A c oss linguis ic s udy on
o hog aphic in luence du ing
audi o y wo d ecogni ion
Albe o Fu goni1, Cla a D. Ma in1,2 & An je S oeh 1,2
Lea ning o ead a ec s speech pe cep ion. Fo example, he abili y o lis ene s o ecognize
consis en ly spelled wo ds as e han inconsis en ly spelled wo ds is a obus inding called he
O hog aphic Consis ency E ec (OCE). P e ious s udies loca ed he OCE a he ime le el and ocused
on languages wi h opaque o hog aphies. This s udy in es iga es whe he he OCE also eme ges
a he phonemic le el and is a gene al phenomenon o languages wi h alphabe ic sc ip s, including
hose wi h anspa en w i ing sys ems. Thi y F ench (opaque language) and 30 Spanish ( anspa en
language) lis ene s pa icipa ed in an audi o y lexical decision ask ea u ing wo ds and pseudowo ds
including ei he only consis en ly spelled phonemes o also inconsis en ly spelled phonemes. Ou
esul s e ealed an OCE in bo h F ench and Spanish which su aced as longe eac ion imes in
esponse o inconsis en ly spelled wo ds and pseudowo ds. Howe e , when analyzing he da a spli by
language, he OCE was only de ec able in F ench bu no in Spanish. Ou indings ha e wo heo e ical
implica ions. Fi s , hey show ha audi o y lexical p ocessing is impac ed by o hog aphic in o ma ion
ha is e ie ed a he phonemic le el, no jus he ime le el. Second, hey sugges ha he OCE may
be modula ed by a language’s opaci y. In conclusion, ou s udy highligh s he dep h o li e acy e ec s
on audi o y language p ocessing and calls o u he in es iga ions in ol ing highly anspa en
languages.
Keywo ds Audi o y wo d p ocessing, O hog aphic consis ency e ec , O hog aphic dep h, F ench,
Spanish, Audi o y lexical decision ask
Reading and w i ing a e among he mos basic cogni i e skills o li e a e people. In e es ingly, lea ning o
ead and w i e in luences speech p ocessing. Fo ins ance, esea ch in o o hog aphic consis ency e ec s has
shown ha consis en (i.e., unambiguous) sound- o-spelling mappings impose less cogni i e e o in wo d
ecogni ion and p ocessing compa ed wi h inconsis en (i.e., ambiguous) mappings, a phenomenon known as
he O hog aphic Consis ency E ec (OCE)1–9. The OCE highligh s he in e play be ween w i en and spoken
language, illus a ing how p edic able o hog aphic pa e ns acili a e quicke and mo e eliable phonological
e ie al. In one o he i s s udies on he OCE, Seidenbe g and Tanenhaus10 demons a ed ha na i e English
lis ene s we e as e o iden i y audi o ily p esen ed wo d pai s as hyming when hey ma ched in spelling (e.g.,
globe and p obe) han when hey misma ched in spelling (e.g., ye and ie). The obus inding ha inconsis en
sound- o-spelling mappings in e e e in audi o y wo d p ocessing sugges s ha he o hog aphic ep esen a ion
o a wo d is au oma ically accessed du ing his p ocess. Unde s anding he OCE is c ucial o insigh s in o
eading luency, language acquisi ion, and he cogni i e p ocesses unde lying li e acy. I also has p ac ical
implica ions o educa ional s a egies and in e en ions aimed a imp o ing eading skills and add essing
eading di icul ies.
Building on ea lie s udies, on o hog aphic e ec s in he audi o y modali y10–12, Ziegle and Fe and9
employed an audi o y lexical decision ask o in es iga e whe he wo ds wi h inconsis en ly spelled imes a e
cogni i ely cos lie o p ocess han wo ds wi h consis en ly spelled imes. The ad an age o audi o y lexical
decision asks is ha hey di ec ly es lexical access and, unlike he wo d pai hyming ask, do no equi e
a p ime o a cue in o de o elici a po en ial OCE. In hei s udy, F ench lis ene s we e quicke and mo e
accu a e a ecognizing wo ds wi h consis en ly spelled a he han inconsis en ly spelled imes, con i ming ha
o hog aphy in luences audi o y wo d p ocessing.
The e a e di e en hypo heses on whe he he OCE is a p e-lexical o a pos -lexical e ec . On he one
hand, Ven u a and colleagues4 a gued in a o o he pos -lexical na u e o he OCE. They claimed ha lexical
access is necessa y o he OCE o occu , as hey obse ed no OCE in a shadowing ask (which elici s online
1Basque Cen e on Cogni ion, B ain and Language, Paseo Mikele egi 69, Donos ia-San Sebas ián 20009, Spain.
2Ike basque - Basque Founda ion o Science, Bilbao, Spain. email: [email p o ec ed]
OPEN
Scien i ic Repo s | (2025) 15:8374 1
| h ps://doi.o g/10.1038/s41598-025-92885-x
www.na u e.com/scien i ic epo s
p ocesses) bu only in a lexical decision ask. Howe e , i is impo an o no e ha sepa a ing pe cep ion and
p oduc ion p ocesses in he shadowing ask is challenging13. On he o he hand, Pe o a and colleagues14
a gued ha he OCE is an online p ocess because ime de ec ion asks inhe en ly equi e wo d segmen a ion.
Addi ionally, hey ound no signi ican di e ence in he magni ude o he OCE be ween he ime de ec ion and
lexical decision asks. Mo e consis en e idence on he p e-lexical na u e o he OCE comes om s udies wi h
elec ophysiological da a, as he e ec was obse ed in a ime window ela ed o p e-lexical p ocessing (i.e., <
400ms) when pa icipan s we e ca ying ou non-me aphonological asks (e.g., seman ic o lexical decision)15–17.
E en hough he e is consis en e idence in a o o he p e-lexical na u e o he OCE, i emains unclea why
his e ec is no consis en ly ound in pseudowo d p ocessing. As discussed la e , he p esen s udy in ends o
shed ligh on pseudowo d p ocessing s a ing om he assump ion ha he OCE is a p e-lexical e ec .
The use o a lexical decision ask enabled Ziegle and Fe and9 o also es o an OCE in pseudowo d
p ocessing. Al hough hey ound no e idence o an e ec , he OCE was ound in pseudowo d p ocessing in
a la e s udy, again wi h F ench lis ene s, by Pa amadilok and colleagues2. Conside ing ha pseudowo ds a e
necessa ily decoded phonemically, i is su p ising ha he e idence o an OCE in pseudowo d p ocessing is
weak. Two a gumen s ha e been pu o wa d o explain his inding9: Fi s , pseudowo ds do no ha e lexical
spellings and he e o e hey canno be as (in)consis en as eal wo ds.
Second, because lexical decision is a classi ica ion ask, pseudowo d s imulus i ems may be “ imed ou ” when
hey do no each an ac i a ion h eshold, which is why hey do no show some o he e ec s ha a e ypically
seen o wo ds.
Ano he possible explana ion may be ha he OCE does no a ec he phonemic le el du ing lexical
access bu , o ou knowledge, no s udy has in es iga ed i he OCE in audi o y wo d ecogni ion ope a es
a he phonemic le el (i.e., an inconsis en phoneme can be spelled in a ious ways, such as F ench / /=<
> o < ph> ) as well as he ime le el. Suppo o a phoneme-le el OCE is p o ided by s udies using isual
wo d ecogni ion asks, w i ing asks, and me alinguis ic asks such as phoneme dele ion, which ha e ound
o hog aphic e ec s on he phonemic le el. Fo example, in a s udy wi h F ench na i e speake s, al hough no
e idence o a phoneme-le el OCE was ound in a isual lexical decision ask, an OCE was elici ed by a w i ing
ask measu ing w i ing la encies and e o a es18. Mo eo e , in a la e s udy using a lexical decision ask, English
na i e speake s ecognized p in ed wo ds con aining only consis en phoneme- o-g apheme mappings mo e
quickly and accu a ely han wo ds con aining inconsis en phoneme- o-g apheme mappings19. Me alinguis ic
asks demons a e ha English na i e speake s manipula e indi idual consis en ly spelled phonemes mo e
e ec i ely han inconsis en ly spelled phonemes20. Simila ly, English na i e speake s ecognize consis en ly
spelled phonemes mo e accu a ely and as e han inconsis en ly spelled phonemes in wo d-ini ial posi ion21. In
sum, e idence sugges s ha o hog aphic (in)consis ency a ec s p ocessing a he phoneme le el, bu i emains
un es ed whe he phoneme-le el o hog aphic inconsis encies a ec audi o y wo d ecogni ion.
Ano he impo an issue ha has no been ho oughly add essed by p e ious esea ch is whe he and o wha
ex en speake s o anspa en languages wi h ew inconsis encies a e sensi i e o he a e i egula i ies in hei
spelling sys em du ing audi o y lexical p ocessing. So a , Po uguese is he only language wi h ai ly anspa en
phoneme- o-g apheme mappings o which he OCE has been es ed and demons a ed4,5. Howe e , Po uguese
s ill has a ela i ely ich phonological in en o y and comp ises conside ably mo e inconsis encies han, o
example, Spanish. The di e ence in o hog aphic anspa ency be ween Po uguese and Spanish becomes
e iden when compa ing child en who a e lea ning o ead in he wo languages: Be ween he i s and ou h
g ades (i.e., be ween six o en yea s o age), Spanish child en make ewe phonological e o s du ing eading
han g ade-ma ched Po uguese child en22. Thus, he p esen s udy explo es he OCE in a highly anspa en
language, namely Spanish.
Finally, i audi o y lexical p ocessing in anspa en languages is a ec ed by he OCE, a e he e any c oss-
linguis ic di e ences in he OCE in alphabe ic w i ing sys ems? Fo ins ance, Pa amadilok and colleagues2
eplica ed Ven u a and colleagues’4 indings on he OCE in Po uguese using F ench as he a ge language,
wi h bo h s udies inding he OCE in an audi o y lexical decision ask bu no in a shadowing ask. Ye , he OCE
a ec ed bo h wo d and pseudowo d p ocessing in F ench, while i only a ec ed wo d p ocessing in Po uguese4.
Taken oge he , hese di e ging esul s sugges ha na i e lis ene s o F ench– an opaque language – show mo e
in e e ence om (inconsis en ) o hog aphy in audi o y lexical p ocessing han na i e lis ene s o Po uguese
– a mo e anspa en language – which could explain why an OCE was also obse ed in F ench pseudowo ds.
Pa amadilok and colleagues2 a gued ha F ench lis ene s ely mo e on o hog aphic ep esen a ions in lexical
p ocessing because hey help wi h he selec ion o he co ec lexical en y (in F ench, /pɛ/ can e e o pain
‘b ead’, pin ‘pine’, peins ‘(you) pain ’, e c.). Al e na i ely, he link be ween phonology and o hog aphy migh be
s onge in F ench han in Po uguese because o di e en o hog aphic anspa ency and i s implica ions in
eading acquisi ion:
F ench-speaking child en need o ely on sound- o-spelling associa ions mo e han Po uguese-speaking
child en when hey lea n o ead. This could also explain why he OCE was obse ed in pseudowo d p ocessing
in F ench. I should be no ed ha F ench and Po uguese we e compa ed a he ime le el, and in wo sepa a e
s udies, which complica es a di ec compa ison. Building on he ini ial insigh s om hese s udies2,4, he e we
conduc a mo e sys ema ic c oss-linguis ic compa ison o unde s and how he OCE a ec s languages.
(1) a he phonemic le el and (2) depending on he opaci y o he spelling sys em.
The p esen s udy
The p esen s udy in es iga es whe he he OCE eme ges a he phonemic le el and whe he i is also p esen in
Spanish, an o hog aphically anspa en language. Speci ically, we asked he ollowing wo esea ch ques ions:
Scien i ic Repo s | (2025) 15:8374 2
| h ps://doi.o g/10.1038/s41598-025-92885-x
www.na u e.com/scien i ic epo s/
(1) Does he OCE eme ge a he phonemic le el? In he audi o y modali y, inconsis encies a e no limi ed o
ime uni s, bu a e also p esen a he phonemic le el: A single phoneme can o en be ep esen ed by mul-
iple g aphemes in w i ing (e.g., / / in F ench can be spelled ei he < > o < ph> ). We p edic ha a wo d
o pseudowo d con aining only consis en phonemes should be ecognized as e han one con aining also
inconsis en phonemes, gi en ha in he o me ins ance, each phoneme co-ac i a es only one o hog aph-
ic ep esen a ion (e.g., he F ench pseudowo d /bun/ can only ac i a e < boune> ). Con e sely, a wo d o a
pseudowo d con aining a leas some inconsis en phonemes should lead o cos lie p ocessing, su acing
as longe eac ion imes, i inconsis en phonemes co-ac i a e mul iple o hog aphic ep esen a ions (e.g.,
he F ench pseudowo d / ub/ could ac i a e bo h < oube > and < phoube> ).
(2) Does he OCE in luence audi o y wo d p ocessing in a anspa en language like Spanish simila ly o how i
impac s audi o y wo d p ocessing in an opaque language like F ench? Conside ing ha he li e a u e on he
OCE is la gely based on s udies using opaque languages, a di ec compa ison o he OCE elici ed in ans-
pa en s. opaque languages can e eal i i is gene alizable o a wide ange o alphabe ic languages. Based
on indings om Po uguese, he mos anspa en language s udied so a , we p edic ha he OCE will
be appa en in Spanish, e en hough i con ains only a ew inconsis en phoneme- o-g apheme mappings.
Thus, we hypo hesize ha e en in highly anspa en languages, inconsis en phonemes elici p ocessing
cos s, su acing as longe eac ion imes.
To answe hese ques ions, na i e lis ene s o F ench (a highly opaque language) and Spanish (a anspa en
language) pa icipa ed in an audi o y lexical decision ask in F ench and Spanish, espec i ely. The use o an
audi o y lexical decision ask was mo i a ed by ea lie esea ch, which sugges s ha his pa adigm can elici an
OCE wi hou employing any p ime o cue9. Spanish and F ench a e sui able o a c oss-linguis ic compa ison
because hey bo h display a simple syllabic complexi y23, allowing us o ocus on he phonemic le el.
Me hodology
Pa icipan s
Thi y na i e F ench lis ene s (17 emale, Mage= 21.83 yea s, SDage=2.13) and 30 na i e Spanish lis ene s (17
emale, Mage= 24.23 yea s, SDage=3.11) pa icipa ed in he s udy. All pa icipan s had uni e si y-le el educa ion.
Based on sel - epo , pa icipan s o bo h g oups had been exposed o hei na i e language since bi h and
hey all p edominan ly used hei na i e language in e e yday li e. No speci ic lea ning impai men s and no
hea ing o unco ec ed ision p oblems we e epo ed. F ench and Spanish pa icipan s we e ma ched on non-
e bal IQ (p = .371), as measu ed by he Kau man B ie In elligence Tes 24. The F ench coho was ec ui ed
om he Uni e si y o Bo deaux (F ance); Spanish pa icipan s we e ec ui ed om he pa icipan pool o
he Basque Cen e on Cogni ion, B ain and Language (Spain). The expe imen was app o ed by he Basque
Cen e on Cogni ion, B ain and Language’s E hics Commi ee [app o al numbe 031218D] and pe o med in
acco dance wi h he Decla a ion o Helsinki. This e hics app o al was also alid a he Uni e si y o Bo deaux
and no addi ional e hics app o al was equi ed. Pa icipan s p o ided w i en in o med consen p io o s a ing
he expe imen and ecei ed mone a y compensa ion o hei pa icipa ion.
Ma e ials
All ma e ials and in o ma ion on s imulus ma ching a e a ailable on he OSF. S imuli we e eco ded by male
na i e speake s o F ench and Spanish in a sound-a enua ing chambe using a Ma an z PMD 671 digi al
eco de wi h a Sennheise ME65 mic ophone and we e digi ized a 44.1kHz. All s imuli we e scaled o 65 dB
and a 50 ms in e al o silence was added o he beginning o each audio ile o allow o su icien loading ime
in he expe imen al so wa e. The s imulus lis o each language consis ed o 60 wo ds and 60 pseudowo ds.
Fo each lexicali y condi ion (wo ds; pseudowo ds), 50% o he s imuli con ained phonemes wi h only one
possible spelling (he ea e , consis en i ems). In he o he 50% o he s imuli, a leas he i s phoneme in
each i em could be spelled in mo e han one way (he ea e , inconsis en i ems). Inconsis en F ench i ems
con ained up o i e inconsis en phonemes, whe eas Spanish i ems con ained up o h ee. The disc epancy
be ween he wo languages was due o di e ences in he opaci y o F ench and Spanish: F ench wo ds commonly
include se e al inconsis en phonemes; Spanish is a anspa en language and has ewe wo ds wi h mul iple
inconsis en phonemes, especially when wo d equency and leng h a e con olled. All inconsis en phonemes
we e consonan s.
Pseudowo ds we e c ea ed om he wo d lis s by changing a leas he consonan s a he beginning o each
syllable, espec ing phono ac ic cons ain s (Table1).
O e all, consis en and inconsis en wo ds wi hin each language we e ma ched on a se ies o po en ially
con ounding a iables such as wo d equency, phonological and o hog aphic leng h, big am and biphone
equency, i s phoneme equency, neighbo hood densi y, and hei uniqueness poin wi hin each language.
C oss-linguis ically, wo ds we e ma ched on equency (Zip ’s log equency), numbe o le e s and he
F ench Spanish
Consis en Inconsis en Consis en Inconsis en
Wo ds po e/pɔʁ / laque/ lak/ u a/ u a/ b azo/bɾaθo/
Pseudowo ds /lɔʁ / /slas/ /lu sa/ /k axo/
Table 1. S imulus examples o F ench and Spanish. No e. Inconsis en phonemes highligh ed in bold.
Scien i ic Repo s | (2025) 15:8374 3
| h ps://doi.o g/10.1038/s41598-025-92885-x
www.na u e.com/scien i ic epo s/
du a ion o he audio eco dings. These a iables we e e ie ed om he Lexique da abase25 and om he EsPal
da abase26 o F ench and Spanish, espec i ely. The phonological and o hog aphic neighbo hood densi ies o
wo ds and pseudowo ds o bo h languages we e de e mined wi h CLEARPOND27. Consis en and inconsis en
pseudowo ds we e ma ched on hese a iables wi hin languages. Ac oss languages, pseudowo ds we e ma ched
on mean phonological neighbo hood.
Appa a us and p ocedu e
The expe imen was adminis e ed on an HP Eli eBook Folio 1040 G3 lap op compu e using OpenSesame
so wa e ( e sion 3.2.428). Audi o y s imuli we e p esen ed o e Sennheise GSP 350 headphones. The expe imen
was un in sound-a enua ing chambe s a he Uni e si y o Bo deaux and a he Basque Cen e on Cogni ion,
B ain and Language’s sa elli e labo a o y a he Uni e si y o he Basque Coun y in Donos ia-San Sebas ián.
Pa icipan s we e es ed on an audi o y lexical decision ask, in which hey had o espond ia key p ess
whe he each s imulus i em was a eal wo d o no . Pa icipan s we e ins uc ed o espond as quickly and as
accu a ely as possible. The espec i e keys we e labeled on he keyboa d; hal o he pa icipan s p essed he le
key o wo ds and he o he hal p essed he igh key o wo ds. Be o e s a ing he main pa o he expe imen ,
he pa icipan s comple ed a p ac ice phase wi h 10 ex a i ems (5 wo ds, 5 pseudowo ds). The pa icipan s only
ecei ed eedback on hei esponses in he p ac ice phase.
Du ing bo h he p ac ice phase and he main expe imen , each ial s a ed wi h a ixa ion poin appea ing
in he cen e o he sc een o 500 ms, which was ollowed by he audi o y s imulus. Reac ion ime measu emen
s a ed a s imulus onse and ended wi h he key p ess. A e pa icipan s ga e hei esponse, he nex ial
was au oma ically ini ia ed. The i ems we e p esen ed in andomized o de ac oss pa icipan s. The ask las ed
app oxima ely 15min.
Da a analyses
Da a and analysis sc ip s a e a ailable on he OSF. We conduc ed he s a is ical da a analyses in RS udio ( e sion
2023.12.1.40229) un on R ( e sion 4.2.030) and using he lme4 package ( e sion 1.1–2931). We ob ained p alues
o s a is ics h ough he lme Tes package ( e sion 3.1-332). We employed an alpha le el o 0.05 o in e p e ing
s a is ical signi icance. We in e se- ans o med eac ion imes as indica ed by he Box–Cox ans o ma ion33
pe o med using he MASS package ( e sion 7.3–5734), and emo ed da a poin s wi h s anda dized esiduals
mo e han 2.5 s anda d de ia ions om 0 as ou lie s using he LMERCon enienceFunc ions package35. We
an a linea mixed-e ec s model wi h in e se- ans o med eac ion imes as he dependen a iable. The model
had de ia ion coded ixed e ec s o Consis ency (consis en =-0.5; inconsis en = 0.5), Lexicali y (wo d=-0.5;
pseudowo d = 0.5), and Language (F ench=-0.5; Spanish = 0.5) wi h an in e ac ion e m including lowe -le el
in e ac ions. The model had andom in e cep s o Pa icipan and I em and by-Pa icipan andom slopes o
Consis ency and Lexicali y. Following Ven u a and colleagues5, we an one model including bo h wo ds and
pseudowo ds.
Due o echnical p oblems, wo Spanish pa icipan s did no ully comple e he ask, esul ing in he loss o
39 da a poin s. The wo F ench consis en pseudowo ds /my / and.
/sadʁ/ we e excluded om he analyses because hey elici ed accu acy a es below chance (1.68% o he
en i e da ase ). The emaining consis en and inconsis en pseudowo ds emained ma ched on all con ounding
a iables.
Resul s
Reac ion ime analyses we e conduc ed on co ec ly answe ed ials (6,840 ials; 96.32% o he da a). An
addi ional 147 da a poin s we e emo ed as ou lie s (2.15% o he da a). The model de ec ed signi ican e ec s
o Consis ency and Lexicali y, showing ha pa icipan s ook longe o espond o inconsis en i ems han o
consis en i ems and o pseudowo ds han o wo ds (Fig.1; Table2). No o he signi ican e ec s o in e ac ions
we e de ec ed (We an an addi ional model in which we excluded F ench ials wi h 4 and 5 inconsis encies o
ensu e ha hese i ems a e no d i ing he obse ed e ec s. This model likewise obse ed e ec s o Consis ency
(Es ima e=-2.475e-05, SE = 1.068e-05, =-2.318, p = .0213) and Lexicali y (Es ima e=-1.040e-04, SE = 1.255e-05,
=-8.282, p < .001) and no o he signi ican main e ec s o in e ac ions.).
To u he explo e he da a, we spli he da a by language and an wo sepa a e linea -mixed e ec models
o see i he OCE was de ec able in each language indi idually. Each model had he same s uc u e o he main
model excep o ha ing Language as an independen a iable. The model un on he F ench da a showed main
e ec s o Consis ency and Lexicali y wi h no in e ac ions (Table3) while he model un on he Spanish da a
showed only a main e ec o Lexicali y (Table4).
Discussion
The O hog aphic Consis ency E ec (OCE) is he p ocessing ad an age o wo ds wi h a consis en spelling,
which has p e iously been obse ed in wo ds wi h consis en compa ed o inconsis en ime spellings in
o hog aphically opaque languages (e.g., globe and p obe a e judged as hyming wo ds as e han ye and ie10).
In con as o p e ious esea ch, we manipula ed o hog aphic inconsis ency a he phonemic a he han he
sup asegmen al ime le el. This manipula ion esul ed in wo ds and pseudowo ds con aining phonemes ha
ei he mapped on o only a single o hog aphic ep esen a ion (consis en wo ds/pseudowo ds) and wo ds and
pseudowo ds con aining some phonemes wi h mul iple po en ial spellings (inconsis en wo ds/pseudowo ds).
Ou esul s indica e ha his phonemic manipula ion also e eals an OCE in audi o y p ocessing o bo h wo ds
and pseudowo ds, ye wi h di e ences be ween F ench and Spanish.
Scien i ic Repo s | (2025) 15:8374 4
| h ps://doi.o g/10.1038/s41598-025-92885-x
www.na u e.com/scien i ic epo s/
Reac ion ime analyses showed ha inconsis encies a he phonemic le el impac ed audi o y lexical
p ocessing: na i e F ench lis ene s we e slowe when iden i ying inconsis en compa ed o consis en i ems. This
no el inding in o ms p e ious esea ch ha in es iga ed he OCE a he ime o syllabic le els4,7,9,36.
Fo example, he OCE ound in Ziegle and Fe and9 was a ibu ed o he (in)consis encies o he sound- o-
spelling mappings o F ench imes (e.g., gamme /gam/ is inconsis en because -/am/ can be spelled -< am>, -<
ame > o -< amme> ) while he indi idual phonemes cons i u ing he ime (/a/ and /m/ in /gam/) a e consis en
when conside ed indi idually. Ye , some s imuli like ga e /ga / a e inconsis en only because one phoneme (/ /
in /ga /) in he ime can be spelled in mul iple ways (< > o < ph > in ga e). I is consequen ly no possible o
Es ima e SE p
(In e cep ) 9.917e-04 1.704e-05 58.205 < 0.001
Consis ency -3.492e-05 1.271e-05 -2.748 0.007
Lexicali y -1.159e-04 1.606e-05 -7.219 < 0.001
Consis ency × Lexicali y -8.467e-08 2.421e-05 -0.003 0.997
Table 3. Resul s able o he eac ion ime analysis o he F ench da a.
Es ima e SE p
(In e cep ) 1.001e-03 1.285e-05 77.913 < 0.001
Consis ency -2.590e-05 1.055e-05 -2.454 0.015
Language 1.985e-05 2.570e-05 0.773 0.442
Lexicali y -1.057e-04 1.238e-05 -8.536 < 0.001
Consis ency × Language 1.734e-05 2.111e-05 0.822 0.412
Consis ency × Lexicali y 1.296e-05 2.062e-05 0.629 0.530
Language × Lexicali y 4.589e-05 2.477e-05 1.853 0.065
Consis ency × Language × Lexicali y 5.817e-05 4.124e.-5 1.411 0.160
Table 2. Resul s able o he eac ion ime analysis.
Fig. 1. Raw eac ion imes (RT) by Consis ency and Lexicali y in F ench and Spanish. No e. Each do shows an
indi idual pa icipan and he black squa e shows he g oup mean.
Scien i ic Repo s | (2025) 15:8374 5
| h ps://doi.o g/10.1038/s41598-025-92885-x
www.na u e.com/scien i ic epo s/
gi e a uni ocal explana ion o he linguis ic le el a ec ed by he OCE. Ou s udy disen angles his ambigui y by
showing ha (in)consis encies o indi idual phonemes can a ec audi o y lexical p ocessing.
Fu he mo e, ou indings ex end p e ious esea ch ha showed a p ocessing disad an age o wo ds
con aining inconsis en ly spelled phonemes in wo d w i ing19 and isual wo d ecogni ion20 as well as p e ious
esea ch on me alinguis ic skills showing ha lis ene s mo e easily manipula e consis en han inconsis en
phonemes21 and ha hey ecognize consis en phonemes as e and mo e accu a ely han inconsis en
phonemes22. This implies, he e o e, ha audi o y lexical p ocessing is a ec ed no only by ime (in)consis ency
bu also by phonemic (in)consis ency. This is only logical, gi en ha he phonemic le el plays a unique ole in
language p ocessing in gene al o people who a e li e a e in alphabe ical languages, as demons a ed by he
inding ha phonemic awa eness is a be e p edic o o success in lea ning o ead han onse / ime awa eness37.
Since li e acy de i es om es ablishing a ela ionship be ween indi idual phonemes and hei o hog aphic
ep esen a ions, ou indings con i m he impo ance o li e acy in audi o y language p ocessing. In ac , i has
ecen ly been a gued38 ha phonemes can be ega ded as a by-p oduc o li e acy as hei ela ion wi h hei
o hog aphic ep esen a ions s ongly impac s audi o y language p ocessing in li e a e people.
P e ious esea ch in es iga ed he OCE mainly in languages wi h opaque o hog aphies (i.e., English
and F ench) o in languages wi h a ai ly anspa en o hog aphy (i.e., Po uguese). Ou s udy shows ha
o hog aphic (in)consis ency does no de ec ably in e ac wi h language, sugges ing ha he OCE a ec ed bo h
F ench and Spanish. I hus seems ha he OCE can be de ec ed in a b oade spec um o alphabe ical languages
anging om he leas o he mos anspa en . Howe e , when spli ing he da a by language, no OCE was
de ec able in he Spanish da a alone. The e o e, ou esul s seem o be mo e inconclusi e in Spanish, a highly
anspa en language, whe e only a nume ical endency appea ed o an OCE in audi o y wo d p ocessing.
As he OCE seems o be p ominen in audi o y language p ocessing when he language is opaque a he han
anspa en , mo e esea ch in o he OCE in anspa en languages is needed o a es o wha ex en he OCE
a ec s anspa en languages.
In e es ingly, ou esul s e ealed an OCE in bo h wo d and pseudowo d p ocessing, indica ing ha na i e
F ench lis ene s s a egically ely on phoneme- o-g apheme con e sions du ing pseudowo d ecogni ion.
This is ema kable as he po en ial op-down p ocesses o wo d ecogni ion canno ha e he same impac on
pseudowo d ecogni ion because hei p ocessing is independen o lexical access o pa ial lexical ac i a ion39.
Howe e , i is possible ha he unde lying mechanisms o he OCE o wo ds and pseudowo ds di e , gi en
ha eal wo ds ha e a ep esen a ion in long- e m memo y while pseudowo ds do no . Consequen ly, e en i
bo h ypes o s imuli igge ed appa en ly simila consis ency e ec s, hey may no ela e o he same unde lying
p ocesses. The OCE obse ed o eal wo ds could be ela ed o compa isons be ween phonological memo y
aces in wo king memo y o o hog aphic memo y aces in long- e m memo y, while he OCE obse ed in
pseudowo ds could be ela ed o phoneme- o-g apheme con e sion e ec s. Ou esul s on he OCE in wo ds
and pseudowo ds a e in line wi h Pa amadilok and colleagues2 bu in con as wi h o he s udies on he OCE
in F ench7,9 and Po uguese4, which did no ind an OCE in pseudowo d ecogni ion. An in e p e a ion o his
disc epancy could be ha he compe i ion be ween di e en o hog aphic ep esen a ions o an inconsis en
phoneme is g ea e han he compe i ion be ween di e en o hog aphic ep esen a ions o an inconsis en ime
a he p e-lexical s age o audi o y language p ocessing.
In o he wo ds, he di e ence be ween he indings on F ench pseudowo d ecogni ion in his s udy and
he lack o an e ec in F ench pseudowo d ecogni ion in o he s udies sugges s ha inconsis encies a he
phonemic le el a e mo e impac ul han inconsis encies a he ime le el, a leas wi h inconsis en phonemes
in ini ial posi ion in he wo d/pseudowo d. Wo d onse playing a majo ole in audi o y wo d ecogni ion, he
impac o inconsis encies in wo d ini ial posi ion migh be la ge . Fu u e s udies should explo e his ques ion by
compa ing he magni ude o he OCE wi h inconsis en phonemes being in ini ial o inal posi ion.
In conclusion, he p esen s udy showed ha na i e lis ene s o opaque languages a e a ec ed by o hog aphic
inconsis encies a he phonemic le el du ing audi o y wo d and pseudowo d p ocessing. Fu u e esea ch should
compa e he magni ude o he OCE a he ime and phonemic le els in opaque and anspa en languages o
disen angle how o hog aphic (in)consis encies a di e en linguis ic le els impac audi o y wo d ecogni ion.
Da a a ailabili y
Da a and analysis code a e a ailable on he OSF: h p s : / / o s . i o / x d 4 h / ? i e w _ o n l y = a d 1 8 d a 4 2 e 4 d 2 4 e b 9 8 9 7 4 0 8 b
e 0 5 e 8 c d 9.
Recei ed: 5 June 2024; Accep ed: 3 Ma ch 2025
Es ima e SE p
(In e cep ) 1.010e-03 1.896e-05 53.266 < 0.001
Consis ency -1.652e-05 1.579e-05 -1.047 0.297
Lexicali y -8.365e-05 1.844e-05 -4.536 < 0.001
Consis ency × Lexicali y 4.246e-05 3.074e-05 1.381 0.170
Table 4. Resul s able o he eac ion ime analysis o he Spanish da a.
Scien i ic Repo s | (2025) 15:8374 6
| h ps://doi.o g/10.1038/s41598-025-92885-x
www.na u e.com/scien i ic epo s/
Re e ences
1. Dich, N. Indi idual di e ences in he size o o hog aphic e ec s in spoken wo d ecogni ion: he ole o lis ene s’ o hog aphic
skills. Appl. Psycholinguis . 32, 169–186. h ps://doi.o g/10.1017/S0142716410000330 (2011).
2. Pa amadilok, C., Mo ais, J., Ven u a, P. & Kolinsky, R. The locus o he o hog aphic consis ency e ec in audi o y wo d
ecogni ion: u he e idence om F ench. Lang. Cogn. P oc. 22, 700–726. h ps://doi.o g/10.1080/01690960601049628 (2007).
3. Pe e, L. & Ziegle , J. C. On-line ac i a ion o o hog aphy in spoken wo d ecogni ion. B ain Res. 1188, 132–138. h p s : / / d o i . o g /
1 0 . 1 0 1 6 / j . b a i n e s . 2 0 0 7 . 1 0 . 0 8 4 (2008).
4. Ven u a, P., Mo ais, J., Pa amadilok, C. & Kolinsky, R. The locus o he o hog aphic consis ency e ec in audi o y wo d
ecogni ion. Lang. Cogn. P oc. 19, 57–95. h ps://doi.o g/10.1080/01690960344000134 (2004).
5. Ven u a, P., Mo ais, J. & Kolinsky, R. The de elopmen o he o hog aphic consis ency e ec in speech ecogni ion: om sublexical
o lexical in ol emen . Cogni ion 105, 547–576. h ps://doi.o g/10.1016/j.cogni ion.2006.12.005 (2007).
6. Ven u a, P., Kolinsky, R., Pa amadilok, C. & Mo ais, J. The de elopmen al u npoin o o hog aphic consis ency e ec s in speech
ecogni ion. J. Exp. Child. Psychol. 100, 135–145. h ps://doi.o g/10.1016/j.jecp.2008.01.003 (2008).
7. Ziegle , J. C., Fe and, L. & Mon an , M. Visual phonology: he e ec s o o hog aphic consis ency on di e en audi o y wo d
ecogni ion asks. Mem. Cogn. 32, 732–741. h ps://doi.o g/10.3758/BF03195863 (2004).
8. Ziegle , J. C., Pe o a, A. & Fe and, L. Feedback consis ency e ec s in isual and audi o y wo d ecogni ion: whe e do we s and
a e mo e han a decade? J. Exp. Psychol. Lea n. 34, 643–661. h ps://doi.o g/10.1037/0278-7393.34.3.643 (2008).
9. Ziegle , J. C. & Fe and, L. O hog aphy shapes he pe cep ion o speech: he consis ency e ec in audi o y wo d ecogni ion.
Psychon B Re . 5, 683–689. h ps://doi.o g/10.3758/BF03208845 (1998).
10. Seidenbe g, M. S. & Tanenhaus, M. K. O hog aphic e ec s on hyme moni o ing. J. Exp. Psychol. -Hum L. 5, 546–554. h p s : / / d o i
. o g / 1 0 . 1 0 3 7 / 0 2 7 8 - 7 3 9 3 . 5 . 6 . 5 4 6 (1979).
11. F os , R., Repp, B. H. & Ka z, L. Can speech pe cep ion be in luenced by simul aneous p esen a ion o p in ? J. Mem. Lang. 27,
741–755. h ps://doi.o g/10.1016/0749-596X(88)90018-6 (1988).
12. Jakimik, J., Cole, R. A. & Rudnicky, A. I. Sound and spelling in spoken wo d ecogni ion. J. Mem. Lang. 24, 165–178. h p s : / / d o i . o
g / 1 0 . 1 0 1 6 / 0 7 4 9 - 5 9 6 X ( 8 5 ) 9 0 0 2 2 - 1 (1985).
13. Cu le , A. Spoken wo d ecogni ion and p oduc ion. In (eds. Mille , J. L. & Eimas, P. D.) Speech, Language, and Communica ion 2,
97–136. h ps://doi.o g/10.1016/B978-012497770-9.50006-6 (1995).
14. Pe o a, A., Gaskell, M. G. & Fe and, L. O hog aphic consis ency and wo d- equency e ec s in audi o y wo d ecogni ion: new
e idence om lexical decision and ime de ec ion. F on Psychol. 2, 1–11. h ps://doi.o g/10.3389/ psyg.2011.00263 (2011).
15. Pa amadilok, C., Pe e, L., Du au, S. & Ziegle , J. C. On-line o hog aphic in luences on spoken Language in a seman ic ask. Jou
Cog Neu osci. 21(1), 169–179. h ps://doi.o g/10.1162/jocn.2009.21014 (2009).
16. Pe e, L., Pa amadilok, C., Mon an , M. & Ziegle , J. C. O hog aphic e ec s in spoken Language: On-line ac i a ion o
phonological es uc u ing? B ain Res. 1275, 73–80. h ps://doi.o g/10.1016/j.b ain es.2009.04.018 (2009).
17. Pe e, L. & Ziegle , J. C. On-line ac i a ion o o hog aphy in spoken wo d ecogni ion. B ain Res. h p s : / / d o i . o g / 1 0 . 1 0 1 6 / j . b a i n
e s . 2 0 0 7 . 1 0 . 0 8 4 (2008).
18. Pee eman, R., Con en , A. & Bonin, P. Is pe cep ion a wo-way s ee ? The case o eedback consis ency in isual wo d ecogni ion.
J. Mem. Lang. 39, 151–174. h ps://doi.o g/10.1006/jmla.1998.2573 (1998).
19. Pe y, C. A phoneme-g apheme eedback consis ency e ec . Psychon Bl Re . 10, 392–397. h ps://doi.o g/10.3758/BF03196497
(2003).
20. Cas les, A., Holmes, V. M., Nea h, J. & Kinoshi a, S. How does o hog aphic knowledge in luence pe o mance on phonological
awa eness asks? Q. J. Exp. Psychol. -A. 56 A, 445–467. h ps://doi.o g/10.1080/02724980244000486 (2003).
21. Cu le , A., T eiman, R. & an Ooijen, B. S a egic deploymen o o hog aphic knowledge in phoneme de ec ion. Lang. Speech. 53,
307–320. h ps://doi.o g/10.1177/0023830910371445 (2010).
22. De io , S., Ma os, F. & Ca y, L. Di e ences in eading acquisi ion de elopmen in wo shallow o hog aphies: Po uguese and
Spanish. Appl. Psycholinguis . 23, 135–148. h ps://doi.o g/10.1017/s0142716402000073 (2002).
23. Seymou , P. H. K., A o, M., E skine, J. M. & Ne wo k, collabo a ion wi h C. A. A. Founda ion li e acy acquisi ion in Eu opean
o hog aphies. B i J. Psychol. 94, 143–174. h ps://doi.o g/10.1348/000712603321661859 (2003).
24. Kau man, A. S. & Kau man, N. L. Kau man B ie In elligence Tes 2nd edn (Ame ican Guidance Se ice, 2004).
25. New, B., Pallie , C., B ysbae , M. & Fe and, L. Lexique 2: A new F ench lexical da abase. Beha . Res. Me h Ins C. 36, 516–524.
h ps://doi.o g/10.3758/BF03195598 (2004).
26. Duchon, A. e al. One-s op shopping o Spanish wo d p ope ies. Beha . Res. Me h. 45, 1246–1258. h p s : / / d o i . o g / 1 0 . 3 7 5 8 / s 1 3 4 2
8 - 0 1 3 - 0 3 2 6 - 1 (2013).
27. Ma ian, V., Ba olo i, J., Chabal, S. & Shook, A. C. L. E. A. R. P. O. N. D. C oss-linguis ic easy-access esou ce o phonological and
o hog aphic neighbo hood densi ies. PLoS ONE. 7, e43230. h ps://doi.o g/10.1371/jou nal.pone.0043230 (2012).
28. Ma hô , S., Sch eij, D. & Theeuwes, J. OpenSesame: An open-sou ce, g aphical expe imen builde o he social sciences. Beha .
Res. Me h. 44, 314–324. h ps://doi.o g/10.3758/s13428-011-0168-7 (2012).
29. Posi eam. RS udio: In eg a ed De elopmen En i onmen o R ( e sion 2023.12.1.402) (Posi So wa e, 2024).
30. R Co e Team. R: A Language and En i onmen o S a is ical Compu ing ( e sion 4.2.0) (R Founda ion o S a is ical Compu ing,
2022).
31. Ba es, D., Mächle , M., Bolke , B. M. & Walke , S. C. Fi ing linea mixed-e ec s models using lme4. J. S a . So w. 67, 1–48. h p s :
/ / d o i . o g / 1 0 . 1 8 6 3 7 / j s s . 0 6 7 . i 0 1 (2015).
32. Kuzne so a, A., B ockho , P. B. & Ch is ensen, R. H. B. Lme Tes package: es s in linea mixed e ec s models. J. S a . So w. 82,
1–26. h ps://doi.o g/10.18637/jss. 082.i13 (2017).
33. Box, G. E. P. & Cox, D. R. An analysis o ans o ma ions. J. R. S a . Soc. 26, 211–252. h p s : / / d o i . o g / 1 0 . 1 1 1 1 / J . 2 5 1 7 - 6 1 6 1 . 1 9 6 4 . T
B 0 0 5 5 3 . X (1964).
34. Venables, W. N. & Ripley, B. D. Mode n Applied S a is ics Wi h S 4 h edn (Sp inge , 2002). h p s : / / d o i . o g / 1 0 . 1 0 0 7 / 9 7 8 - 0 - 3 8 7 - 2 1 7 0
6 - 2
35. T emblay, A. & Ransijn, J. Package ‘LMERCon enienceFunc ions’. Model selec ion and pos -hoc analysis o (G)LMER models (R
package; Ve sion 3.0 (2020).
36. Pa amadilok, C., Mo ais, J., De Vylde , O., Ven u a, P. & Kolinsky, R. The o hog aphic consis ency e ec in he ecogni ion o
F ench spoken wo ds: an ea ly de elopmen al shi om sublexical o lexical o hog aphic ac i a ion. Appl. Psycholinguis . 30,
441–462. h ps://doi.o g/10.1017/S0142716409090225 (2009).
37. Hulme, C. e al. Phoneme awa eness is a be e p edic o o ea ly eading skill han onse - ime awa eness. J. Exp. Child. Psychol. 82,
2–28. h ps://doi.o g/10.1006/jecp.2002.2670 (2002).
38. Mo ais, J. The phoneme: A concep ual he i age om alphabe ic li e acy. Cogni ion 213, 104740. h p s : / / d o i . o g / 1 0 . 1 0 1 6 / j . c o g n i i o
n . 2 0 2 1 . 1 0 4 7 4 0 (2021).
39. Newman, S. D. & Twieg, D. Di e ences in audi o y p ocessing o wo ds and pseudowo ds: an MRI s udy. Hum. B ain Mapp. 14,
39–47. h ps://doi.o g/10.1002/hbm.1040 (2001).
Acknowledgemen s
This wo k was suppo ed by ins i u ional g an s om he Basque Go e nmen [BERC 2022–2025 p og am]
Scien i ic Repo s | (2025) 15:8374 7
| h ps://doi.o g/10.1038/s41598-025-92885-x
www.na u e.com/scien i ic epo s/
and he Spanish S a e Resea ch Agency [BCBL Se e o Ochoa excellence acc edi a ion CEX2020-001010/
AEI/10.13039/501100011033] awa ded o he BCBL. This p ojec has also ecei ed unding om a “La Caixa”
Founda ion ellowship [ID 100010434; ellowship code LCF/BQ/IN18/11660068; co- unded by he Eu opean
Union’s Ho izon 2020 esea ch and inno a ion p og am unde he Ma ie Skłodowska Cu ie g an ag eemen
No. 713673 awa ded o AF], he Eu opean Resea ch Council (ERC) unde he Eu opean Union’s Ho izon 2020
esea ch and inno a ion p og am [g an ag eemen No 819093 o CDM], and om he Spanish S a e Resea ch
Agency [PID2020-113926GB-I00 o CDM; PID2021-123578NA-I00/AEI/10.13039/501100011033/FEDER, UE,
& FJC2020-044978-I o AS]. We hank Ca oline Handley o p oo - eading his manusc ip .
Au ho con ibu ions
Albe o Fu goni: Concep ualiza ion, Me hodology, Fo mal Analysis, In es iga ion, Resou ces, Da a Cu a ion,
W i ing – O iginal D a , Visualiza ion, P ojec Adminis a ion, Funding Acquisi ion; Cla a D. Ma in: Concep-
ualiza ion, Me hodology, W i ing – Re iew & Edi ing, Supe ision, Funding Acquisi ion; An je S oeh : Con-
cep ualiza ion, Me hodology, Fo mal Analysis, Resou ces, W i ing – Re iew & Edi ing, Visualiza ion, Supe i-
sion, Funding Acquisi ion.
Decla a ions
Compe ing in e es s
The au ho s decla e no compe ing in e es s.
Addi ional in o ma ion
Co espondence and eques s o ma e ials should be add essed o A.F.
Rep in s and pe missions in o ma ion is a ailable a www.na u e.com/ ep in s.
Publishe ’s no e Sp inge Na u e emains neu al wi h ega d o ju isdic ional claims in published maps and
ins i u ional a ilia ions.
Open Access This a icle is licensed unde a C ea i e Commons A ibu ion 4.0 In e na ional License, which
pe mi s use, sha ing, adap a ion, dis ibu ion and ep oduc ion in any medium o o ma , as long as you gi e
app op ia e c edi o he o iginal au ho (s) and he sou ce, p o ide a link o he C ea i e Commons licence, and
indica e i changes we e made. The images o o he hi d pa y ma e ial in his a icle a e included in he a icle’s
C ea i e Commons licence, unless indica ed o he wise in a c edi line o he ma e ial. I ma e ial is no included
in he a icle’s C ea i e Commons licence and you in ended use is no pe mi ed by s a u o y egula ion o
exceeds he pe mi ed use, you will need o ob ain pe mission di ec ly om he copy igh holde . To iew a copy
o his licence, isi h p://c ea i ecommons.o g/licenses/by/4.0/.
© The Au ho (s) 2025
Scien i ic Repo s | (2025) 15:8374 8
| h ps://doi.o g/10.1038/s41598-025-92885-x
www.na u e.com/scien i ic epo s/
