The contrastive value of lexical stress in visual word recognition: evidence from Spanish

Albe o Domínguez Ma ínez and Fe nando Cue os Vega 276 One pa icula wo d di e s om o he s by one o mo e le e s in he alphabe ical languages. Some wo ds di e in a single le e such as, in Spanish, casa (house) and cama (bed), o in he o de o le e s and/o syllables, such as bolo (ski le) and lobo (wol ). Some imes a single le e becomes a c i ical p ope y o con as be ween di e en mo phemes, as i happens wi h cas-o (case) and cas-a (house), o di e en gende s o he same s em mo pheme, such as in niñ-o (boy) and niñ-a (gi l). Howe e , a minimal con as ing ea u e can also go e n he di e ence be ween wo ds: he s ess. This is he case o wo ds such as Ánimo (cou age), aNImo (I encou age) and aniMÓ (he encou aged). Pai s o wo ds di e ing only in one speci i c con as i e ea u e, such as casa and cama a e mo e di i cul o ecognize han hose di e ing in many ea u es, such as pe o (dog) and cadena (chain). In he wo las decades many s udies on wo d ecogni ion ha e ocused on he psychological mechanisms allowing he selec ion o one lexical ep esen a ion among a a iable numbe o pa ially simila wo ds (And ews, 1997; G ainge , 1992; Ma ce & Pe ea, 2017; Pe ea & Rosa, 2000). Acco ding o seminal s udies such as Segui and G ainge (1990) we know ha exci a o y and inhibi o y p ocesses a e impo an mechanisms o lexical selec ion and he le e s become de e minan ea u es ha make i possible o dis inguish wo ds. The goal o his a icle is o decide whe he ISSN 0214 - 9915 CODEN PSOTEG Copy igh © 2018 Psico hema www.psico hema.com The con as i e alue o lexical s ess in isual wo d ecogni ion: E idence om Spanish Albe o Domínguez Ma ínez and Fe nando Cue os Vega Uni e sidad de La Laguna and Uni e sidad de O iedo Abs ac Resumen Backg ound: Many pai s o wo ds in Spanish, in pa icula many e bal o ms, di e only in he syllable s essed, such as aNImo (I encou age) and aniMÓ (he encou aged). Consequen ly, wo d s ess may acqui e a lexical con as i e alue ha has been con i med by Dupoux, Pallie , Sebas ian, and Mehle (1997) o Spanish speake s hough no o F ench speake s in audi o y pe cep ion. Me hod: This s udy con as s he p iming e ec p oduced by pai s o w i en wo ds ha di e only in hei s ess pa e n wi h he p iming e ec in epe i ion p iming, s ess only p iming (wi h no o hog aphic ela ion), and mo phological p iming, in isual wo d ecogni ion. Resul s: The esul s, using sho and masked p ime p esen a ion, showed acili a ion o di e en s ess (o hog aphically iden ical) pai s ( asGÓ/RASgo) compa ed o o ally un ela ed pai s (do MÍ/RASgo) bu no acili a ion compa ed o o hog aphically un ela ed (bu s ess ela ed) pai s (PERsa/RASgo). Howe e , iden i y pai s (RASgo- RASgo) p oduced acili a ion compa ed o bo h o hog aphically un ela ed condi ions. A long SOA, o hog aphically ela ed (s ess un ela ed) pai s p oduced signi i can acili a ion, as occu ed wi h mo phologically ela ed pai s (RASga/RASgo), on he o hog aphically un ela ed wo ds (PERsa/ RASgo). Conclusion: These esul s con i m he ea ly and p elexical impo ance o wo d s ess o lexical selec ion in Spanish, as is he case wi h o hog aphic and phonological ea u es. Keywo ds: Lexical s ess, accen , epe i ion p iming, o m p iming, mo phological p iming, isual wo d ecogni ion, p elexical p ocessing. El alo con as i o del acen o léxico en el econocimien o isual de palab as: e idencia del español. An eceden es: muchos pa es de palab as en español, en pa icula muchas o mas e bales, di i e en solo en la sílaba acen uada, al como aNImo y aniMÓ. Así el acen o puede adqui i un alo con as i o que ue con i mado po Dupoux, Pallie , Sebas ian y Mehle (1997) en español, pe o no en ancés, en pe cepción audi i a. Mé odo: es e es udio con as a el e ec o de p imado en pa es de palab as que di i e en en su pa ón de acen uación con el e ec o de epe ición, el p imado solo de acen o (sin elación o og á i ca) y el p imado mo ológico en econocimien o isual de palab as. Resul ados: usando p iming enmasca ado se ob u o acili ación pa a los pa es con di e en e acen o ( asGÓ/RASgo) compa ando con los pa es sin elación (do MÍ/ RASgo), pe o no se p odujo compa ando con los no elacionados de igual acen o (PERsa/RASgo). Sin emba go, los pa es idén icos (RASgo- RASgo) p oduje on acili ación compa ando con ambas condiciones o og á i camen e elacionadas. Con un SOA la go los pa es con di e en e acen o (o og á i camen e iguales ) p oduje on una acili ación signi i ca i a, como ocu ió con los pa es elacionados mo ológicos (RASga/RASgo), sob e los pa es o og á i camen e di e en es (PERsa/RASgo). Conclusión: es os esul ados con i man la impo ancia del p ocesamien o emp ano y p eléxico del acen o pa a la selección léxica en español, como ocu e con las ca ac e ís icas o og á i cas y onológicas de las palab as. Palab as cla e: acen o léxico, p imado de epe ición, p imado o mal, p imado mo ológico, econocimien o isual de palab as, p ocesamien o p eléxico. Psico hema 2018, Vol. 30, No. 3, 276-282 doi: 10.7334/psico hema2018.30 Recei ed: Janua y 30, 2018 • Accep ed: May 16, 2018 Co esponding au ho : Albe o Domínguez Ma ínez Facul ad de Psicología Uni e sidad de La Laguna 38205 Tene i e (Spain) e-mail: [email p o ec ed] The con as i e alue o lexical s ess in isual wo d ecogni ion: E idence om Spanish 277 o no , in eading in Spanish, s ess plays a simila ole in lexical access as le e s do. In ac , many wo ds in Spanish can only be di e en ia ed by he syllable ha is s essed (e.g. BEbe [he/she d inks] and beBÉ [baby]). Mos s udies o lexical s ess ha e been ca ied ou on he pe cep ion o speech (al hough hings and changing, e.g., Sulpizio, Spinelli, & Bu ani, 2015) and ha e p og essed in wo di ec ions (Cu le , 1986). Lexical s ess heo ies a e cen e ed on he con ibu ion o s ess o lexical access as a p elexical cue, while me ical s ess heo ies ocus on he p oblem o lis ene use o s ong and weak syllable al e na ion o iden i y wo d bounda ies in speech (Cu le , 1986; Cu le & Cli on, 1985). English s ess is lexical, associa ed wi h each wo d and can all on any syllable o he wo d. In F ench, in con as , he s ess is no a iable and always alls on he i nal syllable (A ciuli & Slowiaczek, 2007). The lexical con as i e unc ion o English s ess is, howe e , es ic ed o a small g oup o pai s o wo ds. The majo i y o he mos commonly used wo ds (be ween 80% and 90%) ha e a pa e n o s ong-weak (Sw) syllables. The emaining wo ds, weak-s ong (Ws), a e gene ally e bs. When s ess is he only phonological cue ha dis inguishes a pai o wo ds, one o hem, he Sw wo d, is usually a noun and he Ws wo d a e b (Cu le & Ca e , 1987). Membe s o he same g amma ical ca ego y a e no usually con as ed by he s ess. Only his minimal ea u e con as s some excep ions, such as FORbea and o BEAR. The con as i e ole o Spanish s ess Spanish s ess may all on any o he las h ee syllables o a wo d. The mos equen pa e n is penul s ess on owel- i nal wo ds (CAsa “house”). Consonan - i nal wo ds s essed on he i nal syllable a e conside ed egula s (ca TEL “pos e ”) while an epenul s ess is always i egula (PÍca o “ illainous”; Pensado, 1999). Acco ding o Mo ales-F on (2014), abou 64% o he Spanish wo ds ha e he s ess on he penul syllable, 28% on he las syllable and only 8% on he an epenul syllable. Eddin ong (2000) ca ied ou a s ess classi i ca ion o he 4829 mos equen polysyllabic wo ds selec ed om he Alameda and Cue os equency dic iona y o Spanish (1995) suppo ing his linguis ic dis inc ion: om he o al o wo ds ending in a owel, 178 a e s essed on he i nal syllable, 2494 on he penul syllable, and 178 on he an epenul syllable; om he o al o consonan - i nal wo ds (excep plu al wo ds ending in he su i x –s) 778 a e s essed on he i nal syllable, 176 on he penul syllable and 2 on he an epenul syllable. This lexical dis ibu ion o s ess shows a mo e complex pano ama han he simply p edominan egula i y o he Sw wo ds in English. Howe e , some English au ho s ha e de eloped models ha conside s ess as an impo an cue o sea ch in he lexicon, e en in isual wo d ecogni ion. The model o Black and Byng (1986) p edic s ha sea ch o a wo d in he lexicon begins wi h hose ha ing he mos egula s ess pa e n. Gu ié ez, Palma, and San iago (1998) ied o e i y his p edic ion in Spanish using egula wo ds (Sw) and i egula wo ds (Ws) bu only he numbe o e o s and no he eac ion imes seems o a o he Black and Byng’s model. Pe haps he mos impo an di e ence be ween Spanish and English is he capaci y o Spanish s ess o dis inguish be ween wo ds. Each wo d in Spanish has a de i ned and in a iable s ess pa e n. Thus s ess is, some imes, he only ea u e ha esol es he ambigui y de i ed om he exis ence o di e en wo ds made up o he same phoneme s ings (Ánimo, aNImo, aniMÓ [ he cou age, I encou age, he encou aged]). In w i en language his ambigui y is esol ed by he p esence o absence o a punc ua ion ma k: he accen (´), indica ing he syllable ecei ing he s ess. The accen ma ks hose s ess pa e ns which, in Spanish, a e mos i egula (Ánimo, animÓ) and is no used o egula pa e ns, such as aNImo, s essed on he penul syllable o ca TEL, s essed on he i nal syllable. In speech pe cep ion, Dupoux, Pallie , Sebas ian, and Mehle (1997) demons a ed he capaci y o Spanish lis ene s o dis inguish, wi h an ABX pa adigm, be ween pseudowo ds s essed on di e en syllables (A/ BOpelo, B/ boPElo, X) boPElo). F ench lis ene s, howe e , we e unable o dis inguish be ween he s imuli. Also, So o, Cu le , and Sebas ián (2001) ound some esul s in a o o s ess as a p e-lexical cue in Spanish. Gu ié ez- Palma and Palma-Reyes (2008) p imed a wo d wi h co ec ly and inco ec ly s essed wo ds (e.g., écla-TECLA s. eclá-TECLA) a sho and long SOAs (33, 66, 100, and 143 ms) ounding only acili a ion a he longes SOA and inhibi ion a he incong uen p iming also a 100 and 143 ms SOA. They sugges ha s ess a ec s lexical access a a la e s age o lexical access p ocessing. Agains his la e in l uence o he accen ma ks in Spanish, Pe ea, Abu Mallouh, Mohamed, Khali a and Ca ei as (2018) i nd a e y ea ly in l uence o he diac i ic ma ks ha di e en ia e A abic le e s. In a masked p iming expe imen , wi h a p ime du a ion o 50 ms., hey ound an o hog aphic p iming e ec which was independen o whe he he le e changed in i s o m o in he numbe and posi ion o i s diac i ical ma ks wi h espec o he a ge . Also, bo h, o m o hog aphic p iming and diac i ic p iming p oduced simila and longe imes han epe i ion p iming. The pa icipan s in his expe imen we e child en, bu p e iously hey had been ob ained wi h adul s (Pe ea, Abu Mallouh, Mohamed, Khali a and Ca ei as, 2016). The aim o he p esen s udy is o con as he p iming e ec s p oduced by pai s o w i en wo ds ha di e only in hei s ess pa e n, wi h he p iming e ec in se e al o he expe imen al manipula ions such as epe i ion p iming, s ess only p iming, (wi hou o hog aphic ela ion) and mo phological p iming, in isual wo d ecogni ion. De e ing o he p e ious desc ibed s udy o Gu ié ez-Palma and Palma-Reyes (2008) all p imes and a ge s we e legal and exis ing wo ds in Spanish. The gene al aim, gi en he con as i e ole o s ess in Spanish, is o know whe he o no s ess is a ele an ea u e o ea ly and p e-lexical p ocessing o he wo d, as seems o occu wi h o he o hog aphic and phonological ea u es and also, wi h diac i ical poin s in A abic. Expe imen 1 This expe imen a emp s o compa e di e en s ess epe i ion p iming (p ime and a ge wi h o al o hog aphic o e lap bu di e en s ess pa e n [ asGÓ/RASgo, he o e- I ea o ea u e]) wi h epe i ion p iming (RASgo/RASgo I ea /I ea ). This con as aims o disco e whe he s ess disag eemen (wi h o al o hog aphic o e lap) is able o modi y he acili a ion p oduced by he o al o hog aphic o e lap. Repe i ion p iming p oduces consis en acili a ion acco ding o many expe imen al s udies (see he seminal a icles o Fo bach, S anne s, & Hochhaus, 1974; Fo s e & Da is, 1984; Pe ea & Rosa, 2000; and Sca bo ough, Co ese, & Sca bo ough, 1977, o a compa ison o epe i ion and o m p iming). Di e en s ess epe i ion p iming and epe i ion p iming will be compa ed wi h o hog aphically un ela ed bu s ess ela ed p iming (PERsa/RASgo, Pe sian/I ea o ea u e). Albe o Domínguez Ma ínez and Fe nando Cue os Vega 278 Sho exposi ion o he p ime wo d (32 ms.), p eceded by a mask will p e en conscious iden i i ca ion o he wo d and, he e o e, p edic i e biases. This sho SOA may also p o ide a pic u e o ea ly s ages in lexical p ocessing. Ou hypo hesis is ha he amoun o acili a ion will be as g ea as he numbe o ea u es sha ed by p ime and a ge . Repe i ion pai s ha e pe ec o hog aphic and s ess ma ching. Howe e , di e en s ess pai s ma ch only o hog aphically and he e o e, i s ess is p ocessed ea ly, less acili a ion is expec ed. Me hod Pa icipan s Thi y psychology unde g adua es a La Laguna Uni e si y, Spain, pa icipa ed in he expe imen o cou se c edi . All we e Spanish na i e speake s wi h no mal o co ec ed o no mal ision. Ins umen s The s imuli we e p esen ed in he cen e o he sc een o a PC wi h a 70 Hz e esh a e. The le e s, in cou ie on , appea ed as whi e cha ac e s on a da k backg ound. The p imes we e p esen ed in uppe -case le e s while he a ge s appea ed in lowe case. Each cha ac e co e ed app oxima ely 0.38” o isual angle om a dis ance o 60 cm. These display cha ac e is ics we e main ained in all expe imen s. Six y a ge s imuli we e each pai ed wi h an o hog aphically iden ical bu di e en s ess wo d ( asGÓ/ RASgo), he same wo d (RASgo/RASgo) and an o hog aphically un ela ed wo d (PERsa/RASgo). P ime and a ge had he same syllabic s uc u e and leng h in numbe o le e s. Twen y p ime- a ge pai s we e p esen ed in each o he h ee condi ions o each pa icipan in such a way ha hey saw each p ime and a ge once du ing he expe imen . The p ime wo ds had a lowe lexical p in ed equency han he a ge . This equency ela ion was cons ained pai by pai . A lis o i lle s, wo d-wo d un ela ed pai s and wo d- nonwo d ela ed and un ela ed pai s, we e in oduced o educe he pe cen age o o hog aphically ela ed wo d-wo d pai s o 44 %. The pe cen age o ela ed wo d-nonwo d pai s was 17 %. Nine een expe imen al a ge s we e wo ds o ou le e s and o y-one we e i e le e s long. Some i lle s we e ou , i e, six o se en le e s long o app oxima e he cha ac e is ics o he s imuli o no mal Spanish dis ibu ion. All a ge s we e g amma ically ambiguous as hey could be ca ego ized ei he as e bs o as nouns. The p imes o he o hog aphic ca ego y we e unambiguous e bal o ms, whe eas o epe i ion pai s hey we e, ob iously, ambiguous. The ca ego y o un ela ed pai s was made up o a andom selec ion o g amma ically ambiguous and non-ambiguous lexical i ems. P ocedu e The masking p ocedu e was simila o ha used by Fo s e and Da is (1984) and by G ainge , Colé and Segui (1991). The sequence o e en s ha occu ed in each ial was: i s ly, a mask o hash ma ks (#), ha exac ly co e ed he numbe o spaces o he p ime s imulus was p esen ed o 500 ms. Second, he p ime was exposed in he same place o 32 ms. and was immedia ely ollowed by he a ge s imulus, which emained on he sc een un il pa icipan esponse. Pa icipan s we e no in o med o he p ime p esence. They we e old ha one s imulus would appea in he cen e o he sc een and hei ask consis ed o p essing a key as apidly and as accu a ely as possible, “YES” i a wo d and “NO” i a nonwo d appea ed. The compu e eco ded esponse keys and la encies. Da a analysis A main ac o , Type o P iming, wi h h ee le els was in oduced: o hog aphic p iming, epe i ion and un ela ed p iming. Resul s Mean eac ion imes (RTs) o co ec esponses in each condi ion excluding e o s (2.78 %) a e p esen ed in Table 1. ANOVAs we e ca ied ou by pa icipan s (F1) and by i ems (F2). La encies exceeding 1200 ms. o no eaching 200 ms. we e excluded om he analysis (2.94%) o he da a). These wo cu -o poin s we e main ained in all he expe imen s. The ac o Type o P iming p oduced signi i can e ec s on he RTs. (F1(2,58)=12.82, p<0.001, Mse=1169.11; F2(2,118)=12.96, p<0.001) as a esul o he sho e imes o epe i ion p iming. Sepa a e ANOVAs we e ca ied ou o es ablish pai ed compa isons be ween condi ions. The 39 ms di e ence be ween epe i ion p iming and di e en s ess p iming was s a is ically signi i can (F1(1,29)=19.13, p<0.001, Mse=1141.30; F2(1,59)=17.12, p<0.001, Mse=2426,80). Bo h ypes o pai s o e lap in all le e s. The only di e ence is ha he o hog aphic pai s we e s essed on a di e en syllable. Also, epe i ion pai s p oduced a signi i can di e ence o 40 ms. wi h he un ela ed pai s (F1(1,29)=21.03, p<0.001, Mse=1099.62; F2(1,59)=23.19, p<0.001, Mse=2629,90) which do no sha e any le e in he same posi ion bu we e s essed on he same syllable. Howe e , di e en s ess p iming did no p oduce any signi i can e ec on he un ela ed pai s (F1(1,29)=0.01, p=0.90, Mse=1266.39; F2(1,59)=0.62, p<0.435, Mse=3000,31). Expe imen 2 The null acili a ion o he di e en s ess pai s in Expe imen 1 was in e p e ed as he incapaci y o wo wo ds, al hough sha ing Table 1 Mean eac ion imes (and pe cen age o e o s) in expe imen 1 Type o p iming Di e ences Di e en S ess asGÓ/RASgo Repe i ion RASgo/RASgo Un ela ed PERsa/RASgo D1 D2 D3 695 (3.33) 656 (1.33) 696 (3.66) 14039 No e: D1 is he di e ence be ween Di e en S ess and Un ela ed pai s, D2 is he di e ence be ween Repe i ion and Un ela ed pai s and D3 is he di e ence be ween Di e en S ess and Repe i ion pai s The con as i e alue o lexical s ess in isual wo d ecogni ion: E idence om Spanish 279 all le e s, o p oduce acili a ion because o non-ma ching in hei s ess pa e n ( asGÓ/RASgo). The 32 ms o p ime exposi ion was su i cien o p oduce o hog aphic p ocessing as was sugges ed by he e ec i eness o he epe i ion p iming. I o hog aphic and s ess compu a ion occu independen ly du ing he ea ly milliseconds o p ocessing, i is possible o in e p e he null e ec o he o hog aphic pai s as being due o he ac ha he non- ela ed i ems did no cons i u e an adequa e base line. In ac , he o hog aphically un ela ed i ems used in Expe imen 1, such as PERsa/RASgo, a e s ess ela ed wo ds because he s ess alls on he penul syllable in bo h wo ds. I is no unlikely, hen, ha his o hog aphically un ela ed condi ion was p oducing an s ess acili a ion. Expe imen 2 a emp s o cla i y his ques ion by in oducing a mo e adequa e un ela ed condi ion. O hog aphically and s ess un ela ed pai s, such as do MÍ/RASgo, we e in oduced o compa e wi h he di e en s ess epe i ion wo ds ( asGÓ/ RASgo) and o hog aphically un ela ed wo ds (PERsa/RASgo) used in he p e ious expe imen s. I was expec ed ha , main aining he 32 ms. SOA o expe imen 1, asGÓ/RASgo would p oduce some acili a ion wi h espec o he new base line because o he o hog aphic o e lapping. Also, he o hog aphically un ela ed pai s o expe imen 1 (PERsa/RASgo) would p oduce acili a ion on he eally un ela ed pai s because o he s ess ela ionship. Me hod Pa icipan s Thi y psychology unde g adua es a La Laguna Uni e si y, in Spain, pa icipa ed in he expe imen o cou se c edi . All we e Spanish na i e speake s wi h no mal o co ec ed o no mal ision. Ins umen s The p imes o he mo phological ca ego y o expe imen 1 we e emo ed and eplaced by ano he e bal o m ha changes he posi ion o he s ess o he i nal syllable. I he a ge was he i s pe son o he p esen RASgo, he p ime was he hi d pe son o he pas do MÍ. All he o he s imuli emained as in he i s and second expe imen . The pe cen age o o hog aphically ela ed wo d-wo d pai s was 22 % and o wo d-nonwo d pai s was 17%. P ocedu e The same as in expe imen 1 (SOA o 32 ms.) Da a analysis A main ac o , Type o P iming, wi h h ee le els was in oduced: di e en s ess epe i ion p iming ( asGÓ/RASgo), o hog aphically un ela ed p iming (PERsa/RASgo) and o ally un ela ed p iming (do MÍ/RASgo). Resul s Mean eac ion imes (RTs) o co ec esponses in each condi ion excluding e o s (1.9 %) a e p esen ed in Table 3. ANOVAs we e ca ied ou by pa icipan s (F1) and by i ems (F2). La encies exceeding 1200 ms. o no eaching 200 ms. we e excluded om he analysis (2.5 % o he da a). The main ac o , Type o P iming, p oduced signi i can e ec s by pa icipan s (F1(2,58)=4.79, p<0.05, Mse=803.78; F2(2,118)=2.74, p=0.06) as a esul o he acili a ion o s ess di e en ela ed a ge s and o hog aphically un ela ed a ge s on he o ally un ela ed a ge s. Sepa a e ANOVAs we e ca ied ou o es ablish pai ed compa isons be ween condi ions. The 20 ms di e ence be ween s ess di e en epe i ion p iming and he o ally un ela ed condi ion was s a is ically signi i can by pa icipan s and i ems (F1(1,29)=8.01, p<0.01, Mse=717.97; F2(1,59)=4.62, p<0.05, Mse=1747,83). Also he o hog aphically un ela ed pai s p oduced a simila signi i can di e ence o 20 ms. wi h he o ally un ela ed pai s (F1(1,29)=7.93, p<0.01, Mse=729.92; F2(1,59)=4.33, p<0.05, Mse=2867,35). Howe e , no di e ence be ween s ess di e en ela ed and o hog aphically un ela ed pai s was ob ained, as occu ed in Expe imen 1. The esul s o Expe imen 1 we e eplica ed in his expe imen because no di e ences we e ound be ween he asGÓ/RASgo pai s and PERsa/RASgo pai s. Bo h condi ions p oduced in Expe imen 2 signi i can di e ences wi h he o ally un ela ed pai s (do MÍ/RASgo). Howe e , he acili a ion o hese wo ypes o pai s was p oduced by di e en ac o s. S ess di e en ela ed pai s p oduced acili a ion because o he o hog aphic o e lap, whe eas o hog aphically un ela ed pai s p oduced acili a ion because o s ess o e lap. Expe imen 3 The e emains one mo e con as o be ca ied ou . The di e en s ess epe i ion condi ion in expe imen s 1 and 2 is also a condi ion o mo phologically ela ed wo ds. P imes and a ge s such as asGÓ and RASgo a e in l ec ions o he same e bal s em asg-: asGÓ is he hi d pe son o he pas o m o he e b, whe eas RASgo is he i s pe son o he p esen . Some s udies ha e shown acili a ion o p ime and a ge ha di e only in hei gende o numbe o ms (Al a ez, U u ia, Dominguez & Sanchez-Casas, 2011; Dominguez, Segui & Cue os, 2002; Dominguez, De Vega & Ba be , 2004; D ews & Zwi se lood, 1995; see also Amen a & C epaldi, 2012, o a e ision o mo phological p ocessing). The Table 2 Mean eac ion imes (and pe cen age o e o s) in Expe imen 2 Type o p iming Di e ences Di e en S ess asGÓ/RASgo O h.Un ela ed PERsa/RASgo To ally Un ela ed do MÍ/RASgo D1 D2 D3 650 (2.33) 650 (3.00) 670 (2.00) 20 20 0 No e: D1 is he di e ence be ween Di e en S ess and To ally Un ela ed pai s, D2 is he di e ence be ween O hog aphically Un ela ed and To ally Un ela ed pai s and D3 is he di e ence be ween Di e en S ess and O hog aphically Un ela ed pai s Albe o Domínguez Ma ínez and Fe nando Cue os Vega 280 hi d expe imen , hen, ocuses on he compa ison o di e en s ess epe i ion p iming wi h mo phological p iming while using a longe SOA o 224 ms. The pu pose o his expe imen is o con as s ess di e en pai s and he un ela ed pai s o expe imen 1 (PERsa-RASgo) a a longe SOA. Gi en he mo phological and seman ic ela ion be ween wo ds o his ca ego y ( asGÓ- RASgo), we expec eliable acili a ion o he o hog aphically un ela ed pai s (s ess ela ed) a 224 ms. SOA. This acili a ion, i p oduced, will be compa ed wi h mo phological p iming using wo ds sha ing he s em o he e b and changing hei su i x, in his case, he las le e : RASg-a/RASg-o. Me hod Pa icipan s Thi y psychology unde g adua es a La Laguna Uni e si y, in Spain, pa icipa ed in he expe imen o cou se c edi . All we e Spanish na i e speake s wi h no mal o co ec ed o no mal ision. Ins umen s The p imes o he epe i ion ca ego y o he p e ious expe imen we e emo ed and eplaced by ano he e bal o m ha changes he su i x o pe son o ob ain he mo phological p iming ca ego y. I he a ge was he i s pe son o he p esen RASgo, he p ime was he hi d pe son o he p esen RASga. All he o he s imuli emained as in expe imen 1. The pe cen age o o hog aphically ela ed and un ela ed pai s o s imuli was simila o expe imen 1. P ocedu e The same as he p e ious expe imen , excep o he ime o exposi ion o he p ime, which, in his expe imen was o 224 ms. Da a analysis A main ac o , Type o P iming wi h h ee le els was in oduced: di e en s ess epe i ion p iming ( asGÓ/RASgo), mo phological p iming (RASga/RASgo) and un ela ed p iming (PERsa/RASgo). Resul s Mean eac ion imes (RTs) o co ec esponses in each condi ion excluding e o s (3.56 %) a e p esen ed in Table 2. ANOVAs we e ca ied ou by pa icipan s (F1) and by i ems (F2). La encies exceeding 1200 ms. o no eaching 200 ms. we e excluded om he analysis (7,44 % o he da a). The main ac o , Type o p iming, p oduced signi i can e ec s on RTs (F1(2,58)=9.96, p<0.001, Mse=2255.64; F2(2,118)=16.80, p<0.001) as a esul o he acili a ion o di e en s ess o hog aphically ela ed a ge s and mo phologically ela ed a ge s on he un ela ed a ge s. Sepa a e ANOVAs we e ca ied ou o es ablish pai ed compa isons be ween condi ions. The 35 ms di e ence be ween di e en s ess p iming and he un ela ed condi ion we e s a is ically signi i can (F1(1,29)=8.04, p<0.01, Mse=2190.70; F2(1,59)=16.75, p<0.001, Mse=2804,65). Also he mo phological pai s p oduced a signi i can di e ence o 55 ms. wi h he un ela ed pai s (F1(1,29)=26.52, p<0.001, Mse=1654.44; F2(1,59)=28.36, p<0.001, Mse=3335,35). Howe e , he di e ence be ween di e en s ess p iming and mo phological p iming (20 ms.) did no each signi i can acili a ion (F1(1,29)=2.02, p=0.166, Mse=2921.76; F2(1,59)=2.97, p=0.09, Mse=2777,72). Expe imen 1 used a e y sho p ime exposi ion o 32 ms., which was seen o be insu i cien o in l uence mo phologically he a ge in he s ess di e en pai s. The 224 ms p ime exposi ion in his expe imen did allow mo phological p ocessing. In his espec , he 55 ms. acili a ion e ec o mo phological ela i es (RASga/RASgo) suppo s his assump ion o he need o longe p ime exposi ion o allow mo phological p ocessing. Discussion A se ies o h ee expe imen s was ca ied ou o s udy he con as i e alue o he s ess and accen ma ks in Spanish. The esul s, wi h he masked p iming p ocedu e, allow suppo ing an ea ly p ocessing o lexical s ess. Expe imen 1 showed ha epea ing a wo d as p ime and a ge is no su i cien o ob ain acili a ion, i he s ess pa e n is no iden ical in bo h wo ds. I he s ess alls on di e en syllables ( asGÓ-RASgo), no acili a ion is ob ained wi h espec o he un ela ed condi ion. In con as , he same s ess pa e n ( epe i ion p iming) p oduces acili a ion (RASgo-RAS-go). The absence o acili a ion wi h a e y sho SOA (32 ms.) o s ess di e en pai s could be in e p e ed as he balance be ween acili a ion due o o hog aphic o e lap and inhibi ion due o he lack o s ess ag eemen leading o compe ing lexical ep esen a ions. Expe imen 1 allows us o in e he impo ance o s ess in lexical access, based on he null acili a ion o di e en s ess o hog aphically iden ical pai s in ela ion o epe i ion p iming. Howe e , a e i i ca ion o his e ec was necessa y a sho SOA. Expe imen 2 in oduced a new con ol condi ion wi hou o hog aphic o s ess ela ion (do MÍ- Table 3 Mean eac ion imes (and pe cen age o e o s) in Expe imen 3 Type o p iming Di e ences Di e en S ess asGÓ/RASgo Mo phological RASga/RASgo Un ela ed PERsa/RASgo D1 D2 D3 722 (2.99) 702 (2.55) 757 (5.16) 35 55 20 No e: D1 is he di e ence be ween Di e en S ess and Un ela ed pai s, D2 is he di e ence be ween Mo phological and Un ela ed pai s and D3 is he di e ence be ween Di e en S ess and Mo phological pai s The con as i e alue o lexical s ess in isual wo d ecogni ion: E idence om Spanish 281 RASgo). The goal was o compa e his condi ion wi h he con ol condi ion o he p e ious expe imen (PERsa-RASgo) o ob ain, as in ac occu ed, a ne acili a ion due only o s ess o e lap wi hou o hog aphic ela ion. The o hog aphic condi ion ( asGÓ-RASgo) also p oduced signi i can acili a ion on he o ally un ela ed pai s (do MÍ-RASgo). S ess compu a ion and o hog aphic p ocessing seem o wo k in a pa allel and independen way because o hog aphic p iming wi hou s ess o e lap p oduced acili a ion bu s ess p iming wi hou o hog aphic ela ion also p oduced simila acili a ion. Because p ime and a ge in he di e en s ess epe i ion condi ion a e also mo phological ela i es, Expe imen 3 used a longe ime window (SOA 224 ms.) o eco e he mo pho-seman ic p ocessing. Pai s such as asGÓ-RASgo a e in l ec ional o ms o he same e bal s em. I was expec ed ha he ea ly inhibi ion p oduced by a di e en s ess pa e n would be subs i u ed by la e mo phological and seman ic acili a ion. The esul s con i med he ini ial p edic ion. The o hog aphic pai s p oduced, in his expe imen , signi i can acili a ion. E en hough his acili a ion was 20 ms. smalle han ha ob ained o mo phological pai s (RASga-RASgo), he di e ence was no s a is ically signi i can and migh poin o a lesse impo ance o le e s in lexical disc imina ion as compa ed o s ess. In summa y, s ess could be conside ed a p elexical elemen ha he isual wo d ecogni ion sys em uses o a i e a he lexicon as also Gu ié ez-Palma and Palma-Reyes (2008) poin -ou , bu in con as wi h hei esul s, he con as i e alue o s ess in ou expe imen s has a e y ea ly e ec , pe haps due o he illegal pa e n o he o hog aphic ma ks used in he p imes o hei expe imen s, ha slow he ecogni ion o he a ge . Ou esul s a e in ag eemen wi h hose o Pe ea e al. (2016, 2018) suppo ing ea ly p ocessing o diac i ical ma ks in A abic language. These diac i ical ma ks a e poin s si ua ed, in di e en numbe and posi ion, on he le e s ha change he lexical alue o he wo d. Many A abic wo ds di e physically only in he p esence and/o loca ion o he diac i ical poin s. The accen in Spanish seems o ul i ll his ea ly con as i e unc ion. I phonemes, syllables o le e s a e conside ed p elexical uni s in ecognizing a wo d, i is logical ha lexical s ess is also a p elexical uni ha es ic s, by means o ecogni ion o he s essed syllable, he lexical i eld o sea ch o he amoun o ac i a ed uni s in he lexicon. This lexical unc ion o s ess appea s o be especially ac i e in Spanish, a language which, in con as wi h English, uses s ess on many occasions as he minimal ea u e ha dis inguishes pai s o wo ds wi hin he same g amma ical ca ego y o be ween ca ego ies (So o, Cu le , & Sebas ián, 2001). Spanish e bs cons i u e a special g oup o wo ds ha use s ess o dis inguish some o hei o ms. Two poin s, a leas , emain un esol ed. Fi s ly, wi h e e ence o he na u e o s ess p ocessing, we do no know whe he his is a isual o a phonological ope a ion du ing eading. Spanish s ess is o hog aphically ma ked wi h an accen (´) on he owel o he s essed syllable and is go e ned by speci i c ules amilia o eade s and w i e s. The p esence o absence o he accen and he consonan al o ocalic end o he wo ds indica es which is he s essed syllable o w i en wo ds. These ules allow he eade o access co ec ly he s ess pa e n o known o unknown wo ds. The e o e s ess, a p osodic and phonological cue, is in e ed in eading using only isual and p elexical in o ma ion. Ou da a do no allow o know whe he o no his isual ea u e is ans o med in o a phonological pa e n ha pe mi s lexical ac i a ion o sea ch o occu in a phonological lexicon, as may occu wi h le e s in o phonemes using g apheme o phoneme ules (Col hea , 1978). The second p oblem is mo e echnical bu no less impo an . The a ge s used in he expe imen s a e g amma ically ambiguous. RASgo as an indi idual wo d is a noun o a e bal o m ( i s pe son p esen o he e b asGAR). This aspec migh be conside ed a p oblem because some linguis s conside ha s ess assignmen is go e ned by di e en ules o e bs and nouns (Roca, 1988), bu his opinion is no ee o con o e sy (Edding on, 2000; Ha is, 1989). Ano he ques ion ha migh be impo an in ou s imuli is ha some o hese a e no only g amma ically ambiguous bu also seman ically ambiguous. RASgo means ea u e as a noun and o ea as a e b. Some o he a ge s a e only g amma ically ambiguous. Fo example, RObo as a e b, means o s eal and, as a noun, means obbe y. This is a e y in e es ing aspec in ela ion o s ess assignmen ha should be conside ed o u u e esea ch bu i was no , o he momen , he ocus o his esea ch. Essen ially, ou conclusion is ha he s ess pa e n o a wo d is p elexically compu ed. S ess seems o ha e, in Spanish, a c ucial unc ion in compa ison o o he languages, such as English, because he numbe o pai s o wo ds di e ing only in his cha ac e is ic is abundan . English s ess does no ha e, pe haps, his lexical unc ion bu i has, a he , a me ical u ili y o de i ne he bounda ies o wo ds in speech pe cep ion, such as has been a ibu ed by Cu le . I is no likely, hen, in English, o ob ain a esul such as ha ound in Spanish, as in isual wo d ecogni ion he wo ds appea segmen ed on he pape . Acknowledgemen s This esea ch was suppo ed by he Spanish Minis e io de Economía y Compe i i idad, PSI2015-64174-P and PSI2013- 47959-P. Re e ences Alameda, J.R. & Cue os, F. (1995). Dicciona io de ecuencias de las unidades lingü í s icas del cas ellano [Spanish dic iona y o equencies o he linguis ic uni s]. O iedo: Se icio de Publicaciones de la Uni e sidad de O iedo. Al a ez, C.J., U u ia, M., Dominguez, A., Sanchez-Casas, R. (2011). P ocessing in l ec ional and de i a ional mo phology: Elec ophysiological e idence om Spanish. Neu oscience Le e s, 490(1), 6-10. doi: 10.1016/j.neule .2010.12.015 Amen a, S. & C epaldi, D. (2012) Mo phological P ocessing as We Know I : An Analy ical Re iew o Mo phological E ec s in Visual Wo d Iden i i ca ion. F on ie s in Psychology, 3(232), 1-12 doi: 10.3389/ psyg.2012.00232 And ews, S. (1997) The e ec s o o hog aphic simila i y on lexical e ie al: esol ing neighbo hood con l ic s. Psychological Bulle in & Re iew, 4(4), 439-461. doi: 10.3758/BF03214334 Albe o Domínguez Ma ínez and Fe nando Cue os Vega 282 A ciuli, J. & Slowiaczek, L.M., (2007). The whe e and when o linguis ic wo d-le el p osody. Neu opsychologia, 45(11), 2638-2642. doi: 10.1016/j.neu opsychologia.2007.03.010 Black, M. & Byng, S. (1986). P osodic cons ain s on lexical access in eading. Cogni i e Neu opsychology, 3(4), 369-409. doi: 10.1080/02643298608252028 Col hea , M. (1978). Lexical access in simple eading ask. In G. Unde wood (Ed.), S a egies o In o ma ion p ocessing (pp. 151-216). San Diego, CA Academic P ess. Cu le , A. (1986) Fo bea is a homophone: lexical p osody does no cons ain lexical access. Language and Speech, 29(3), 201-220. doi: 10.1177/002383098602900302 Cu le , A. & Ca e , D.M. (1987). The p edominance o s ong ini ial syllables in he English ocabula y. Compu e Speech and Language, 2(3-4), 133-142. doi: 10.1016/0885-2308(87)90004-0 Cu le , A. & Cli on, C.E. (1985). The use o p osodic in o ma ion in wo d ecogni ion. In H. Bouma & D.G. Bouwhuis (Eds.) A en ion and Pe o mance X (pp 183-196). Hillsdale. NJ: Law ence E lbaum Associa es. Domí nguez, A., de Vega, M. & Ba be , H. (2004). E en Rela ed Po en ials elici ed by mo phological, homog aphic and seman ic p iming. Jou nal o Cogni i e Neu oscience, 16(4), 598-608. doi: 10.1162/ 089892904323057326 Domí nguez, A., Seguí , J. & Cue os, F. (2002). The ime cou se o in l ec ional mo phological p iming. Linguis ics, 40(2), 235-259. doi: 10.1515/ling.2002.011 Dupoux, E.; Pallie , C.; Sebas ián, N.; Mehle , J. (1997). A des essing dea ness in F ench. Jou nal o Memo y and Language, 36(3), 406- 421. doi: 10.1006/jmla.1996.2500 Edding on, D. (2000). Spanish s ess assignmen wi hin Analogical Modeling o Language. Language, 76(1), 92-109. doi: 10.1353/ lan.2000.0022 Fo bach, G.B., S anne s, R.F. & Hochhaus, L. (1974). Repe i ion and p ac ice e ec s in a lexical decision ask. Memo y & Cogni ion, 2(2), 337-339. 10.3758/BF03209005 Fo s e , K. & Da is, C. (1984). Repe i ion p iming and he equency a enua ion in lexical access. Jou nal o Expe imen al Psychology: Lea ning, Memo y and Cogni ion, 10(4), 680-698. doi: 10.1037/0278- 7393.10.4.680 G ainge , J. (1992). O hog aphic neighbo hoods and isual wo d ecogni ion. In R. F os and L. Ka z (Eds.) O hog ahy, phonology, mo phology and meaning (pp. 131-146. Ams e dam: Else ie Science Publishe s. G ainge , J., Cole, P. & Segui, J. (1991). Masked mo phological p iming in isual wo d ecogni ion. Jou nal o Memo y and Language, 30(3), 370-384. doi: 10.1016/0749-596X(91)90042-I Gu ié ez, N., Palma, A., y San iago, J. (1998, Sep embe ). On he ole o lexical s ess in isual wo d ecogni ion. Pape p esen ed a he XI Eu opean Socie y o Cogni i e Psychology con e ence, Je usalem, Is ael. Gu ié ez-Palma, N. & Palma-Reyes, A. (2008). On he use o lexical s ess in eading Spanish. Reading and W i ing, 21(6), 645-660. 10.1007/s11145-007-9082-x Ha is, J.W. (1989). How di e en is e b s ess in Spanish. P obus ,1(3), 241-258. doi: 10.1515/p bs.1989.1.3.241 Ma ce , A., & Pe ea, M. (2017). Is ne al NEUTRAL? Visual simila i y e ec s in he ea ly phases o w i en-wo d ecogni ion. Psychonomic Bulle in & Re iew, 24(4), 1180–1185. doi: 10.3758/s13423-016-1180-9 Mo ales-Fon , A. (2014). El acen o [The accen ]. In R. A. Núñez., S. Colina., y T. G. B adley (Eds.). Fonología gene a i a con empo ánea de la lengua española [Con empo a y gene a i e phonology o he Spanish language] (2nd Edi ion) (pp. 235-265). Washing on, DC: Geo ge own Uni e si y P ess. Pensado, C. (1999). Mo ología y onología, enómenos mo o onológicos [Mo phology and phonology, mo phophonological phenomena]. In I. Bosque & V. Demon e (Eds.) G amá ica Desc ip i a de la Lengua Española. Vol. 3: En e la o ación y el discu so. Mo ología [Desc ip i e G amma o he Spanish language. Vol.3: Be ween he sen ence and de discou se] (pp. 4423-4504). Mad id. Espasa Calpe. Pe ea, M., Abu Mallouh, R., Mohammed, A., Khali a, B., & Ca ei as, M. (2016). Do diac i ical ma ks play a ole a he ea ly s ages o wo d ecogni ion in A abic? F on ie s in Psychology, 22(7), 1255. doi:10.3389/ psyg.2016.01255 Pe ea, M., Abu Mallouh, R., Mohammed, A., Khali a, B., & Ca ei as, M. (2018). Does isual le e simila i y modula e masked o m p iming in young eade s o A abic? Jou nal o Expe imen al Child Psychology, 169, 110-117. doi: 10.1016/j.jecp.2017.12.004 Pe ea, M., & Rosa, E. (2000). Repe i ion and o m p iming in e ac wi h neighbo hood densi y a a sho s imulus onse asynch ony. Psychonomic Bulle in and Re iew, 7(4), 668-677. doi: 10.3758/ BF03213005 Roca, I. (1988). Theo e ical implica ions o Spanish wo d s ess. Linguis ic Inqui y, 19(3), 393-423. Re ie ed om h p://www.js o . o g/s able/25164902. Sca bo ough, D. L., Co ese, C. & Sca bo ough, H. (1977). F equency and epe i ion e ec s in lexical memo y. Jou nal o Expe imen al Psychology: Human Pe cep ion and Pe o mance, 3(1), 1-17. doi: 10.1037/0096-1523.3.1.1 Segui, J. & G ainge , J. (1990). P iming wo d ecogni ion wi h o hog aphic neighbo s: E ec s o ela i e p ime a ge equency. Jou nal o Expe imen al Psychology: Human Pe cep ion and Pe o mance, 16(1), 65-76. doi: 10.1037/0096-1523.16.1.65 So o, S., Cu le , A. and Sebas ián, N. (2001). Segmen al and sup asegmen al misma ch in lexical access. Jou nal o Memo y and Language, 45(3), 412-432. doi: 10.1006/jmla.2000.2783 Sulpizio, S., Spinelli, G., & Bu ani, C. (2015). S ess a ec s a icula o y planning in eading aloud. Jou nal o Expe imen al Psychology: Human Pe cep ion and Pe o mance, 41(6), 453-461. doi: 10.1037/ xhp0000163