beha io al
sciences
A icle
Volun a y Language Swi ching in he Con ex o
Bilingual Aphasia
Nicholas G unden 1, Gio gio Piazza 2, Ca men Ga cía-Sánchez 3and Ma co Calab ia 4,*
1Cen e o B ain and Cogni ion, Pompeu Fab a Uni e si y, 08005 Ba celona, Spain; ng [email p o ec ed]
2Basque Cen e on Cogni ion, B ain and Language (BCBL), 20009 Donos ia-San Sebas ián, Spain;
[email p o ec ed]
3Hospi al de la San a C eu i San Pau, 08041 Ba celona, Spain; CGa [email p o ec ed]
4Facul y o Heal h Sciences, Uni e si a Obe a de Ca alunya, 08018 Ba celona, Spain
*Co espondence: [email p o ec ed]; Tel.: +34-934-505-418
Recei ed: 23 July 2020; Accep ed: 14 Sep embe 2020; Published: 18 Sep embe 2020
Abs ac :
As s udies o bilingual language con ol (BLC) seek o explo e he unde pinnings o
bilinguals’ abili ies o juggle wo languages, di e en ypes o language swi ching asks ha e been
used o unco e swi ching and mixing e ec s and he eby e eal wha p oac i e and eac i e con ol
mechanisms a e in ol ed in language swi ching. Volun a y language swi ching asks, whe e a
bilingual pa icipan can swi ch eely be ween hei languages while naming, a e being u ilized mo e
o en due o hei g ea e ecological alidi y compa ed o cued swi ching pa adigms. Because his
ype o ask had no ye been applied o language swi ching in bilingual pa ien s, ou s udy sough
o explo e olun a y swi ching in bilinguals wi h aphasia (BWAs) as well as in heal hy bilinguals.
In Expe imen 1, we eplica ed p e iously epo ed esul s o swi ch cos s and mixing bene i s wi hin
ou own bilingual popula ion o Ca alan-Spanish bilinguals. Wi h Expe imen 2, we compa ed bo h
he pe o mances o BWAs as a g oup and as indi iduals agains con ol g oup pe o mance. Resul s
illus a ed a complex pic u e o language con ol abili ies, indica ing a ying deg ees o associa ion
and dissocia ion be ween ac o s o BLC. Gi en he di e si y o impai men s in BWAs’ language
con ol mechanisms, we highligh he need o examine BLC a he indi idual le el and h ough he
lens o heo e ical cogni i e con ol amewo ks in o de o u he pa se ou how bilinguals egula e
hei language swi ching.
Keywo ds:
bilingual aphasia; olun a y language swi ching; bilingual language con ol; p oac i e
con ol; eac i e con ol
1. In oduc ion
Bilinguals ha e he uncanny abili y o manage hei languages. Once achie ing mode a e
p o iciency, a bilingual has he powe o main ain a language h oughou a con e sa ion and a oid
blu ing ou unwan ed in usions om hei o he languages. Then, gi en he need, many bilinguals
can seamlessly “ lip he sc ip ” and swi ch in and ou o languages o communica e wi h he people
a ound hem. This se o abili ies is usually e med as bilingual language con ol (BLC) and includes
a numbe o cogni i e p ocesses [
1
–
3
]. Ou lined in he Adap i e Con ol Hypo hesis (ACH; [
4
,
5
]),
a comp ehensi e desc ip ion o BLC includes a leas eigh con ol p ocesses: goal main enance,
con lic moni o ing, in e e ence supp ession, salien cue de ec ion, selec i e esponse inhibi ion, ask
engagemen and disengagemen , and oppo unis ic planning.
In o de o s udy how bilinguals e ec i ely swi ch be ween languages, mos s udies hus a ha e
employed expe imen al asks wi h cued swi ching be ween languages (e.g., [
6
–
13
]; o ecen e iews
see [
14
,
15
]). In hese asks, subjec s a e explici ly shown wha language hey need o name a gi en
Beha . Sci. 2020,10, 141; doi:10.3390/bs10090141 www.mdpi.com/jou nal/beha sci
Beha . Sci. 2020,10, 141 2 o 23
s imulus in wi h a isual cue (e.g., a colo , lag o a gi en language egion, e c.) F om he esul ing da a,
i is possible o measu e wo ypes o cos , swi ch and mixing cos s. Swi ch cos s a e calcula ed as he
di e ence in naming la encies be ween “swi ch” ials, whe e he a ge language changed compa ed
o he p e ious ial, and “ epea ” o “s ay” ials, whe e he naming language was he same as in he
p e ious ial. Mixing cos s a e calcula ed as he di e ence be ween epea and single ials, whe e
single ials a e hose named in he non-mixed condi ions, such ha only one naming language is
in use (e.g., [
16
,
17
]). Howe e , in ligh o mo e ecen indings o mixing bene i s in some bilingual
popula ions [
18
,
19
], i would be mo e accu a e o e e o hese gene al measu es as swi ching e ec s
and mixing e ec s. Concep ually, swi ching e ec s can be hough o as e lec ions o he abili y o
esol e c oss-language in e e ence, language engagemen , and disengagemen , while mixing e ec s
a e ela ed o wo king memo y mechanisms, such as he demand in main aining ask goals ha a e
p esen in a dual-language si ua ion [17,20,21].
Resea ch wi h pa hological popula ions pe o ming hese asks has epo ed use ul insigh s on
how BLC wo ks. Fo ins ance, Calab ia e al. [
22
] adminis e ed a cued language swi ching ask o a
pa ien wi h pa hological language swi ching, wi h esul s showing ha he pa ien ’s pe o mance on
he ask was exac ly he same as he pe o mance in he mo e na u alis ic connec ed speech condi ion.
Tha is, he pa ien exhibi ed c oss-language in usions om hei non-dominan in o hei dominan
language in he swi ching ask, jus as hey did when hey we e equi ed o desc ibe complex pic u es o
when hey engaged in no mal con e sa ion. Fu he mo e, language swi ching asks ha e been use ul
in assessing he in eg i y o BLC mechanisms. Calab ia e al. [
23
] ound ha , in seman ic demen ia,
swi ching abili ies measu ed ia language swi ching asks may be spa ed despi e a ma ked deg ada ion
o seman ic memo y and anomic de ici s. Finally, a se ies o s udies in pa ien s wi h neu odegene a ion
in he basal ganglia ha e highligh ed an inc eased impac on BLC de ici s compa ed o o he con ol
mechanisms; esul s indica ed ha language swi ching abili ies can be mo e a ec ed han non-linguis ic
con ol abili ies in bilingual pa ien s wi h Pa kinson’s disease (PD) [
20
] and ha language swi ching
is clea ly dissocia ed om o he language con ol abili ies [
21
]. These s udies hus e eal a c ucial
dis inc ion be ween BLC de ici s a ec ing con ol pa hways in bilingual language p oduc ion and
gene alized language de ici s, as PD pa ien s did no exhibi any ype o language diso de .
Gi en hese indings, i is plausible hen o hypo hesize ha pa ien s wi h aphasia may also
ha e some speci ic impai men in language swi ching abili ies. The e is e idence ha , in some
ci cums ances, bilingual pa ien s wi h aphasia may show in olun a y c oss-language in usions and/o
language mixing by blending mo phological ea u es o he wo languages wi hin a wo d o wo ds in
a sen ence [
24
–
31
]. Mo e ecen ly, we ha e also unco e ed e idence ha pa ien s wi h aphasia wi hou
in olun a y language mixing o swi ching ha e shown language con ol de ici s a he lexical le el
ha p e en hem om pe o ming e ec i e wo d e ie al [
32
,
33
]. In bilinguals wi h aphasia (BWA)
such as hese who do no show p onounced pa hological swi ching, language di icul ies ha e been
a gued o s em om p oblems in con olling and managing he inhibi ion o hei languages [
34
,
35
].
The e o e, i is easonable o specula e ha hei con ol de ici s may ex end o di icul ies in language
disengagemen and engagemen , as well as language main enance when hey ind hemsel es in
con ex s whe e swi ching back and o h be ween he wo languages is equi ed. Howe e , li le is
cu en ly known abou hese de ici s in bilingual indi iduals wi h pos -s oke aphasia ha do no
demons a e o e in olun a y language mixing o swi ching.
Add essing his issue, we aimed o iden i y he BLC de ici s ha may p e en BWAs om
e icien ly engaging wi h hei language swi ching abili ies. To do so, we explo ed he pe o mance o
Ca alan-Spanish BWA and heal hy con ols on a language swi ching ask, speci ically ocusing on wo
key aspec s.
Fi s , we used a olun a y language swi ching ask as a ela i ely new expe imen al app oach
o his issue and a me hod o cap u ing a mo e na u al swi ching beha io o ou popula ion o
bilinguals. Volun a y swi ching is ega ded as a mo e ecological measu e o language swi ching [
3
],
especially o bilinguals imme sed in dual-language o dense code-swi ching con ex s [
4
] such as is
Beha . Sci. 2020,10, 141 3 o 23
he case o mos Ca alan-Spanish bilinguals in he Ba celona me opoli an a ea. In pas s udies, his
ype o language swi ching ask has been employed as an al e na i e o cued swi ching pa adigms
when s udying he unde lying con ol mechanisms o endogenous language swi ching wi hin heal hy
indi iduals [
18
,
19
,
36
–
38
] bu he e is no s udy o da e ha inco po a es pa ien s wi h bilingual
aphasia. Consequen ly, any indings in he olun a y swi ching ask o BWA will hold clinical and
ecological signi icance.
Second, ou s udy o language swi ching abili ies in BWAs ocused on wo main con ol componen s
measu ed ia swi ching and mixing e ec s. In an ongoing deba e su ounding he na u e o BLC
mechanisms, i has ecen ly been p oposed ha hese con ol mechanisms a e di e en ially in ol ed in
single- s. dual-language con ex s o speech p oduc ion (e.g., [
16
,
17
]). Fu he mo e, ecen indings
om cued language swi ching asks in bilingual pa ien s wi h PD ha e demons a ed ha swi ch and
mixing cos s a e possibly ela ed o wo quali a i ely di e en mechanisms [
20
,
21
]. Wi hin he con ex
o he dual mechanisms o con ol (DMC) amewo k o non-linguis ic execu i e con ol [
39
,
40
], hese
wo dual-language e ec s ha e been associa ed wi h wo di e en ypes o con ol, eac i e con ol and
p oac i e con ol. Reac i e con ol, measu ed by swi ching e ec s, is de ined as a bo om-up, ansien ,
and s imulus-d i en ype o con ol whe eas p oac i e con ol, measu ed by mixing e ec s, is op-down,
mo e sus ained, and goal-di ec ed. This dual-mechanisms concep o con ol has subsequen ly been
applied o explain he unde lying mechanisms o BLC [
17
]: eac i e con ol is engaged when bilinguals
ha e o sol e o c oss-language in e e ence in swi ch ials, while p oac i e con ol comes in o play
when hey ha e o main ain hei wo languages ac i e du ing a dual-language naming condi ion.
In o ming ou cu en s udy, hese wo con ol mechanisms ha e been s udied wi hin he con ex o
olun a y language swi ching wi h a ying esul s. In some cases, i has been epo ed ha pa icipan s
did no show swi ch cos s when u ilizing ce ain olun a y naming s a egies, sugges ing an e ec i e
eliance upon eac i e con ol (“bo om-up swi ching”; [
41
]). Howe e , o he expe imen s ha e shown
he opposi e, whe e pa icipan s s ill had swi ch cos s in he olun a y swi ching condi ion [
18
,
19
].
In con as o cued language swi ching, p oac i e con ol o bilinguals in olun a y language swi ching
has la gely e ealed a mixing bene i , whe e bilinguals named pic u es as e in mixed-language s.
single-language con ex s [18,19,38].
P esen S udy
Ou s udy in ol ed wo expe imen s: Expe imen 1, whe e we sough o eplica e p e ious
s udies’ explo a ion o olun a y language swi ching [
18
] wi hin ou dis inc bilingual popula ion
o Ca alan-Spanish bilinguals; and Expe imen 2, whe e we compa ed pe o mance o BWAs and
heal hy bilingual con ols on he olun a y language swi ching ask. In bo h expe imen s, we es ed
pa icipan s wi h he ollowing linguis ic p o ile: ea ly (lea ning bo h languages be o e 6 yea s old) and
balanced (la gely equi alen equency o language usage) bilinguals wi h high le els o p o iciency in
bo h languages.
Wi hin Expe imen 1, young Ca alan-Spanish bilinguals pa icipa ed in a pic u e naming ask
in ol ing blocks wi h a ying a ge languages: wo single-language blocks (one in Spanish and
one in Ca alan) ollowed by a dual-language block whe e hey we e ins uc ed o name he pic u e
in ei he language. The design o he expe imen is a eplica ion o he de B uin e al. [
18
] s udy,
wi h he excep ion o cogna e s a us in s imuli; in ou s udy, hal o he ials we e cogna e wo ds
(wi h phonological o e lap be ween he Ca alan and Spanish wo ds o a gi en i em) and hal we e
non-cogna e wo ds (wi hou phonological o e lap). Because he modi ica ion is mino , we la gely
expec ed simila esul s o hose ob ained by de B uin e al. [
18
]. Fi s , we expec ed ha he equency o
language swi ching would be a ound 50%, whe e abou hal o he ials in he dual-language condi ions
will be named in Ca alan and abou hal in Spanish. Second, as was ound by de B uin e al. [
18
],
we expec ed a mixing bene i o ou bilinguals, measu ed by la ge naming la encies o single ials
(single-language condi ion) compa ed o epea ials (dual-language condi ion). Thi d, despi e his
hypo hesized bene i o mixing o bilinguals in he dual-language condi ion, we also expec ed o ind
Beha . Sci. 2020,10, 141 4 o 23
a swi ch cos wi hin he dual-language con ex . This would sugges ha swi ching and he usage
o language engagemen and disengagemen con ol p ocesses would elici a cos o he language
sys em. The magni ude o he swi ch cos is expec ed o be he same o he wo languages based on
p e ious s udies’ indings wi h high-p o icien bilinguals in cued language swi ching [
9
,
10
,
42
,
43
] and
by de B uin e al. [
18
] in olun a y language swi ching. Finally, as an es ablished body o li e a u e
main ains he p esence o a cogna e acili a ion e ec in heal hy bilinguals [
44
,
45
], we likewise p edic
ha ou pa icipan s will bene i in cogna e ials and demons a e smalle naming la encies.
Wi hin Expe imen 2, Ca alan-Spanish bilinguals wi h aphasia and age-ma ched heal hy con ols
comple ed a olun a y language swi ching ask, as in Expe imen 1. Fo he BWA g oup, we cen e ed
ou p edic ions on he hypo hesis ha BWA may show de ici s in moni o ing hei wo languages
du ing a dual-language naming condi ion. E idence suppo ing his hypo hesis comes om ou
p e ious s udy wi h bilingual aphasia, in which we looked a he ela ionship be ween seman ic
in e e ence in pic u e naming and con lic moni o ing [
32
]. Resul s showed ha bilinguals wi h
aphasia esponded slowe compa ed o heal hy con ols when pe o ming an execu i e con ol ask
equi ing hem o moni o o bo h cong uen and incong uen s imuli (a lanke ask) bu we e as
e icien as heal hy con ols when hey had o sol e o s imulus incong uence (con lic cos ). C i ically,
BWA pe o mance on he lanke ask co ela ed wi h hei delayed naming la encies on a pic u e
naming ask and was hus in e p e ed as e idence o a gene alized con lic moni o ing de ici . In he
cu en expe imen , de ici s in moni o ing could ha e epe cussions when a emp ing o main ain wo
languages ac i e in he dual-language condi ion and impac he p oac i e con ol sys em ( o p oac i e
non-linguis ic de ici s in bilingual pa ien s wi h aphasia, see [46]).
In ligh o hese p e ious indings, ou p edic ions o BWA compa ed o con ols we e he
ollowing: Fi s , i BWAs ha e de ici s ha would g ea ly hinde hei pe o mance in he dual-language
condi ion, we p edic ed ha hei swi ching equency would be lowe han 50%, as hey p esumably
s ick mo e o one language a he han al e na ing be ween he wo. Second, i hey do exhibi a lowe
a e o swi ching han con ols, we p edic ed ha he pe o mance in he dual-language condi ion
o BWAs would esemble he single-language condi ion and hus we would no expec o ind a
mixing bene i epo ed in p e ious s udies (de B uin e al., 2018). Thi d, we expec ed ha pa ien s
wi h aphasia and heal hy con ols would show simila magni udes o swi ch cos s due o p e ious
e idence ha bilinguals wi h aphasia demons a e p ese ed eac i e con ol on a non-linguis ic con ol
ask [
46
]. Finally, as has been es ablished in heal hy bilinguals as well as bilinguals wi h aphasia [
47
],
we p edic ed ha BWAs would bene i om a cogna e acili a ion e ec .
2. Expe imen 1: Young Adul Bilinguals
2.1. Me hods
2.1.1. Pa icipan s
A o al o 20 uni e si y-age, heal hy pa icipan s (12 women, 8 men) we e ec ui ed om a olun ee
da abase a Pompeu Fab a Uni e si y. Sociodemog aphic and language backg ound in o ma ion was
collec ed ia a p e-expe imen al ques ionnai e (see Table 1). All pa icipan s ga e w i en in o med
consen in acco dance wi h he Decla a ion o Helsinki. The expe imen al p o ocol was app o ed by
he “Pa c de Salu MAR” Resea ch E hics Commi ee ( e e ence numbe : 2018/8029/I). Pa icipan s
we e compensa ed o hei ime du ing he expe imen .
Beha . Sci. 2020,10, 141 5 o 23
Table 1. Sociodemog aphic and linguis ic measu es o young Ca alan-Spanish bilinguals.
M SD
Age (yea s) 22.35 2.58
Educa ion (yea s) 16.35 2.58
BSWQ Subsco es (max. 15)
DL Swi ch 8.65 1.84
NDL Swi ch 8.20 1.91
Con ex ual Swi ch 8.20 2.61
Unin ended Swi ch 6.65 2.74
BSWQ O e all Swi ch (max. 60) 31.70 6.73
Dominan Language (DL)
Age o Acquisi ion (yea s) 0.75 1.16
P o iciency (max. 7)
Speaking 7.00 —
Comp ehension 6.75 0.64
Reading 6.85 0.49
W i ing 7.00 —
Language Usage (%) 54.97 14.48
Non-dominan Language (NDL)
Age o acquisi ion (yea s) 1.75 1.74
P o iciency (max. 7)
Speaking 7.00 —
Comp ehension 6.75 0.55
Reading 6.75 0.55
W i ing 6.95 0.22
Language Usage (%) 34.19 14.10
In he ques ionnai e gi en o pa icipan s, age o acquisi ion and deg ees o p o iciency in eading,
w i ing, speaking, and comp ehension ( a ed on 7-poin scales, wi h “1” indica ing low p o iciency
and “7” indica ing high p o iciency) we e eco ded. Based on esponses, subjec s who ook pa in
his s udy we e all conside ed o be ea ly, balanced Ca alan-Spanish bilinguals, as indica ed by hei
acquisi ion o bo h hei languages be o e 5 yea s o age and ha ing balanced p o iciency in bo h
languages. Addi ionally, because hese measu es did no indica e a clea L1 o L2, we labeled he
pa icipan s’ languages as hei dominan language (DL) and non-dominan language (NDL), acco ding
o sel - epo ed dominance. In ou sample, 10 pa icipan s we e labeled as Ca alan-dominan and he
emaining 10 pa icipan s we e Spanish-dominan . F equency o usage o bo h Spanish and Ca alan was
epo ed as well. Finally, in o de o ob ain measu es o swi ching beha io ou side he expe imen al
se ing, pa icipan s’ swi ching was assessed using he Bilingual Swi ching Ques ionnai e (BSWQ; [
48
]).
This ques ionnai e yielded 5 swi ching sco es: L1-Swi ch, L2-Swi ch, Con ex ual Swi ching, Unin ended
Swi ching, and O e all Swi ching. Following ou classi ica ion o ou bilinguals’ wo languages,
L1-Swi ch and L2-Swi ch we e e med DL-Swi ch and NDL-Swi ch, espec i ely. See Table 1 o
sociodemog aphic and language in o ma ion.
2.1.2. P ocedu e
The isual s imuli o his ask consis ed o 60 di e en pic u es aken om Snodg ass and
Vande wa [
49
]. S imuli we e balanced ac oss Ca alan and Spanish o measu es o loga i hmic
equency (Ca alan: M=1.19, SD =0.52; Spanish: M=1.11, SD =0.46; (59) =0.90, p=0.39), wo d
leng h (Ca alan: M=5.86, SD =1.44; Spanish: M=5.70, SD =1.25; (59) =0.66, p=0.41), and numbe
o phonemes (Ca alan: M=5.27, SD =1.32; Spanish: M=5.35, SD =1.25; (59) =0.82, p=0.91). The
equencies o he Spanish and Ca alan names we e ob ained om he LEXESP [
50
] and he Ca alan
Dic iona y o F equencies [
51
] da abases, espec i ely. In addi ion o hese linguis ic ac o s, hal o
he pic u e names we e cogna e wo ds while he o he hal we e non-cogna e wo ds be ween he
wo languages.
Beha . Sci. 2020,10, 141 6 o 23
A he onse o he expe imen , subjec s unde wen a amilia iza ion exe cise whe e hey we e
p esen ed wi h all expe imen al pic u es and asked o ead aloud he co ec esponses o each pic u e
p in ed below i in bo h Spanish and Ca alan. While s imuli wi h high name ag eemen we e selec ed
o his expe imen , his ini ial p esen a ion o pic u es se ed o s eng hen his name ag eemen o
s imuli ac oss pa icipan s.
Expe imen al asks included h ee blocks: wo single-language blocks and a dual-language block.
Each single-language block consis ed o 120 naming ials (60 pic u es p esen ed 2 imes each) in ei he
Spanish o Ca alan. The o de o hese single-language blocks was coun e balanced ac oss subjec s and
ins uc ions o he ask we e always gi en in he a ge language o he block. These single-language
blocks we e hen ollowed by a dual-language block (360 ials, 60 pic u es epea ed 6 imes each),
whe e pa icipan s we e asked o name i ems wi h “whiche e language comes mos na u ally” bu
also o “swi ch languages mul iple imes” h oughou he ask. The language o hese ins uc ions
was coun e balanced ac oss pa icipan s so as o balance any p iming e ec s in subsequen naming.
Dual-language blocks included wo ca ch ials a he beginning o he pic u e naming which we e
hen dis ega ded du ing analysis. Because o he longe du a ion o hese blocks, subjec s we e gi en
wo oppo uni ies o es du ing he naming ials. Each o hese b eaks was also ollowed by ano he
ca ch ial ha was hen aken ou o analyses.
Fo all ials ega dless o block, subjec s we e p esen ed wi h a ixa ion c oss o 500 ms, ollowed
by he image o be named by he pa icipan . Said image emained on he sc een o 2000 ms
(i espec i e o when a pa icipan began/ended hei esponse) and audio was eco ded du ing his
ime o cap u e esponses gi en. The pic u e naming ask was adminis e ed o pa icipan s using
DMDX so wa e [52].
2.1.3. Da a Analysis
Ou o he 20 young bilingual pa icipan s ec ui ed o his s udy, 1 pa icipan was excluded due
o an abno mally high numbe o e o s (42 e o s; g oup e o a e age =11.95). The emaining 19
pa icipan s we e included in all subsequen s a is ical analyses.
The dependen a iables o his s udy, naming la encies and accu acy o pa icipan s’ esponses,
we e analyzed o -line wi h CheckVocal [
53
]. Fo each audio ile, he beginning o he pa icipan ’s
esponse was ma ked and he esponse was coded as co ec o inco ec . Naming la encies o inco ec
esponses and hose exceeding 2 SDs abo e o below a gi en subjec ’s mean naming la ency we e
excluded. Naming e o s we e classi ied unde he ollowing ypes [
54
]: (a) omissions: non-in elligible
e bal esponse gi en du ing eco ding window o no esponse; (b) seman ic e o s: e bal esponse
seman ically ela ed o a ge wo d; (c) o mal e o s: dele ion, subs i u ion, o addi ion o phonemes
in a ge wo d; (d) un ela ed e o s: e bal esponse wi h no seman ic o o he ela ion o a ge wo d;
(e) c oss-language in usions: co ec naming o pic u e bu in non- a ge language (only applicable o
single-language blocks); and ( ) au o-co ec ion: inco ec e bal esponse ollowed by co ec ion wi h
a ge wo d.
In o de o examine e ec s o swi ching and mixing, ials in single-language blocks we e classi ied
as “single” ials whe eas, in dual-language blocks, ials we e labeled as “ epea ” o “swi ch”,
depending on whe he he language used o name an i em was he same ( epea ) o di e en (swi ch)
compa ed o he p e ious ial. Fo e o ials ha we e able o be ca ego ized as ei he Spanish o
Ca alan (seman ic e o s, non-ambiguous o mal e o s, un ela ed e o s and c oss-language in usions),
he ollowing ial was classi ied as epea o swi ch ials in e e ence o he language o he e o .
Howe e , o e o ials ha we e no able o be ca ego ized by language (i.e., no esponse and
ambiguous o mal e o s), he ials di ec ly a e hem we e unclassi iable as epea /swi ch and we e
excluded om analyses. O e all, an a e age o 4.58 ials (SD =6.27), wi h a maximum o 25 ials,
we e unclassi iable pe subjec .
Wi h his da a, we i s analyzed he dis ibu ion o he swi ch ials pe language and pe cogna e
s a us in o de o assess whe he he swi ching equency was a ound 50%, as done by de B uin e al. [
18
].
Beha . Sci. 2020,10, 141 7 o 23
Subsequen ly, we an wo sepa a e epea ed-measu es ANOVAs o RTs and accu acy as dependen
a iables, including T ial Type (Single, Repea , o Swi ch), Language (Dominan s. Non-dominan )
and Cogna e S a us (Cogna e s. Non-cogna e) as wi hin-subjec ac o s. I he assump ion o sphe ici y
(Mauchly’s es ) was iola ed, he G eenhouse–Geisse co ec ion was applied by adjus ing he deg ees
o eedom.
2.2. Resul s
Swi ching equencies.
Pa icipan s on a e age swi ched on 43.99% (SD =7.43) o he mixed
language ials. O hose swi ch ials, he e we e simila pe cen ages o swi ching o cogna es
(
M=49.73
,SD =3.26) and non-cogna es (M=50.27, SD =3.26; (18) =
−
0.37, p=0.718). Likewise,
swi ch ials we e equally sha ed be ween swi ches in o dominan language (M=50.31, SD =0.97) and
swi ches in o non-dominan language (M=49.69, SD =0.97; (18) =1.39, p=0.182).
Naming la encies.
Analyses showed ha he main e ec o T ial Type, F(1.17, 36) =5.52, p=0.02,
ηp2
=0.24, was signi ican . Pos hoc analyses e ealed a signi ican swi ch cos o pa icipan s, whe e
swi ch ials (M =769 ms, SD =96) we e signi ican ly slowe han epea ials (M =753 ms, SD =91;
p=0.002
). The di e ence in magni ude o swi ch cos was no signi ican ly di e en ac oss languages
(DL swi ch cos : M =2.36 ms, SD =3.00; NDL swi ch cos : M =2.07, SD =3.79; (18) =0.245, p=0.809).
Fu he mo e, naming o single ials (M =788 ms, SD =95) was signi ican ly slowe han o epea
ials (M =753 ms, SD =91; p=0.04), sugges ing a mixing bene i , whe e pa icipan s imp o ed hei
naming la encies in dual-language condi ions compa ed o single-language condi ions (see Figu e 1).
Beha . Sci. 2020, 10, x 7 o 23
Naming la encies. Analyses showed ha he main e ec o T ial Type, F(1.17, 36) = 5.52, p = 0.02,
η
p2
= 0.24, was signi ican . Pos hoc analyses e ealed a signi ican swi ch cos o pa icipan s, whe e
swi ch ials (M = 769 ms, SD = 96) we e signi ican ly slowe han epea ials (M = 753 ms, SD = 91;
p = 0.002). The di e ence in magni ude o swi ch cos was no signi ican ly di e en ac oss languages
(DL swi ch cos : M = 2.36 ms, SD = 3.00; NDL swi ch cos : M = 2.07, SD = 3.79; (18) = 0.245, p = 0.809).
Fu he mo e, naming o single ials (M = 788 ms, SD = 95) was signi ican ly slowe han o epea
ials (M = 753 ms, SD = 91; p = 0.04), sugges ing a mixing bene i , whe e pa icipan s imp o ed hei
naming la encies in dual-language condi ions compa ed o single-language condi ions (see Figu e 1).
Figu e 1. Pe o mance on olun a y swi ching ask o young, heal hy Ca alan-Spanish bilinguals: (a)
Naming la encies show signi ican di e ences be ween swi ch and epea ials (swi ch cos ) and
be ween epea and single ials (mixing bene i ) wi h no e ec o language. (b) Dis ibu ion o
p opo ional swi ching and mixing e ec s in bilingual pa icipan s.
The main e ec o Cogna e S a us was also signi ican , F(1,18) = 25.77, p < 0.001, η
p2
= 0.59, whe e
cogna es (M = 759 ms, SD = 88) we e named as e han non-cogna es (M = 783 ms, SD = 93). Nei he
he main e ec o Language, F(1,18) = 0.18, p = 0.68, no any in e ac ions be ween ac o s we e ound
o be signi ican .
Accu acy. Analyses e ealed a signi ican main e ec in T ial Type, F(1.51, 36) = 25.14, p < 0.001,
η
p2
= 0.58. Pos hoc es s o T ial Type e ec s showed a signi ican dec ease in accu acy du ing single
ials (M = 97.24%, SD = 1.6) compa ed o bo h epea (M = 99.12%, SD = 0.61; p < 0.001) and swi ch
(M = 99.17%, SD = 0.78; p < 0.001) ials. The main e ec o Cogna e S a us was also signi ican , F(1,18)
= 4.87, p = 0.04, η
p2
= 0.21, whe e cogna es (M = 98.51%, SD = 1.43) we e named mo e accu a ely han
non-cogna es (M = 97.50%, SD = 1.47). The main e ec o Language, F(1,18) = 2.99, p = 0.101, was no
ound o be signi ican .
The Cogna e S a us × T ial Type in e ac ion was also signi ican , F(2, 36) = 3.92, p = 0.03, η
p2
=
0.18. Pos hoc analyses showed ha non-cogna e single ials (M = 96.24%, SD = 2.86) we e named
signi ican ly less accu a ely han all o he non-cogna e ials and all cogna e ials. The e was also a
signi ican di e ence in accu acy be ween cogna e single ials (M = 97.94%, SD = 2.76) and cogna e
epea ials (M = 99.47%, SD = 1.05).
Co ela ions be ween Language Measu es and Volun a y Swi ching E ec s. To in es iga e
whe he pa icipan s’ language backg ounds a ec ed ask pe o mance, we an co ela ions be ween
language measu es and swi ching/mixing e ec s. Co ela ion analyses we e un compa ing he
magni udes o swi ching and mixing e ec s (calcula ed as an indi idual’s swi ch cos di ided by he
g oup a e ages o epea and swi ch ials × 100) o all BSWQ sco es, ages o acquisi ion, p o iciency
sco es, and equency o usage o bo h dominan and non-dominan languages. No signi ican
co ela ions we e ound be ween measu es o language p o ile and swi ching o mixing e ec s.
Figu e 1.
Pe o mance on olun a y swi ching ask o young, heal hy Ca alan-Spanish bilinguals:
(
a
) Naming la encies show signi ican di e ences be ween swi ch and epea ials (swi ch cos ) and
be ween epea and single ials (mixing bene i ) wi h no e ec o language. (
b
) Dis ibu ion o
p opo ional swi ching and mixing e ec s in bilingual pa icipan s.
The main e ec o Cogna e S a us was also signi ican , F(1,18) =25.77, p<0.001,
ηp2
=0.59, whe e
cogna es (M=759 ms, SD =88) we e named as e han non-cogna es (M=783 ms, SD =93). Nei he
he main e ec o Language, F(1,18) =0.18, p=0.68, no any in e ac ions be ween ac o s we e ound
o be signi ican .
Accu acy.
Analyses e ealed a signi ican main e ec in T ial Type, F(1.51, 36) =25.14, p<0.001,
ηp2
=0.58. Pos hoc es s o T ial Type e ec s showed a signi ican dec ease in accu acy du ing single
ials (M=97.24%, SD =1.6) compa ed o bo h epea (M=99.12%, SD =0.61; p<0.001) and swi ch
(M=99.17%, SD =0.78; p<0.001) ials. The main e ec o Cogna e S a us was also signi ican ,
F(1,18) =4.87, p=0.04,
ηp2
=0.21, whe e cogna es (M=98.51%, SD =1.43) we e named mo e accu a ely
han non-cogna es (M=97.50%, SD =1.47). The main e ec o Language, F(1,18) =2.99, p=0.101, was
no ound o be signi ican .
Beha . Sci. 2020,10, 141 8 o 23
The Cogna e S a us
×
T ial Type in e ac ion was also signi ican , F(2, 36) =3.92, p=0.03,
ηp2=0.1
8.
Pos hoc analyses showed ha non-cogna e single ials (M=96.24%, SD =2.86) we e named
signi ican ly less accu a ely han all o he non-cogna e ials and all cogna e ials. The e was also a
signi ican di e ence in accu acy be ween cogna e single ials (M=97.94%, SD =2.76) and cogna e
epea ials (M=99.47%, SD =1.05).
Co ela ions be ween Language Measu es and Volun a y Swi ching E ec s.
To in es iga e
whe he pa icipan s’ language backg ounds a ec ed ask pe o mance, we an co ela ions be ween
language measu es and swi ching/mixing e ec s. Co ela ion analyses we e un compa ing he
magni udes o swi ching and mixing e ec s (calcula ed as an indi idual’s swi ch cos di ided by he
g oup a e ages o epea and swi ch ials
×
100) o all BSWQ sco es, ages o acquisi ion, p o iciency
sco es, and equency o usage o bo h dominan and non-dominan languages. No signi ican
co ela ions we e ound be ween measu es o language p o ile and swi ching o mixing e ec s.
3. Expe imen 2: Bilinguals wi h Aphasia and Heal hy Con ols
3.1. Me hods
3.1.1. Pa icipan s
Se en Ca alan-Spanish bilinguals wi h aphasia (BWA) we e ec ui ed o ake pa in his s udy
om he Speech Pa hology Clinic a he Hospi al de la San a C eu i San Pau. Ten age-ma ched
bilingual adul s we e also included in his expe imen as neu ologically heal hy con ols. P io o he
expe imen , all pa icipan s comple ed a language backg ound ques ionnai e whe e hey we e asked o
epo sociodemog aphic and linguis ic ac o s, as in Expe imen 1. In his expe imen , language usage
was measu ed wi h a se o ques ions abou he equency o Spanish and Ca alan usage in di e en
se ings and ac oss di e en pe iods o he indi idual’s li e. These esponses we e ans o med in o a
inal sco e, exp essed as a pe cen age o usage whe e 0% was solely Spanish, 100% was solely Ca alan,
and 50% ma ked a balanced usage o he wo languages. Fu he mo e, all esponses on he language
ques ionnai es, wi h he excep ion o hose in he BSWQ, we e gi en by BWAs in e e ence o p emo bid
le els. See Table 2 o compa isons on sociodemog aphic and linguis ic cha ac e is ics be ween g oups.
This s udy was ca ied ou in acco dance wi h he ecommenda ions o he “Pa c de Salu
MAR-Resea ch E hics Commi ee.” All pa icipan s ga e w i en in o med consen in acco dance wi h
he Decla a ion o Helsinki and wi h he esea ch p o ocol ha was app o ed by he “Pa c de Salu
MAR-Resea ch E hics Commi ee” ( e e ence numbe : 2018/8029/I).
3.1.2. Language Assessmen
Along wi h sociodemog aphic and language backg ound da a, clinical da a o BWAs we e also
compiled. Pa ien s we e assessed using he Spanish e sion o he Wes e n Aphasia Ba e y (WAB; [
55
])
in o de o asce ain hei cu en ype and se e i y o aphasia. This clinical assessmen ool was
adminis e ed and sco ed by a clinical neu opsychologis a Hospi al de la San a C eu i San Pau. The
WAB is a comp ehensi e es o language unc ions wi h a ela i ely sho es adminis a ion ime
(30–60 min) and includes ou language sub es s which assess spon aneous speech, comp ehension,
epe i ion, and naming o calcula e an Aphasia Quo ien (AQ). Based on his sco e, he se e i y and
ype o aphasia can be de e mined. In ou sample, 4 BWAs we e classi ied as ha ing mild aphasia and
he emaining 3 as ha ing mode a e aphasia. In e ms o aphasia ype, 3 had anomic aphasia, 2 had
We nicke’s aphasia, 1 wi h conduc ion aphasia, and 1 wi h ansco ical mo o aphasia. Pa ien s we e
only es ed in Spanish as a Ca alan e sion o he WAB is no cu en ly a ailable. The e iologies and
mon hs since he onse o b ain lesions we e also epo ed o each pa ien . Finally, pa ien s did no
p esen any clinically signi ican mo o speech diso de s a he ime o he expe imen . See Table 3 o
clinical da a.
Beha . Sci. 2020,10, 141 9 o 23
Table 2.
Sociodemog aphic and linguis ic measu es o Bilinguals wi h Aphasia (BWAs) and con ols.
Bilinguals wi h Aphasia (BWA) Con ols
M SD M SD p-Values
Age (yea s) 54.43 6.16 47.67 6.87 0.061
Educa ion (yea s) 14.28 2.93 15 2.16 0.57
Language Usage (%) 59.43 10.85 48 20.66 0.203
BSWQ Subsco es (max. 15)
DL Swi ch 9.86 1.86 7.8 1.14 0.012
NDL Swi ch 9.57 0.79 6.3 1.57 <0.001
Con ex ual Swi ch 9.00 1.41 7.1 2.77 0.118
Unin ended Swi ch 7.86 0.69 6.7 2.06 0.176
O e all Swi ch (max. 60) 36.29 3.45 27.9 6.42 0.007
Dominan Language (DL)
Age o Acquisi ion (yea s) 0 0.00 0 0.00 —
P o iciency (max. 7)
Speaking 7 0.00 7 0.00 —
Comp ehension 7 0.00 6.8 0.42 0.17
Reading 7 0.00 6.9 0.32 0.343
W i ing 6.5 0.85 6.9 0.32 0.191
Non-dominan Language
(NDL)
Age o Acquisi ion (yea s) 4.57 4.08 1.7 2.67 0.098
P o iciency (max. 7)
Speaking 7 0.00 7 0.00 —
Comp ehension 6.75 0.66 6.9 0.32 0.54
Reading 7 0.00 6.9 0.32 0.343
W i ing 6.75 0.66 6.9 0.32 0.54
Table 3. Indi idual clinical and linguis ic da a o bilinguals wi h aphasia (BWAs).
Type o
Aphasia
Mon hs
Pos -Onse Se e i y E iology Dominan Language
(DL)
Non-Dominan Language
(NDL)
Pa ien 1
WERNICKE
100
MODERATE
CVA CAT SPAN
Pa ien 2 COND. 156 MILD TUMOR CAT SPAN
Pa ien 3 ANOMIC 83
MODERATE
CVA SPAN CAT
Pa ien 4 ANOMIC 53 MILD CVA CAT SPAN
Pa ien 5 TRANS. M. 129
MODERATE
CVA CAT SPAN
Pa ien 6
WERNICKE
114 MILD CVA CAT SPAN
Pa ien 7 ANOMIC 88 MILD CVA CAT SPAN
Aphasia
Quo ien
(AQ)
Spon.
Speech Comp. Rep. Naming BAT-C DL
(max. =47)
BAT-C NDL
(max. =47)
p- alues
BAT-C DL
s. NDL
Pa ien 1 56.3 13 6.25 3.2 5.7 25 15 0.07
Pa ien 2 84.5 18 9.25 6.4 8.6 36 26 0.05
Pa ien 3 71.4 12 7.5 8.1 8.1 35 44 0.02
Pa ien 4 84.1 17 8.75 7.1 9.2 29 31 0.83
Pa ien 5 74.8 12 8.5 8 8.9 28 32 0.51
Pa ien 6 75.7 16 6.75 6.2 8.9 37 35 0.8
Pa ien 7 87.2 15 10 9.7 8.9 41 41 -
COND.,
Conduc ion;
TRANS. MOTOR,
T ansco ical mo o ;
CVA,
Ce eb o ascula acciden ;
CAT,
Ca alan;
SPAN,
Spanish;
Spon. Speech,
Spon aneous Speech;
Comp.,
Comp ehension;
Rep.,
Repe i ion;
BAT-C,
Bilingual Aphasia
Tes —Pa C.
Pa ien s’ language abili ies we e u he es ed using pa C o he Bilingual Aphasia Tes
(BAT; [
56
]), which assesses c oss-language abili ies o e ou sub es s: Wo d Recogni ion, Wo d
T ansla ion, Sen ence T ansla ion, and G amma ical Judgmen . In Wo d Recogni ion, pa ien s we e
asked o selec he co ec ansla ion o each wo d om a lis o 10 possible choices (5 wo ds pe
language; max. sco e =10). In he Wo d T ansla ion ask, pa ien s needed o e bally supply he
ansla ion o a wo d spoken by he examine (10 wo ds pe language; max. sco e =20). Inc easing
Beha . Sci. 2020,10, 141 16 o 23
mo e e o s. Due o his aspec o expe imen al design ha can be capi alized upon by oppo unis ic
swi ching, as i is unde s andable o bilinguals o gene ally demons a e an accu acy ad an age in
dual-naming condi ions.
In addi ion o eplica ing he de B uin e al. indings, we sough o u he explo e olun a y
swi ching wi h bo h cogna e and non-cogna e wo ds. As in p e ious s udies o Ca alan-Spanish
bilinguals [
44
] as well as s udies in ol ing o he bilingual popula ions [
45
], ou bilingual pa icipan s
in Expe imen 1 showed an o e all cogna e acili a ion e ec in educed naming la encies and
inc eased accu acy o cogna es compa ed o non-cogna es. This cogna e e ec did no in e ac wi h
swi ching and mixing cos s in naming la encies, bu cogna es did in luence accu acy ac oss ial ypes,
whe e non-cogna e, single-language ials we e named wi h he wo s accu acy. This in e ac ion
seems o indica e ha he cogna e e ec is mo e po en when bo h languages a e held in a s a e o
gene alized ac i a ion (dual-language condi ion) and how cogna e phonological simila i y is mo e
eadily accessible in his condi ion e sus he single-language condi ion, whe e only one language is
he a ge o ac i a ion. Ou s udy hus adds o he li e a u e o olun a y swi ching by showing ha
cogna e wo ds could p o ide g ea e acili a ion o co ec esponses when wo languages a e globally
engaged and olun a ily oggled be ween by a bilingual.
Viewed in he con ex o he Adap i e Con ol Hypo hesis (ACH; [
4
]) and he dual-mechanisms o
con ol (DMC) amewo k [
39
,
40
], he mixing bene i seen he e in Expe imen 1 could e lec a highly
ained p oac i e con ol sys em wi hin BLC, whe e young bilinguals who in e ac in dual-language
en i onmen s a e able o main ain hei languages “a he eady” wi h li le di icul y. Fu he mo e,
his mixing bene i indica es ha he global ac i a ion o wo languages has been mas e ed o he poin
whe e i is mo e cos ly o ha e o selec i ely inhibi one language o pe o m in a single-language
con ex . Finally, as he swi ch cos s in Expe imen 1 mi o hose seen in simila olun a y swi ching
s udies [
18
,
19
,
38
] and hose in cued-swi ching s udies, engaging eac i e con ol mechanisms seems
o lead o empo al cos s o he a e age bilingual, unless expe imen al manipula ions o indi idual
s a egies a e used o speci ically a o bo om-up swi ching [37].
In Expe imen 2, we sough o apply he olun a y language swi ching pa adigm o he s udy
o bilinguals wi h aphasia (BWAs) and how hei oli ional swi ching beha io compa ed wi h ha
o age-ma ched con ols. Due o a high deg ee o a iabili y in his pa ien popula ion on language
pe o mance measu es and a modes sample size o bo h expe imen al g oups, ou analyses included
g oup-le el compa isons as well as indi idual-le el analyses o BWAs. Wi h hese wo laye s o
in e p e a ion, we discuss a eas o commonali y ac oss g oups as well as he he e ogenei y o indi idual
pe o mances in ou pa ien sample.
A he g oup le el, BWAs we e able o swi ch wi h he same equency be ween hei languages
as con ols du ing he dual-language blocks, a inding con a y o ou p edic ions a he onse
o his expe imen . Ini ially, we had p edic ed ha pa ien s wi h aphasia would pe o m ewe
swi ches han con ols when gi en he op ion o choose hei language, based on p e ious indings
showing he p esence o con ol de ici s in BWAs [
57
,
58
], especially o con lic moni o ing [
32
].
Al hough he e a e cases o pa hological and uncon olled swi ching in pa ien s wi h b ain lesions
(e.g., [
22
,
24
,
25
,
27
,
31
,
59
,
60
]), ou BWA g oup demons a es ha no all BWA pa ien s a e impai ed in
hei abili y o olun a ily swi ch be ween hei languages. One possible explana ion o his is ha he
pa ien s included in his s udy we e all conside ed o ha e mild o mode a e aphasia and did no epo
signi ican ly highe a es o unin ended o in olun a y swi ching beha io on he BSWQ compa ed o
con ols. Mo eo e , some o he p e ious s udies ha ha e documen ed pa hological language mixing
and swi ching in pa ien s ha e sugges ed ha his beha io is mainly due o damaged subco ical
a eas (le cauda e, [
24
]; [
31
]) o wi hin he on o-s ia al sys em, as he BLC model would p edic o
language ac i a ion and selec ion [
1
]. The e o e, his i s esul could indica e ha ou BWAs ha e
spa ed unc ionali y in hese b ain a eas o a leas did no sus ain damage o he poin o a ec ing hei
language swi ching a a clinical le el; howe e , his is pu ely specula i e gi en ha we did no ha e
access o neu oimaging da a de ailing pa ien s’ speci ic lesion loca ions. To conclude whe he BWAs
Beha . Sci. 2020,10, 141 17 o 23
had comple ely spa ed BLC, we explo ed whe he hey also exhibi ed spa ed eac i e and p oac i e
con ol as a g oup below and how hey beha ed a an indi idual le el in he nex sec ion.
In gene al, BWAs we e o e all slowe and less accu a e in naming han con ols, a esul ha
we ha e p e iously obse ed in his clinical popula ion on naming asks [
32
,
61
]. Addi ionally, he
esponse du a ions gi en by BWAs we e signi ican ly la ge han hose o con ol, bu his ac o did no
in e ac wi h o he expe imen al a iables; his inding, along wi h o he lack o clinically signi ican
mo o speech diso de s, allows us o subsequen ly disca d a icula o y p ocesses as ac o ing in o
language swi ching pe o mance o his BWA sample. Fu he mo e, as in Expe imen 1, cogna e
s a us was examined he e as a ac o in swi ching beha io . In e es ingly, he e was a g ea e cogna e
acili a ion in bo h naming la encies and accu acy o BWAs compa ed o con ols. Adding o o he
s udies which ha e also ound his cogna e e ec o bilinguals wi h aphasia [
47
,
62
], his inding is
pa icula ly ele an o p oposed cogna e-based in e en ions in he clinical ealm [
63
], as cogna e
e ec s ound in a olun a y swi ching ask ca y mo e ecological alidi y in day- o-day naming o
bilinguals and hus indica e mo e po en ial o gene alizabili y in cogna e bene i s.
Fo eac i e con ol, we did no ind a signi ican di e ence be ween he wo g oups. Indeed, bo h
pa ien s and con ols collec i ely expe ienced swi ch cos s, whe e swi ch ials we e named slowe
han epea ials. Sepa a ed by g oups, con ols su e ed clea swi ch cos s ac oss all le els o cogna e
s a us and language, wi h he excep ion o non-cogna es in hei non-dominan language. This impac
o swi ching, oge he wi h he simila swi ching esul s om Expe imen 1, sugges s a pe asi e cos
o swi ching languages in his ask [
18
,
19
], whe e disengaging and engaging wi h ano he language
equi es ac i a ion o eac i e con ol p ocesses and, ba ing speci ic s a egies o ci cum en cos s,
leads o slowe naming esponses o bilinguals.
Fo p oac i e con ol, he only signi ican e ec o naming in a dual-language con ex s.
single-language con ex was a mixing cos , whe e con ols named cogna es in hei non-dominan
language slowe du ing epea ials; all o he mixing e ec s ac oss languages, ial ypes, and cogna e
s a uses o con ols we e non-signi ican . BWAs as a g oup did no show any signi ican mixing
e ec s. Thus, ou esul s seem o indica e ha global e ec s o in a-block language mixing a e nei he
bene icial no cos ly o balanced adul bilinguals. Mixing e ec s and p oac i e con ol a e u he
examined on he indi idual le el o pa ien s in he ollowing sec ion.
Accu acy on he language swi ching ask o Expe imen 2 showed a simila pa e n as seen
in Expe imen 1, whe e single language ials we e named less accu a ely han epea ials in he
dual-language naming condi ion; howe e , his di e ence was mo e p onounced in he BWA g oup.
While his pa e n is again likely a p oduc o pa icipan s ha ing he op ion o use ei he language
as a “li eline” in he dual-language condi ion bu no in he single-language condi ion, BWAs’ la ge
dec ease in single-language accu acy migh also e lec de ici s in e ie ing a ge language wo ds.
In e es ingly, i we look a he e o dis ibu ion o bo h g oups in he single-language blocks, BWAs
showed a ma ked inc ease in c oss-language in usions, he mos equen ype o e o o his
g oup (an a e age o 5.36% o ials o BWAs and 0.67% o ials o con ols in single-language
blocks; see Table 4) behind omission e o s. Fu he mo e, combining c oss-language in usions wi h
c oss-language au o-co ec ions (see E o Dis ibu ion sec ion o de ini ion) unde he umb ella e m
‘c oss-language e o s,’ he inc eased equency o hese e o s in BWAs (5.76% o all single-language
ials) could be d i ing he g ea e di e ence ac oss ial ypes in g oup accu acy.
While he aim o his s udy was no o compa e olun a y language swi ching pe o mance ac oss
di e en bilingual age g oups, esul s om young adul bilinguals in Expe imen 1 and he con ol
g oup om Expe imen 2 did di e in some impo an espec s. Conside ing he e ec o language
mixing in dual-language epea ials e sus single-language ials, young bilinguals demons a ed a
mixing bene i while olde bilinguals did no . This di e ence con as s wi h he indings o de B uin,
Samuel, and Duñabei ia [
19
], whe e olde bilingual adul s, like hei younge coun e pa s, exhibi ed
a mixing bene i . Addi ionally, while ou young bilinguals showed a posi i e e ec o cogna es on
naming accu acy, we we e unable o ind his e ec in he olde bilinguals. Bo h o hese disc epancies
Beha . Sci. 2020,10, 141 18 o 23
in pe o mance dese e o be explo ed in u u e s udies bu , as ou ocus he e emains on esul s
s emming om pa ien s wi h aphasia, his analysis alls ou side he scope o his s udy.
4.2. Indi idual Le el Analyses o Pa ien s
Zooming in on he indi idual le el, he complexi ies o language swi ching in pa ien s become
mo e salien . In ou sample, indi idual BWAs showed a leas wo ypes o naming pe o mance in
hei accu acy: some BWAs su e ed om c oss-language e o s in he single naming condi ion in a
much mo e subs an ial way (P 1 and 5) while mos o hem did no p oduce such e o s o p oduced
e y ew (P 2, P 3, P 4, P 6, and P 7). In ligh o his, we a gue ha in e p e a ions o esul s hold
mo e alidi y i we ake in o accoun hese wo p o iles sepa a ely, as hey a e indica i e o di e en
de ici s in he con ol mechanisms unde lying language swi ching.
Wi hin e o ype dis ibu ions, he p esence o mo e c oss-language e o s in he single naming
condi ion is po en ially an indica ion o some BLC de ici s. Reminiscen o how pa ien s wi h
pa hological language swi ching (e.g., [
22
]) pe o m bu o a lesse deg ee, ce ain BWAs we e unable
o es ic he lexicaliza ion o ha speci ic language du ing he ask and su e ed in usions (whe he
au o-co ec ed o no ) om he unin ended language. Howe e , hese pa ien s did no ha e di icul ies
in p oac i e con ol as hey we e able o main ain he wo languages wi h no mixing e ec s when
hey named i ems in he dual-language si ua ion. Looking a he da a o hese wo pa ien s, P 1
and P 5, we can see ha hey bo h ha e mixing pe o mance wi hin he no mal ange as compa ed
o heal hy con ols. Wi h his ype o pe o mance, we could specula e ha hese pa ien s had some
speci ic de ici s wi hin goal main enance o in e e ence con ol, elemen s o BLC p oposed by G een
and Abu alebi [
4
] in he ACH; hei c oss-language e o s sugges di icul ies in main aining he
goal o speaking in jus one language o o supp essing he c oss-language in e e ence om he
unin ended language. Howe e , de ici s o he o he con ol mechanisms (salien cue de ec ion,
selec i e esponse inhibi ion, ask disengagemen , ask engagemen , and oppo unis ic planning)
p oposed by his amewo k canno explain he beha io o hese pa ien s, as hese de ici s would
impac he dual-language naming condi ions as well.
Al e na i ely, we migh also a gue ha he p esence o c oss-language e o s in he con ex o
aphasia deno es a s a egy o compensa ing o wo d e ie al de ici s in a a ge language [
64
] which
we a e only able o de ec when p oduc ion is es ic ed o said language [
22
]; adop ing his echnique
could ha e clea bene i s in dual-language con ex s bu gene aliza ion o single-language con ex s
would lead o he obse ed le els o c oss-language e o s. E idence o his in e p e a ion comes
om signi ican ly highe a ings o DL and NDL swi ching o pa ien s compa ed o con ols on he
BSWQ. The ques ions comp ising hese wo subscales add ess swi ching in o he o he language when
he a ge language is inaccessible (i.e., “When I canno ecall a wo d in Ca alan, I end o immedia ely
p oduce i in Spanish”; [
48
]) and hus highe a ings would indica e mo e ins ances o compensa o y
language swi ching when he a ge language is inaccessible. This ype o s a egy o c oss-language
acili a ion o cueing wi h he non- a ge language has been p oposed as a po en ially olun a y o
in olun a y ac ic used by BWAs [
65
]. Howe e , i is impo an o highligh ha his compensa ion is
no necessa ily ee o swi ch cos s. Indeed, P 5 had signi ican ly la ge swi ching cos s as compa ed o
con ols, sugges ing ha , despi e he pa ien ’s abili y o compensa e wi h he o he language o some
measu es, hei eac i e con ol was no so e ec i e as in heal hy indi iduals. O e all, expe imen al
esul s om hese wo pa ien s could indica e a g adien in se e i y o pa hological language swi ching,
whe e bilinguals can expe ience clea impai men s in language- es ic ed lexical e ie al wi hou
eaching a poin o uncon olled o pa hological language swi ching.
In e e ence o he clinical p o iles o hese wo pa ien s, key simila i ies and di e ences shed
ligh on he ela ionship be ween diagnos ic assessmen s and expe imen al pe o mance. Based on
hei WAB sco es, bo h P 1 and P 5 we e classi ied as mode a e in e ms o he se e i y o hei
language diso de . Howe e , hei ypes o aphasia we e pola opposi es: P 1 demons a ed de ici s in
audi o y comp ehension and epe i ion bu luid speech, compa ible wi h We nicke’s aphasia, while P
Beha . Sci. 2020,10, 141 19 o 23
5 showed a non- luen pa e n o speech bu p ese ed comp ehension and epe i ion, cha ac e is ics
o ansco ical mo o aphasia. Thus, while he WAB’s classi ica ions o aphasia may help o o ien
diagnosis and ea men s a egies in he clinical se ing o BWAs, hei ype o aphasia does no
necessa ily align wi h p ese a ion o impai men o BLC; a he , he o e all se e i y and pe asi eness
o de ici s may be e indica e whe he hei con ol o e hei di e en languages is also impac ed.
The subsequen se o da a, s emming om he emaining pa ien s wi hou p oblems in main aining
a language in he single-language condi ion, can be in e p e ed in e ms o associa ions and dissocia ions
be ween de ici s in language con ol mechanisms. Among hese pa ien s, wo showed pa allel impac s
in said mechanisms. One pa ien (P 3) p oduced mixing and swi ching e ec s wi hin no mal anges
as compa ed o con ols bu hei poo accu acy sugges ed a gene al p oblem in lexical e ie al;
his pa e n o pe o mance is compa ible wi h he pa ien ’s diagnosis o mode a e anomic aphasia.
In e es ingly, al hough P 3 commi ed a la ge numbe o omission e o s ac oss ials (see Table 5), she
did so wi hou any clea de ici in language ask engagemen o disengagemen , as she was able o
manage c oss-language in e e ence and pe o m swi ches du ing he dual-language condi ion. These
esul s clea ly sugges ha wo d p oduc ion de ici s in bilinguals a e o some ex en dissocia ed om
he de ici s in con olling he wo languages [
23
], as he neu al models o bilingualism would also
sugges . Al hough hese wo sys ems o e lap in a numbe o b ain a eas [
2
,
5
,
66
], some a e speci ic o
language con ol and he absence o language con ol de ici s in BWAs could indica e ha said a eas
a e spa ed om b ain damage. A he o he end o he spec um, we also saw an example whe e bo h
eac i e and p oac i e con ol a e a ec ed in he case o P 2. As we ha e shown in p e ious s udies
wi h clinical popula ions [
20
,
21
], bo h con ol sys ems can be a ec ed when pa ien s a e asked o
pe o m he swi ching ask wi h he wo languages, despi e ha ing some spa ed abili ies o swi ching
in he non-linguis ic domain. He e, his gene alized impac migh be explained by di e en e iology
ha he pa ien had compa ed o he o he s; his pa ien ’s language de ici s de eloped ollowing a
umo esec ion whe eas he es o he pa ien s had pos -s oke aphasia. While his is again la gely
specula i e gi en ha we did no ha e ex ensi e in o ma ion on he ex en o he esec ion, we migh
hypo hesize ha he dis inc ype o b ain damage is he sou ce o such wide-sp ead pa hological
pe o mance in bo h domains o language con ol.
Be ween hese wo ex emes o pa allel e ec s, we ha e h ee pa ien s who show dissocia ion
be ween eac i e and p oac i e de ici s. Despi e i being di icul o in e p e wha d i es pa ien s o
ha e p oblems in one domain o he o he , he esul s seem o sugges ha in some cases hese wo
con ol mechanisms a e no a comple ely uni ied sys em. P 6 and 7 had de ici s in p oac i e bu no
in eac i e con ol, whe eas P 4 showed a la ge bene i while naming pic u es in he dual-language
condi ion bu no signi ican de ia ion om con ols in eac i e con ol. These ypes o dissocia ions a e
no no el indings, as p e ious s udies ha e desc ibed hem in some pa hological condi ions [
46
] and in
heal hy indi iduals. Fo ins ance, Seo and P a [
67
] ha e ecen ly ound ha hese p ocesses may ely
on di e en b ain a eas, whe e p oac i e con ol is mo e associa ed wi h ac i a ion o he do sola e al
p e on al co ex and p emo o co ex and eac i e con ol wi h ac i a ion in he an e io cingula e
in bilinguals. Ma, Li, and Guo [
17
] showed ha p oac i e and eac i e con ol a e di e en ially
a ec ed by ask condi ions (p epa a ion ime) and hei in ol emen depends on pa icipan s’ language
dominance. Fu he mo e, De B uin e al. [
18
,
19
] ha e shown ha when indi iduals a e ee o swi ch
a will in dual-language condi ions, hey ha e bene i s in global language mixing bu hen exhibi cos s
o local swi ches be ween he wo languages, sugges ing a ce ain deg ee o dissocia ion be ween
hese wo con ol mechanisms.
Acco ding o he DMC amewo k p oposed by B a e [
39
], p oac i e con ol helps indi iduals
o main ain wo asks ac i e in dual condi ion, simila o he concep o a wo king memo y sys em.
In he con ex o bilingualism, i has been p oposed ha p oac i e con ol in ol ed in main aining
he ac i a ion o he wo languages in hose con ex s in which bilinguals a e equi ed o swi ch [
20
].
Hence, we migh ini ially in e p e inc eased mixing cos s in hese wo pa ien s (P 6 and P 7) as
a consequence o a de ici in dealing wi h he ac i a ion o wo languages, speci ically in con lic
Beha . Sci. 2020,10, 141 20 o 23
moni o ing as Abu alebi and G een [
5
] sugges in hei amewo k. Howe e , since he pa ien s did
no show any de ici o eac i e con ol (demons a ing simila swi ch cos s o heal hy con ols), we
ha e o conclude ha hey had p ese ed con lic esolu ion mechanisms as hey we e able o deal
wi h he c oss-language in e e ence in swi ch ials. Simila ly, we can exclude any majo di icul ies in
language engagemen o disengagemen as hey we e able o swi ch om one language o he o he a
he same a e as heal hy con ols.
Finally, simila o he wo pa ien s wi h ele a ed c oss-language e o s, hese i e emaining
BWAs do no seem o e eal a link be ween hei aphasia ypes and hei pe o mance on he olun a y
language swi ching ask. No ably, hei aphasia ypes do no coincide wi h one ano he (P 2:
conduc ion; P s 3, 4, and 7: anomic; P 6: We nicke) and ha e no clea ela ionship wi h associa ions
o dissocia ions in hei espec i e con ol de ici s; his is unde s andable gi en ha he WAB was
no designed o de ec de ici s in BLC. Howe e , while classi ica ion based on aphasia ype may no
coincide wi h pa e ns o BLC de ici s, he WAB’s index o se e i y may, wi h mo e se e e cases o
aphasia co esponding o g ea e di icul ies in language- es ic ed lexical e ie al. In con as o he
mode a e le els o aphasia sha ed by P 1 and P 5, ou ou o i e pa ien s in his second g oup, hose
pa ien s ha did no show de ici s in main aining languages, exhibi ed only mild o ms o aphasia.
Obse ing his eme ging pa e n, u u e s udies should con inue o explo e whe he se e i y o aphasia
co ela es wi h de ici s in main aining a single language while naming.
In sum, indi idual pa ien analyses indica e wo p o iles o BLC de ici s. The i s ype o
BWA he e is cha ac e ized by dec eased capaci ies in language goal main enance and/o in e e ence
supp ession and consequen ially commi ing la ge numbe s o c oss-language e o s when es ic ed
o one language. While hese pa ien s’ de ici s in BLC do no each a le el o signi icance o be classi ied
as “pa hological swi che s,” hey could be iewed as mild cases o pa hological language swi ching
ha become e iden in expe imen al pa adigms. The second, mo e nebulous ype o BWA appea s o
ha e p ese ed unc ionali y in language moni o ing bu exhibi s a iabili y in wha aspec s o BLC
a e impac ed. Gene ally, hese pa ien s do seem o e eal a ce ain dissocia ion be ween p oac i e and
eac i e language con ol p ocesses.
5. Conclusions
The p esen s udy aimed a in es iga ing he ole o p oac i e and eac i e con ol in he olun a y
language swi ching h ough he lens o bilingual aphasia. Resul s om young indi iduals eplica e
p e ious esea ch while hose om pa ien s wi h aphasia show a mo e complex pic u e ha equi ed
he in eg a ion o bo h g oup and indi idual le el analyses. Gi en he complexi y o ou esul s, i is
di icul o de e mine whe he aphasia a ec s BLC o no in all pa ien s because his migh depend
on se e al ac o s. Howe e , i is impo an o highligh wo main indings ha help bo h clinicians
and u u e esea ch. Fi s , i is essen ial o ame he in es iga ion o he BLC de ici s in pa ien s wi h
aphasia by using heo e ical amewo ks o cogni i e con ol [
5
,
39
], as hey allow us he oppo uni y
o desc ibe spa ed and a ec ed mechanisms wi h speci ici y. Second, a ine-g ained analysis o he
pe o mance o each pa ien is essen ial o iden i y BLC de ici p o iles as well as associa ions and
dissocia ion be ween he di e en language con ol mechanisms (p oac i e s. eac i e).
Au ho Con ibu ions:
The au ho s o his s udy each con ibu ed in he ollowing ways: Concep ualiza ion, M.C.;
me hodology, N.G., G.P. and M.C.; o mal analysis, N.G., G.P. and M.C.; in es iga ion, N.G., G.P., C.G.-S., and M.C.;
da a cu a ion, N.G. and G.P.; w i ing—o iginal d a p epa a ion, N.G., G.P., C.G.-S., and M.C.; w i ing— e iew
and edi ing, N.G. and M.C.; supe ision, C.G.-S. and M.C.; unding acquisi ion, M.C. All au ho s ha e ead and
ag eed o he published e sion o he manusc ip .
Funding:
This esea ch was suppo ed wi h unding p o ided o M.C. by he pos doc o al Ram
ó
n y Cajal
ellowship (RYC-2013-14013) as well as unding om Agencia Es a al de In es igaci
ó
n (AEI, Na ional Resea ch
Agency) and Fondo Eu opeo de Desa ollo Regional (FEDER, Eu opean Regional De elopmen Fund) unde
p ojec s PSI2017-87784-R and RED2018-102615-T.
Acknowledgmen s:
The au ho s would like o acknowledge and hank M
ó
nica Sando al-Gonz
á
lez o he help
in da a collec ion and analysis.
Beha . Sci. 2020,10, 141 21 o 23
Con lic s o In e es : The au ho s decla e no con lic o in e es .
Re e ences
1.
Abu alebi, J.; G een, D.W. Con ol mechanisms in bilingual language p oduc ion: Neu al e idence om
language swi ching s udies. Lang. Cogn. P ocess. 2008,23, 557–582. [C ossRe ]
2.
Calab ia, M.; Cos a, A.; G een, D.W.; Abu alebi, J. Neu al basis o bilingual language con ol. Ann. Acad. Sci.
2018,1426, 221–235. [C ossRe ] [PubMed]
3.
Decle ck, M.; Philipp, A.M. A e iew o con ol p ocesses and hei locus in language swi ching. Psychon.
Bull. Re . 2015,22, 1630–1645. [C ossRe ] [PubMed]
4.
G een, D.W.; Abu alebi, J. Language con ol in bilinguals: The adap i e con ol hypo hesis. J. Cogn. Psychol.
2013,25, 515–530. [C ossRe ] [PubMed]
5.
Abu alebi, J.; G een, D.W. Neu oimaging o language con ol in bilinguals: Neu al adap a ion and ese e.
Bilingualism 2016,19, 689–698. [C ossRe ]
6.
B anzi, F.M.; Della Rosa, P.A.; Canini, M.; Cos a, A.; Abu alebi, J. Language con ol in bilinguals: Moni o ing
and esponse selec ion. Ce eb. Co ex 2015,26, 2367–2380. [C ossRe ]
7.
B anzi, F.M.; Calab ia, M.; Bosca ino, M.L.; Cos a, A. On he o e lap be ween bilingual language con ol and
domain-gene al execu i e con ol. Ac a Psychol. 2016,166, 21–30. [C ossRe ]
8.
B anzi, F.M.; Ma in, C.D.; Abu alebi, J.; Cos a, A. The a e -e ec s o bilingual language p oduc ion.
Neu opsychologia 2014,52, 102–116. [C ossRe ]
9.
Cos a, A.; San es eban, M. Lexical access in bilingual speech p oduc ion: E idence om language swi ching
in highly p o icien bilinguals and L2 lea ne s. J. Mem. Lang. 2004,50, 491–511. [C ossRe ]
10.
Cos a, A.; San es eban, M.; I ano a, I. How do highly p o icien bilinguals con ol hei lexicaliza ion p ocess?
Inhibi o y and language-speci ic selec ion mechanisms a e bo h unc ional. J. Exp. Psychol. Lea n. Mem.
Cogn. 2006,32, 1057–1074. [C ossRe ]
11.
Guo, T.; Liu, H.; Mis a, M.; K oll, J.F. Local and global inhibi ion in bilingual wo d p oduc ion: MRI e idence
om Chinese-English bilinguals. Neu oimage 2011,56, 2300–2309. [C ossRe ] [PubMed]
12.
Jackson, G.M.; Swainson, R.; Cunning on, R.; Jackson, S.R. ERP co ela es o execu i e con ol du ing
epea ed language swi ching. Biling. Lang. Cogn. 2001,4, 169–178. [C ossRe ]
13.
Meu e , R.F.I.; Allpo , A. Bilingual Language Swi ching in Naming: Asymme ical Cos s o Language
Selec ion. J. Mem. Lang. 1999,40, 25–40. [C ossRe ]
14.
Baus, C.; B anzi, F.M.; Cos a, A. On he mechanism and scope o language con ol in bilingual speech
p oduc ion. Camb idge Handb. Biling. P ocess. 2015, 508–526. [C ossRe ]
15.
Cos a, A.; B anzi, F.M.;
Á
ila, C. Bilingualism: Swi ching. In Neu obiology o Language; Else ie : Ams e dam,
The Ne he lands, 2016; pp. 419–430.
16.
Ch is o els, I.K.; Fi k, C.; Schille , N.O. Bilingual language con ol: An e en - ela ed b ain po en ial s udy.
B ain Res. 2007,1147, 192–208. [C ossRe ] [PubMed]
17.
Ma, F.; Li, S.; Guo, T. Reac i e and p oac i e con ol in bilingual wo d p oduc ion: An in es iga ion o
in luen ial ac o s. J. Mem. Lang. 2016,86, 35–59. [C ossRe ]
18.
De B uin, A.; Samuel, A.G.; Duñabei ia, J.A. Volun a y language swi ching: When and why do bilinguals
swi ch be ween hei languages? J. Mem. Lang. 2018,103, 28–43. [C ossRe ]
19.
De B uin, A.; Samuel, A.G.; Duñabei ia, J.A. Examining bilingual language swi ching ac oss he li espan in
cued and olun a y swi ching con ex s. J. Exp. Psychol. Hum. Pe cep . Pe o m. 2020. [C ossRe ]
20.
Ca aneo, G.; Calab ia, M.; Ma ne, P.; Gi onell, A.; Abu alebi, J.; Cos a, A. The ole o execu i e con ol in
bilingual language p oduc ion: A s udy wi h Pa kinson’s disease indi iduals. Neu opsychologia
2015
,66,
99–110. [C ossRe ]
21.
Ca aneo, G.; Cos a, A.; Gi onell, A.; Calab ia, M. On he speci ici y o bilingual language con ol: A s udy
wi h Pa kinson’s disease pa ien s. Bilingualism 2020,23, 570–578. [C ossRe ]
22.
Calab ia, M.; Ma ne, P.; Rome o-Pinel, L.; Juncadella, M.; Cos a, A. Losing con ol o you languages: A case
s udy. Cogn. Neu opsychol. 2014,31, 266–286. [C ossRe ] [PubMed]
23.
Calab ia, M.; Je e ies, E.; Sala, I.; Mo enas-Rod
í
guez, E.; Ill
á
n-Gala, I.; Mon al, V.; Fo ea, J.; Lle
ó
, A.;
Cos a, A. Mul ilingualism in seman ic demen ia: Language-dependen lexical e ie al om deg aded
concep ual ep esen a ions. Aphasiology 2019. [C ossRe ]
Beha . Sci. 2020,10, 141 22 o 23
24.
Abu alebi, J.; Miozzo, A.; Cappa, S.F. Do subco ical s uc u es con ol ‘language selec ion’in polyglo s?
E idence om pa hological language mixing. Neu ocase 2000,6, 51–56.
25.
Aglio i, S.; Bel amello, A.; Gi a di, F.; Fabb o, F. Neu olinguis ic and ollow-up s udy o an unusual pa e n
o eco e y om bilingual subco ical aphasia. B ain 1996,119, 1551–1564. [C ossRe ]
26.
Aglio i, S.; Fabb o, F. Pa adoxical selec i e eco e y in a bilingual aphasic ollowing subco ical lesions.
Neu o epo An In . J. Rapid Commun. Res. Neu osci. 1993,4, 1359–1362. [C ossRe ]
27.
Ansaldo, A.I.; Saidi, L.G.; Ruiz, A. Model-d i en in e en ion in bilingual aphasia: E idence om a case o
pa hological language mixing. Aphasiology 2010,24, 309–324. [C ossRe ]
28.
Ga c
í
a-Caballe o, A.; Ga c
í
a-Lado, I.; Gonz
á
lez-He mida, J.; A ea, R.; Recimil, M.J.; Juncos Rabad
á
n, O.;
Lamas, S.; Ozai a, G.; Jo ge, F.J. Pa adoxical eco e y in a bilingual pa ien wi h aphasia a e igh
capsulopu aminal in a c ion. J. Neu ol. Neu osu g. Psychia y 2007,78, 89–91. [C ossRe ]
29.
Kong, A.P.H.; Abu alebi, J.; Lam, K.S.Y.; Weekes, B. Execu i e and language con ol in he mul ilingual b ain.
Beha . Neu ol. 2014,2014. [C ossRe ]
30.
Leemann, B.; Lagana o, M.; Schwi e , V.; Schnide , A. Pa adoxical swi ching o a ba ely-mas e ed second
language by an aphasic pa ien . Neu ocase 2007,13, 209–213. [C ossRe ]
31.
Ma iën, P.; Abu alebi, J.; Engelbo ghs, S.; De Deyn, P.P. Pa hophysiology o language swi ching and mixing
in an ea ly bilingual child wi h subco ical aphasia. Neu ocase 2005,11, 385–398. [C ossRe ]
32.
Calab ia, M.; G unden, N.; Se a, M.; Ga c
í
a-S
á
nchez, C.; Cos a, A. Seman ic P ocessing in Bilingual Aphasia:
E idence o Language Dependency. F on . Hum. Neu osci. 2019,13, 1–15. [C ossRe ] [PubMed]
33.
Calab ia, M.; G unden, N.; Iaia, F.; Ga c
í
a-S
á
nchez, C. In e e ence and acili a ion in phonological encoding:
Two sides o he same coin? E idence om bilingual aphasia. J. Neu olinguis . 2020,56, 100935. [C ossRe ]
34.
Abu alebi, J.; G een, D. Bilingual language p oduc ion: The neu ocogni ion o language ep esen a ion and
con ol. J. Neu olinguis . 2007,20, 242–275. [C ossRe ]
35.
G een, D.W.; Abu alebi, J. Unde s anding he link be ween bilingual aphasia and language con ol. J.
Neu olinguis . 2008,21, 558–576. [C ossRe ]
36.
Gollan, T.H.; Fe ei a, V.S. Should I s ay o should I swi ch? A cos –bene i analysis o olun a y language
swi ching in young and aging bilinguals. J. Exp. Psychol. Lea n. Mem. Cogn. 2009,35, 640–665. [C ossRe ]
37.
Gollan, T.H.; Kleinman, D.; Wie enga, C.E. Wha ’s Easie : Doing wha you wan , o being old wha o do?
Cued e sus olun a y language and ask swi ching. J. Exp. Psychol. Gen. 2014,143, 2167–2195. [C ossRe ]
38.
Je o i´c, M.; Duñabei ia, J.A.; De B uin, A. How Do Bilinguals Swi ch be ween Languages in Di e en In e ac ional
Con ex s? A Compa ison be ween Volun a y and Manda o y Language Swi ching; Camb idge Uni e si y P ess:
Camb idge, UK, 2020; pp. 1–46.
39.
B a e , T.S. The a iable na u e o cogni i e con ol: A dual-mechanisms amewo k Shi ing he emphasis
o a iabili y in cogni i e con ol. T ends Cogn. Sci. 2012,16, 106–113. [C ossRe ]
40.
De Pisapia, N.; B a e , T.S. A model o dual con ol mechanisms h ough an e io cingula e and p e on al
co ex in e ac ions. Neu ocompu ing 2006,69, 1322–1326. [C ossRe ]
41.
Kleinman, D.; Gollan, T.H. Speaking Two Languages o he P ice o One: Bypassing Language Con ol
Mechanisms ia Accessibili y-D i en Swi ches. Psychol. Sci. 2016,27, 700–714. [C ossRe ]
42.
Calab ia, M.; B anzi, F.M.; Ma ne, P.; He n
á
ndez, M.; Cos a, A. Age- ela ed e ec s o e bilingual language
con ol and execu i e con ol. Bilingualism 2015,18, 65–78. [C ossRe ]
43.
Calab ia, M.; He n
á
ndez, M.; B anzi, F.M.; Cos a, A. Quali a i e di e ences be ween bilingual language
con ol and execu i e con ol: E idence om ask-swi ching. F on . Psychol. 2011,2, 1–10. [C ossRe ]
44.
Cos a, A.; Ca amazza, A.; Sebas ian-Galles, N. The Cogna e Facili a ion E ec : Implica ions o Models o
Lexical Access posi ha only he lexical node selec ed in he i s s age. J. Exp. Psychol. Lea n. Mem. Cogn.
2000,26, 1283–1296. [C ossRe ] [PubMed]
45.
Hoshino, N.; K oll, J.F. Cogna e e ec s in pic u e naming: Does c oss-language ac i a ion su i e a change
o sc ip ? Cogni ion 2008,106, 501–511. [C ossRe ] [PubMed]
46.
Dash, T.; Ka , B.R. Bilingual language con ol and gene al pu pose cogni i e con ol among indi iduals
wi h bilingual aphasia: E idence based on nega i e p iming and lanke asks. Beha . Neu ol.
2014
,2014.
[C ossRe ]
47.
Robe s, P.M.; Deslau ie s, L. Pic u e naming o cogna e and non-cogna e nouns in bilingual aphasia.
J. Commun. Diso d. 1999,32, 1–23. [C ossRe ]
Beha . Sci. 2020,10, 141 23 o 23
48.
Rod iguez-Fo nells, A.; K äme , U.M.; Lo enzo-Se a, U.; Fes man, J.; Mün e, T.F. Sel -assessmen o indi idual
di e ences in language swi ching. F on . Psychol. 2012,3, 1–15. [C ossRe ]
49.
Snodg ass, J.G.; Vande wa , M. A s anda dized se o 260 pic u es: No ms o name ag eemen , image
ag eemen , amilia i y, and isual complexi y. J. Exp. Psychol. Hum. Lea n. Mem.
1980
,6, 174–215. [C ossRe ]
50.
Sebas ian-Galles, N.; Cue os, F.; Ma
í
, M.A.; Ca ei as, M. LEXESP: L
é
xico in o ma izado del español; Edicions
de la Uni e si a de Ba celona: Ba celona, Spain, 2000.
51.
Ra el i Fon anals, J. Dicciona i de F eqü
è
ncies: Llengua no Li e
à
ia; Ins i u d’Es udis Ca alans: Ba celona,
Spain, 1996.
52.
Fo s e , K.; Fo s e , J. DMDX: A Windows display p og am wi h millisecond accu acy senso es_psico
use _log. Beha . Res. Me hods Ins um. Compu . 2003,35, 116–124.
53.
P o opapas, A. CheckVocal: A p og am o acili a e checking he accu acy and esponse ime o ocal
esponses om DMDX. Beha . Res. Me hods 2007,39, 859–862. [C ossRe ]
54.
Schwa z, M.F. Theo e ical analysis o wo d p oduc ion de ici s in adul aphasia. Philos. T ans. R. Soc. B Biol.
Sci. 2014,369. [C ossRe ]
55. Ke esz, A.; Pascual-Leone, A.; Pascual-Leone, G. Ba e ía de las a asias Wes e n. Wes . Aphasia Ba e . 1990.
56.
Pa adis, M.; Libben, G. The Assessmen o Bilingual Aphasia; Law ence E lbaum Associa es: Hillsdale, NJ, USA,
1987.
57.
G ay, T.; Ki an, S. The ela ionship be ween language con ol and cogni i e con ol in bilingual aphasia.
Bilingualism 2016,19, 433–452. [C ossRe ]
58.
Fa oqi-Shah, Y.; Sampson, M.; P ange , M.; Baughman, S. Cogni i e con ol, wo d e ie al and bilingual
aphasia: Is he e a ela ionship? J. Neu olinguis . 2018,45, 95–109. [C ossRe ]
59.
Ansaldo, A.I.; Joane e, Y. Language mixing and language swi ching ollowing aphasia. In Ad ances in he
Neu olingusi ics o Bilingualism: Essays in Hono o Michel Pa adis; Fab o, F., Ed.; Fo um: Udine, I aly, 2002;
pp. 261–274.
60.
Ansaldo, A.I.; Ma co e, K. Language swi ching in he con ex o Spanish-English bilingual aphasia.
In Communica ion Diso de s in Spanish Speake s: Theo e ical, Resea ch, and Clinical Aspec s; Cen eno, J.G.,
Ande son, R.T., Oble , L.K., Eds.; Mul ilingual Ma e s: Cle edon, UK, 2007; pp. 214–230.
61.
Calab ia, M.; He n
á
ndez, M.; Ca aneo, G.; Suades, A.; Se a, M.; Juncadella, M.; Reñ
é
, R.; Sala, I.; Lle
ó
, A.;
O iz-Gil, J.; e al. Ac i e bilingualism delays he onse o mild cogni i e impai men . Neu opsychologia
2020
,
107528. [C ossRe ] [PubMed]
62.
Ve ey , N.; De Le e , M.; Hemelsoe , D.; San ens, P.; Duyck, W. Cogna e e ec s and execu i e con ol in a
pa ien wi h di e en ial bilingual aphasia. Appl. Neu opsychol. 2013,20, 221–230. [C ossRe ] [PubMed]
63.
Kohne , K. Cogni i e and cogna e-based ea men s o bilingual aphasia: A case s udy. B ain Lang.
2004
,
91, 294–302. [C ossRe ]
64.
Ansaldo, A.I.; Ma co e, K.; Sche e , L.; Raboyeau, G. Language he apy and bilingual aphasia: Clinical
implica ions o psycholinguis ic and neu oimaging esea ch. J. Neu olinguis . 2008,21, 539–557. [C ossRe ]
65.
Go al, M.; Le y, E.S.; Oble , L.K.; Cohen, E. C oss-language lexical connec ions in he men al lexicon:
E idence om a case o ilingual aphasia. B ain Lang. 2006,98, 235–247. [C ossRe ]
66.
Luk, G.; G een, D.W.; Abu alebi, J.; G ady, C. Cogni i e con ol o language swi ching in bilinguals: A
quan i a i e me a-analysis o unc ional neu oimaging s udies. Lang. Cogn. P ocess.
2012
,27, 1479–1488.
[C ossRe ]
67.
Seo, R.; P a , C.S. P oac i e and eac i e language con ol in he bilingual b ain. B ain Sci.
2019
,9, 161.
[C ossRe ]
©
2020 by he au ho s. Licensee MDPI, Basel, Swi ze land. This a icle is an open access
a icle dis ibu ed unde he e ms and condi ions o he C ea i e Commons A ibu ion
(CC BY) license (h p://c ea i ecommons.o g/licenses/by/4.0/).
