feedback was given on test trials.
For the ‘‘central’’ attention condition, the partici-
pant was asked to ignore the four distracters in the
periphery and attend to the four crosses in the central
vision. The participant was instructed to hit the key if
exactly ‘‘one white’’ cross appeared in the central vi-
sion (target). If not so (nontarget), the participant was
asked to refrain from hitting the key. Nontarget
included two or more white crosses or no white cross
at all.
For the ‘‘peripheral’’ condition, the participant was
asked to ignore the distracters in the center and press
any key if exactly ‘‘one white’’ cross appeared in periph-
eral vision (target), and if this was not so (nontarget),
refrain from hitting the key. Nontarget included two or
more white crosses or no white cross at all. The partic-
ipant was reminded to fixate eyes on the center while
attending to target stimuli in the periphery.
Location of the target within central (or periph-
eral) vision was randomly distributed across trials.
Figure 1 illustrates the event sequence for one trial
in central condition. The event sequence for periph-
eral condition is identical except that the target is
located in the periphery region. The response
interval was set at 1,200 ms. If the participant saw
a target and responded correctly by hitting a key, this
was counted as a hit. A false alarm occurred when
a nontarget appeared and the participant hit a key. If
the target appeared on the screen and the participant
failed to hit the key, this was coded as an omission.
After the 1,200-ms interval ended, the next trial
began.
The attention-switching condition involved two
types of cognitive load: switching between targets in
the central and peripheral regions over trials and
refraining from hitting the key or spacebar when
a nontarget stimulus appeared after a repeated se-
quence of three targets. Participants were asked to
hit the key when they saw one white cross-symbol
stimulus in ‘‘either’’ central or peripheral visual
regions, and do nothing if this was not met (see Figure
2). The switching condition had 200 trials with 150
targets and 50 nontargets. The increased number of
targets in the switching condition as compared to cen-
tral and peripheral conditions was intended to induce
a high frontal lobe cognitive load, requiring more cog-
nitive flexibility and readiness to inhibit responses to
nontargets after a continuous series of targets (Lavie,
Hirst, de Fockert, & Viding, 2004). Flexibility in shift-
ing attention and immediately inhibiting responses
under cognitive load in this Go/No-Go paradigm
[ Pobierz całość w formacie PDF ]

  • zanotowane.pl
  • doc.pisz.pl
  • pdf.pisz.pl
  • braseria.xlx.pl

    •