Hemispheric specialization for language according to grapho ...

Brain and Cognition 69 (2009) 465–471

Contents lists available at ScienceDirect

Brain and Cognition journal homepage: www.elsevier.com/locate/b&c

Hemispheric specialization for language according to grapho-phonemic transformation and gender. A divided visual field experiment Emilie Cousin, Marcela Perrone, Monica Baciu * Laboratoire de Psychologie et Neurocognition, UMR CNRS 5105, Université Pierre Mendès-France, BP 47, 38040 Grenoble Cedex 09, France

a r t i c l e

i n f o

Article history: Accepted 9 October 2008 Available online 18 November 2008 Keywords: Hemisphere Specialization Divided visual field Phonology Gender Grapho-phonemic translation Transparent Non-transparent Healthy

a b s t r a c t This behavioral study aimed at assessing the effect of two variables on the degree of hemispheric specialization for language. One of them was the grapho-phonemic translation (transformation) (letter-sound mapping) and the other was the participants’ gender. The experiment was conducted with healthy volunteers. A divided visual field procedure has been used to perform a phoneme detection task implying either regular (transparent) grapho-phonemic translation (letter-sound mapping consistency) or irregular (non-transparent) grapho-phonemic translation (letter-sound mapping inconsistency). Our results reveal a significant effect of grapho-phonemic translation on the degree of hemispheric dominance for language. The phoneme detection on items with transparent translation (TT) was performed faster than phoneme detection on items with non-transparent translation (NTT). This effect seems to be due to faster identification of TT than NTT when the items were presented in the left visual field (LVF)-right hemisphere (RH). There was no difference between TT and NTT for stimuli presented in the right visual field (RVF)-left hemisphere (LH). This result suggests that grapho-phonemic translation or the degree of transparency can affect the degree of hemispheric specialization, by modulating the right hemisphere activity. With respect to gender, male participants were significantly more lateralized than female participants but no interaction was observed between gender and degree of transparency. Ó 2008 Elsevier Inc. All rights reserved.

1. Introduction For the majority of individuals, language functions depend on the left dominant hemisphere (Chiarello, Kacinik, Manowitz, Otto, & Leonard, 2004; Springer & Deutsch, 2000). The present research used the divided visual field procedure (DVF) since it is, among the various methods used to assess the hemispheric specialization for language operations, a reliable and easy-to-use behavioral technique. This procedure exploits the fact that visual pathways are partially crossed (Bourne, 2006; Chiarello et al., 2004). Due to this anatomical specificity, a stimulus that is flashed in half of one visual field will be first processed by the opposite hemisphere. Thus, when presenting a stimulus in the left visual field (LVF) it will be first processed by the right hemisphere (RH) and when presenting a stimulus in the right visual field (RVF) it will first be processed by the left hemisphere (LH). The logic underlying this procedure for assessing language hemispheric dominance is that linguistic stimuli are processed faster and more efficiently if they are presented first to the hemisphere specialized for language, generally the left one (Bourne, 2006; Hunter & Brysbaert, 2008). Furthermore, the DVF procedure allows the assessment of LH dominance for phono* Corresponding author. E-mail addresses: [email protected], [email protected] (M. Baciu). 0278-2626/$ - see front matter Ó 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.bandc.2008.10.001

logical (Coney, 2002; D’Hondt & Leybaert, 2003; Lukatela, Carello, Savic, & Turvey, 1986; Tremblay, Ansado, Walter, & Joanette, 2007; Tremblay, Monetta, & Joanette, 2004, 2007) and semantic (Tremblay, Ansado, et al., 2007; Tremblay, Monetta, et al., 2007) processes. Semantic (Kahlaoui & Joanette, 2008; Schmidt, DeBuse, & Seger, 2007; Tremblay, Ansado, et al., 2007; Tremblay et al., 2004, 2007) and phonological (Cousin, Peyrin, & Baciu, 2006; Lavidor, Johnston, & Snowling, 2006; Tremblay, Ansado, et al., 2007; Tremblay et al., 2004, 2007) operations seem to induce various degrees of hemispheric lateralization. Phonological processes operate with the auditory representations that are related to the graphophonemic translations of visual stimuli (Alario, Schiller, DomotoReilly, & Caramazza, 2003; Simon, Bernard, Lalonde, & Rebai, 2006; Walter, Cliche, Joubert, Beauregard, & Joanette, 2001). Experiments conducted with the DVF procedure and neuroimaging (Baciu et al., 2001; Kareken, Lowe, Chen, Lurito, & Mathews, 2000; Lurito, Kareken, Lowe, Chen, & Mathews, 2000) revealed that they are significantly more lateralized than semantic processes which deal with meaning (Coney, 2002; D’Hondt and Leybaert, 2003; Lukatela et al., 1986; Tremblay, Ansado, et al., 2007; Tremblay et al., 2004, 2007). Although phonological operations largely depend on the left hemisphere, the left hemispheric dominance is not exclusive (absolute) because there are some arguments suggesting right hemisphere abilities during phonological processing. A large

466

E. Cousin et al. / Brain and Cognition 69 (2009) 465–471

majority were provided by neuropsychological observations (Zaidel & Peters, 1981). Using DVF, Smolka and Eviatar (2006) described a commissurotomy patient (L.B from Zaidel & Peters, 1981) who was able to perform a simple phonological task (i.e. ideographic reading) not only when the stimuli were presented to the dominant LH (RVF) but also when they were presented to the non-dominant RH (LVF). Because in this patient the two hemispheres did not communicate, the authors concluded that the RH could solve easy language operations and thus has some language capacities. The degree of hemispheric specialization for language operations can be modified by manipulating tasks and/or psycholinguistic characteristics of stimuli, such as concreteness (Chiarello, Liu, & Shears, 2001), word imageabilty (Chiarello, Shears, Liu, & Kacinik, 2005), orthographic depth (Frost & Katz, 1989) or the type of grapho-phonemic translation (Cousin et al., 2006; Crossman & Polich, 1988; Khateb et al., 2000; Tremblay, Ansado, et al., 2007; Tremblay et al., 2004, 2007; Waldie & Mosley, 2000; Simon et al., 2006; Walter et al., 2001). Processing visual words implies the translation from graphemic (letter) to phonemic (corresponding sound) representations. When the grapho-phonemic relation is regular (transparent translation, TT) there is a consistent letter-sound mapping. When it is irregular (non-transparent translation, NTT) the letter-sound mapping is inconsistent. Languages with regular grapho-phonemic relationships such as Italian, Spanish or German are considered as transparent. English is less transparent because the grapho-phonological conversion rules are not always consistent. There are different degrees of grapho-phonemic transparency. French language is more non-transparent (opaque) than Italian, but more transparent than English. Thus, in French there are words with regular grapho-phonemic translation and others with irregular grapho-phonemic translation (Yvon et al., 1998). For instance, the phonological form of the word ‘‘auto” (self) contains one phoneme with an irregular (non-transparent) relation between orthography (‘‘au”) and sound (/o/). But the phonological form of the word ‘‘auto” contains another phoneme with a regular (transparent) translation between orthography (‘‘o”) and sound (/o/). Another example is the phonological form of the French word ‘‘femme” (woman). The first letter ‘‘e” from the words corresponds to the sound /a/. This is an example of irregular or NTT letter-sound correspondence. Using the DVF procedure, Tremblay et al. (2004) explored whether the degree of grapho-phonemic transparency affects the degree of hemispheric specialization for language. The authors showed a significant effect of this factor on the degree of hemispheric specialization. The results suggest that a phoneme detection task in TT stimuli can be performed by both hemispheres. That is because phoneme detection in TT phonemes would mainly require a visual global analysis, which is more based on letter identification than grapho-phonemic translation. Both hemispheres are able to perform a visual global analysis, but only the LH is able to perform grapho-phonemic translations. The results of Tremblay et al. (2004) are in agreement with those founded by Lukatela et al. (1986) which had manipulated the phonological ambiguity in Serbo-croat. They have found that phonological ambiguity had strong effects in the right hemisphere. This result presupposes that both hemispheres have different abilities in order to process phonology. Moreover, Smolka and Eviatar (2006) have performed a DVF study by using an interference paradigm in unvoweled Hebrew script to assess hemispheres phonological abilities (i.e. phonemically, orthographically and figurally incorrect vowel information conflicted with the consonant of words in each visual field). Their interference pattern obtained indicates that the left hemisphere automatically transforms grapheme into phonological information whereas right hemisphere process grapheme as visual forms.

In this perspective, the Simon et al. study (2006) explored the effect of orthographic transparency on event-related potentials (ERP) in bilingual French-Arab readers. French, unlike Arab, was expected to favor the use of grapheme-phoneme conversion rules during reading. Previous results obtained by the same team showed that N320 is a good candidate for grapheme–phoneme conversion processing in languages such as French. Indeed, when reading in their mother tongue, only French subjects clearly elicited a N320. Moreover, the comparisons between activations elicited by Arabic words in Arab subjects and French monolingual people also confirm that the N170 component represents an important orthographic stage. These results may have implications concerning visual word recognition and suggest the existence of at least two mechanisms in the access to word meaning. The Simon et al. study (2006) does not directly tell us about transparency but based on their results we can make indirect assumptions about how specific ERP components reflect transparency. More specifically, the N320 was elicited by the graphemephoneme translation condition. Assimilated to our ‘‘non-transparent” condition, the N320 component can be indirectly considered as a marker of transparency: the N320 would be elicited or has higher amplitude for ‘‘non-transparent” than for ‘‘transparent” condition. Although some other studies show a relationship between N320 and transparency, the interpretation of it is not related to linguistic processes. More precisely a study performed by Tremblay, Ansado, et al. (2007) and Tremblay, Monetta, et al. (2007) for a phonological judgement in ‘‘transparent” and ‘‘nontransparent” words compared whether a variation of graphemesto-phonemes difficulty has an impact on the N320 and N400. Although the N320 was elicited by both conditions, the authors surprisingly showed that the amplitude of this ERP component was higher for ‘‘transparent” (easy) than for ‘‘non-transparent” (difficult) condition. The authors interpret their results according to the idea advanced by Proverbio, Vecchi, and Zani (2004), namely that N320 reflects a stage in the graphemes-to-phonemes conversion mechanism and its larger amplitude represents the participant’s confidence in their responses. The N320 and N170-180 elicited during a phonological task are recorded on the anterior fronto-central region (N320) and on the occipito-temporal region (N170-180). They are both left hemisphere lateralized (Proverbio et al., 2004). Overall, these findings suggest that the effect of psycholinguistic variables such as transparency on the amplitude of ERPs do not necessary reflect processing stages but rather this effect could more simply index the brain’s sensitivity to the orthographic regularity. Thus, the origin of the relationship between ERP components and psycholinguistic variables such as transparency is not completely understood and further investigation is necessary for clarification. In a previous behavioral DVF experiment (Cousin et al., 2006) we compared the degree of hemispheric lateralization for two phonological (rhyme detection and phoneme detection) tasks. We have shown that rhyme detection was more lateralized than phoneme detection. The reduced RVF-LH advantage for phoneme compared to rhyme detection was induced by different degrees of grapho-phonemic translation. Actually, the target-phoneme during the phoneme detection task, /b/, required a regular (transparent) grapho-phonemic translation from the letter ‘‘b”. As mentioned above, both hemispheres are equally able to perform this task (Tremblay, Ansado, et al., 2007; Tremblay et al., 2004, 2007) since performing this task involves visual detection. This result is in agreement with a PET study (Cappa, Perani, Schnur, Tettamanti, & Fazio, 1998) showing that a letter detection task activates very weakly the language regions and mainly involves the visual regions, mostly in a bilateral fashion. If the TT phonemes detection can be assimilated to a letter detection task, it proves again that it mainly involves non lateralized visual regions.


467

Other variables affect the degree of hemispheric specialization for language, such as handedness (Bradshaw & Taylor, 1979; Piazza, 1980; Tremblay et al., 2004) and gender (Coney, 2002; Majeres, 1999). Behavioral (Coney, 2002; Majeres, 1999) and neuroimaging studies (Clements et al., 2006; Pugh et al., 1996; Shaywitz et al., 1995; Voyer, 1996) showed that females are less lateralized than males for phonological processing. No difference has been observed for semantic tasks as a function of gender (Frost et al., 1999; Pugh et al., 1996; Shaywitz et al., 1995). The aim of this DVF experiment carried out with healthy righthanded males and females was to assess the effect of grapho-phonemic translation (T or NTT) and gender on the hemispheric specialization for language during a phoneme detection task.

in white ‘‘Courrier New” font size 24 and centered on the middle of a white screen. The pseudo-words were presented randomly among participants. The ‘‘control condition” stimuli were 96 items (Fig. 1) composed of unreadable characters (Karalyn Patterson font). They were the same length as the linguistic stimuli. We used the control items to make sure that the inter-hemispheric difference we expected to obtain for the linguistic stimuli was not due to visual processing but to a language activity, as no inter-hemispheric difference was expected for the control items. Each trial (language or control stimulus) started with a fixation cross and was followed by a visual mask (composed of seven stars) in order to remove retinal persistence.

2. Material and methods

2.3. Tasks and procedure

2.1. Participants

The subjects were instructed to perform a language (phoneme detection in pseudo-words) and a control (visual detection in unreadable characters) task. During phoneme detection the participants had to judge whether the pseudo-words contained the sound /o/. During the presentation of the control items, the participants were instructed to judge whether the items contained at least one character which overshot the others. More specifically it means that participants had to detect if among all the Karalyn Patterson characters forming an unreadable word, there was at least one which was spatially higher than the other ones. After receiving the instructions, the participants went through a short training session with items that were different from those presented during the DVF experiment. During the experiment, each participant was tested individually in a darkened quiet room. The subjects were seated in front of a computer monitor (screen resolution 1024 by 768 pixels), located at 110 cm in front of them. The experiment was built by means of E-Prime software (E-Prime Psychology Software Tools Inc., Pittsburgh, USA). Each trial began with a fixation cross of 500 msec duration in order to keep the gaze direction the center of the screen. Then, the stimulus was displayed, either in the LVF or in the RVF was displayed for 130 msec. This short presentation time insured a mono-hemispheric presentation. Each item was followed by a visual mask composed of symbols (sequence of seven stars) which lasted 30 msec. The inner and the outer edges of the lateralized presented stimuli located 2° and 6° from fixation, respectively. Finally, the trial ended with a fixation cross of 1500 msec duration. The answers (yes or no) were transmitted by means of two response keys pressed by the index and the middle fingers of the

Seventeen healthy males (mean of age 23.6 y) and seventeen healthy females (mean of age 22.9 y) participated in the experiment. They were native French speakers. They were graduate and undergraduate students and received course credit at the University for their participation. They were all right-handed according to the Edinburgh Handedness Inventory (Oldfield, 1971) and had normal or corrected-to-normal vision. They all gave their informed consent to the study. 2.2. Stimuli We presented two types of stimuli: ‘‘language condition” and ‘‘control condition”. The ‘‘language condition” stimuli were 96 legal pseudo-words (Fig. 1) built by exchanging two or three letters in French concrete 7 letter nouns. They were displayed randomly, 48 in the LVF (i.e. 24 targets and 24 non-targets items) and 48 in the RVF (i.e. 24 targets and 24 non-targets items). The target-items (items to be detected) contained the phoneme /o/ varying by grapho-phonemic transparency. Half of target-items were TT (contained the grapheme ‘‘o” and the phoneme /o/; the graphophonemic transformation was considered regular). The other half were NTT (contained the grapheme ‘‘au” and the phoneme /o/; the grapho-phonemic transformation was considered irregular). The target-phoneme was placed an equivalent number of times at the beginning [i.e. autril (NTT) or obalet (TT)], in the middle [i.e. phaumi (NTT) or clotir (TT)] and at the end [i.e. amilau (NTT) or damulo (TT)] of the pseudo-words. All the stimuli were written

Fig. 1. Example of stimuli presented during the experiment. A language trial (left side) consists of: fixation cross, (a) task item (pseudo-words containing or not the phoneme /o/) and visual mask. A control trial (right side) consists of: fixation cross, (b) control item (unreadable words containing or not at least one Karalyn Patterson character which is higher than the other ones).

468


right hand (for one half of the items) and of the left hand (for another half of the items). The response hand was counterbalanced between subjects. The reaction time (RT) and the accuracy (% correct) were recorded for each participant. The entire experiment lasted 12 min. Data processing was performed by using an analysis of variance (ANOVA). We used a two-step analysis. The first step consisted of a general analysis of variance ANOVA including all subjects and all items of both experimental states (language and control), in order to make sure that the inter-hemispheric difference in fact was due to language processing. We expected to observe a hemispheric specialization only for the language condition but not for the control condition. If the difference between the visual fields (hemispheres) was significant for one condition, the second step was to examine whether there was gender and grapho-phonemic translation effect on hemispheric specialization. Moreover, we checked (t-test) that the Correct Responses Rates (% CR) for the main condition (RVF-LH vs. LVF-RH) were above the chance level. All results are reported in Table 1.

Fig. 2. This figure shows in terms of response time (RT msec) a significant interaction between visual field of presentation and experimental conditions. The inter-hemispheric difference was significant only for language condition (p < 0.05).

3. Results 3.1. First step analysis We performed an analysis of variance including all stimuli, pseudo-words and control items. After verifying the assumption of homogeneity of variance, a 2 2 ANOVA was performed on RT (F1) and CR rate (F2) with two within-subject factors with two conditions, Visual field of presentation (LVF-RH, RVF-RH) and Experimental state (Language; pseudo-words, Control; Karalyn Patterson items). All results are presented in terms of reaction times (mean response time, RT) and Accuracy (% correct responses, CR). 3.1.1. Response time For all subjects, a significant interaction was obtained between visual field of presentation and experimental state [F1(1,33) = 4.06, MSe = 5565.6, p < 0.05] suggesting different patterns of inter-hemispheric asymmetry across conditions. We performed planned comparisons on this interaction and the results showed significant effect of the visual field of presentation for pseudo-words [F1(1,33) = 7.30, MSe = 9764.2, p < 0.05] with significantly faster responses for items presented in the RVF (837.6 msec) than for items presented in the LVF (902.44 msec). We did not obtain a significant effect of the visual field of presentation for the control condition [F1(1,33) = 1.50, MSe = 1959.1, p = 0.22]. The RT was similar for items presented in the RVF (747.8 msec) and for items presented in the LVF (761 msec) (Fig. 2).

3.1.2. Accuracy For all subjects, a significant interaction was obtained between visual field of presentation and experimental states [F2(1,33) = 8.67, MSe = 0.004, p < 0.05] suggesting different patterns of inter-hemispheric asymmetry across conditions. We performed planned comparisons on this interaction and the results showed a significant effect of the visual field of presentation for pseudo-words [F2(1,33) = 15.63, MSe = 0.01, p < 0.001] with more accurate responses for items presented in the RVF (70%) than for items presented in the LVF (61%). We did not obtain any significant effect of the visual field during the control condition [F2(1,33) = 3.39, MSe = 0.0036, p = 0.08], the %CR was similar for items presented in the RVF (78%) and for items presented in the LVF (74%) (Fig. 3). 3.2. Second analysis The previous ANOVA analysis showed hemispheric specialization only for the language condition. Thus, in this second ANOVA analysis we considered only the language condition and we assessed the effect of two variables, gender and grapho-phonemic translation on hemispheric specialization. After verifying the assumption of homogeneity of variance, a 2 2 2 ANOVA was performed on RT (F1) and CR rate (F2) with two within-subject fac-

Table 1 The mean correct responses rates (% CR) and the mean response time (RT msec) for stimuli presented in the left visual field (LVF-RH) and right visual field (RVF-LH) for each gender. t-Test for CR rate against chance level (50%) values and corresponding pvalues are reported. Gender

Male

Visual field of presentation n participants Mean response time (msec) Standard error (msec) Mean correct response (%) Standard error (%) t-Value (against chance level) p-Value

LVF-RH 12 851 246.4 65% 16% 4.04 0.0003

Female RVF-LH 12 765.2 208.1 75% 17% 6.21 5.9E07

LVF-RH 12 896.5 225.9 57% 12% 2.42 0.0002

RVF-LH 12 898.6 118.8 66% 17% 4.09 0.0215

Fig. 3. This figure shows in terms of correct response rate (CR %) significant interaction between visual field of presentation and experimental conditions. The inter-hemispheric difference was significant only for language condition (p < 0.05).


tor with two conditions, Visual field of presentation (LVF-RH, RVFRH) and Grapho-phonemic transparency (NTT, TT) and one between-subject factor with two conditions, Gender (Male, Female). All results are presented in terms of reaction times (mean response time, RT) and Accuracy (% correct responses, CR). 3.2.1. Response time A significant effect of the visual field of presentation was obtained [F1(1,32) = 4.62, MSe = 14347.2, p < 0.05], pseudo-words presented in the RVF (831.2 msec) were processed faster than those presented in the LVF (875.3 msec). We also obtained a significant effect of the type of grapho-phonemic translation [F1(1,32) = 6.43, MSe = 13264.4, p < 0.05], TT induced faster responses (828.2 msec) than NTT (878.3 msec). A significant interaction was obtained between the visual field of presentation and gender. It suggests different patterns of inter-hemispheric asymmetry in male and in female participants [F1(1,32) = 5.16, MSe = 14347.2, p < 0.05]. We performed planned comparisons on this interaction and the results showed significant effect of the visual field of presentation for males [F1(1,32) = 9.77, MSe = 14347.2, p < 0.05] with faster responses for items presented in the RVF (765.5 msec) than for items presented in the LVF (856.3 msec). We did not obtain any significant effect of the visual field [F1(1,32) = 0.007, MSe = 14347.2, p = 0.93] in females, RTs were similar for items presented in the RVF (896.9 msec) and for items presented in the LVF (894.4 msec) (Fig. 4). The significant interaction between gender and visual field of presentation showed visual field presentation effect (i.e. faster responses for items presented in the RVF) only in the male population. For this reason, in the following section, we explored the effect of grapho-phonemic translation only in male participants. 3.2.2. Response time for the male population A significant interaction was obtained between visual field of presentation and the grapho-phonemic translation type for males [F1(1,32) = 5.16, MSe = 73.80, p < 0.01] suggesting different patterns of inter-hemispheric asymmetry for NTT and for TT. We performed planned comparisons on this interaction and the results showed a significant effect of the visual field of presentation for NTT [F1(1,32) = 14.61, MSe = 13095.1, p < 0.001] with significantly faster responses for items presented in the RVF (777.2 msec) than for

Fig. 4. This figure shows in terms of response time (RT msec) the significant effect of gender on the hemispheric specialization according to the visual field of presentation. Males responded faster for stimuli presented in the RVF-LH than for those presented in the LVF-RH. Females showed no difference.

469

items presented in the LVF (927.2 msec). We did not obtain any significant effect of the visual field of presentation [F1(1,32) = 0.98, MSe = 8632.5, p = 0.32] for TT, the RT was similar for items presented in the RVF (753.8 msec) and for items presented in the LVF (785.5 msec). The grapho-phonemic translation effect was specifically due to stimuli presented in the LVF-RH. Actually, when stimuli were presented first in the LVF-RH, the subjects responded faster for TT (785.5 msec) than for NTT (977.2 msec) [F1(1,32) = 15.05, MSe = 11349.41, p < 0.001] (see Fig. 5). 3.2.3. Accuracy A significant effect of the visual field of presentation was obtained [F2(1,32) = 15.17, MSe = 0.021, p < 0.005]. The pseudo-words presented in the RVF (70%) were processed more efficiently than those presented in the LVF (61%). We also obtained a significant effect of the grapho-phonemic translation [F2(1,32) = 20.07, MSe = 0.032, p < 0.005], as TT (72%) were processed more efficiently than NTT (59%). Significant interactions were not found between visual field of presentation and gender [F2(1,32) = 0.02, MSe = 0.021, p = 0.88], nor between grapho-phonemic translation type and visual field of presentation [F2(1,32) = 0.27, MSe = 0.015, p = 0.60]. 3.3. Synthesis of results The main results obtained in this study are: Left hemispheric specialization during phoneme detection is observed only in male participants. Significant effect of grapho-phonemic type: phoneme detection NTT induced higher degree of lateralization than phoneme detection TT. Modulation of RH involvement by the type of grapho-phonemic translation: TT phoneme detection can be equally performed by both hemispheres as visual processing is necessary to perform this task.

Fig. 5. This figure shows in terms of response time (RT msec) significant interaction between visual field of presentation and grapho-phonemic translation. Although phoneme detection task induced significant advantage of the RVF (LH), the interhemispheric difference was significantly greater (p < 0.001) for NTT than for TT. The reduction of the inter-hemispheric difference was due to faster responses on the LVF-RH for TT than for NTT.

470


4. Discussion The aim of this DVF experiment was to explore the hemispheric dominance for language in a French phoneme detection task. We investigated the effects of grapho-phonemic translation and gender. All the participants were left hemisphere lateralized in the language condition but not in the control condition. This result was in agreement with previous behavioral (Chiarello et al., 2004; Cousin et al., 2006; Smolka & Eviatar, 2006; Tremblay, Ansado, et al., 2007; Tremblay et al., 2004, 2007; Waldie & Mosley, 2000) and neuroimaging (Baciu, Juphard, Cousin, & Le Bas, 2005; Baciu et al., 2001; Kareken et al., 2000; Lurito et al., 2000; Simon et al., 2006) studies. 4.1. Effect of gender The second analysis of variance has shown that overall, only males were hemisphere lateralized during phoneme detection. They showed a significant advantage of the right visual field of presentation. This result is in agreement with those provided by Coney (2002) in a DVF experiment, in which the authors explored pseudo-homophone detection (to judge if the pseudo-word ‘‘sounds like an ordinary English word”) and a rhyme-matching task. For these tasks, both involving grapho-phonemic transformation, males were significantly more lateralized than females. These results are in line with neuroimaging studies showing higher lateralization in males during language tasks (Eviatar, Hellige, & Zaidel, 1997; Shaywitz et al., 1995). Using fMRI for mapping language during a phonological task (rhyming detection), Clements et al. (2006) showed bilateral activation of the inferior frontal gyrus (Broca’s area to the left) for females and only left for males. Our results are also supported by neuropsychological data. Aphasia caused by left hemisphere lesions is less frequently observed in females than in males (Springer & Deutsch, 2000). In terms of accuracy, males and females were equally lateralized. The left hemisphere lateralization for phoneme detection in males but not in females confirms results obtained by other studies. For instance, Lukatela et al. (1986) have shown that phonological difficulty induces supplementary right hemisphere involvement in females with respect to males. The authors say that this pattern is consistent with suggestions in the literature that if the right hemisphere is linguistically competent, particularly with respect to something as difficult as phonology, then it is more likely to be so in females. The ‘‘extra” language capacity of the right hemisphere underlies the traditional female verbal superiority (and spatial inferiority) as spatial processing space has been usurped by language (Bradshaw, Gates, & Nettleton, 1977). Our result showing the effect of gender on the hemispheric asymmetry during a phonological task is in agreement with this view. 4.2. Effect of grapho-phonemic translation As mentioned above, for some of the tested items, the graphophonemic translation was regular (TT) and for others it was irregular (NTT). Since only male participants showed significant hemispheric specialization for this task, we only explored the effect of grapho-phonemic translation in males. We observed a significant effect of grapho-phonemic translation on the degree of hemispheric specialization, as observed in previous studies (Bourne, 2006; Walter et al., 2001). This effect was only observed for the items presented in the LVF-RH, inducing faster responses for TT than for NTT. There were no significant differences between TT and NTT when the stimuli were presented in the RVF-LH. This result suggests that TT induces the participation of both hemispheres. As grapho-phonemic translation is regular, the TT task is performed on the basis of visual processing of the letter (grapheme). The visual letter identification would only involve the visual

regions and their activation is not lateralized (Cappa et al., 1998). Indeed, a letter detection task seems to involve processes that do not depend on a predominant hemisphere. This observation is in line with other results provided on Arabic (Eviatar, Ibrahim, & Ganayim, 2004) and Japanese (Hatta, 1977; Sasanuma, Itoh, Mori, & Kobayashi, 1977) linguistic items. Japanese has two modalities of writing. The ideogram code uses symbols (Kanji) and the phonogram code (Kana) uses ‘‘graphemes”; the latter implies that reading can be sequential. The results show that the RH is predominantly involved in Kanji (needs holistic visual processing) and the LH is predominantly involved in Kana reading (needs grapheme-phoneme translation). Furthermore, a recent fMRI study (Nakamura et al., 2005) showed left lateralized activations during phonogram presentation, while ideograms activate both hemispheres. The type of grapho-phonemic translation, regular or irregular, seems to modulate RH involvement. This hemisphere could be able to perform a language task if the stimuli do not require graphophonemic translation, as the TT items in our study. In this situation, the processes involved in performing the task seem to be more visual than linguistic. Thus, the RH could be crucial to perform a very quick recognition process on words, based on visualorthographic characteristics (Halderman & Chiarello, 2005). The visual word form recognition (i.e. abstract representation of visual word form) could be performed within the left temporo-occipital area near the fusiform gyrus, region called visual word form area (VWFA). However, Cohen et al. (2003) have shown that patients with lesions of the left VWFA are still able to read letter-by-letter and this ability would be due to the right homolog of the VWFA, activated in these patients (Cohen et al., 2003). In conclusion, optimal phonological processing seems to require the participation of both hemispheres (Chiarello et al., 2004; Tremblay et al., 2004). Whereas the left hemisphere automatically transforms graphemes into the corresponding phonemes, the right hemisphere uses a visual strategy and phonemes are processed like visual objects (Smolka & Eviatar, 2006). Our results revealed that the effect of grapho-phonemic transformation on the degree of hemispheric dominance was significant for response time but not for accuracy. If there is no interaction in the accuracy data, it does not mean that the two hemispheres are equally active. The RT values were different for the two visual fields. We just do not know how to relate this measure to the hemispheres’ abilities. Accuracy could also be a measure too imprecise for depicting effects as small as those related to variation of the grapho-phonemic translation.

5. Conclusion The results of this DVF experiment show left hemisphere dominance for a phoneme detection task only in male participants. The stimuli were processed faster and more accurately when they were presented in the RVF-LF than in the LVF-RH. In males, when stimuli were presented in the LVF-RH, the task was performed faster for the TT than the NTT condition. This result suggests a modulation of the RH involvement according to grapho-phonemic translation. References Alario, F. X., Schiller, N. O., Domoto-Reilly, K., & Caramazza, A. (2003). The role of phonological and orthographic information in lexical selection. Brain and Language, 84, 372–398. Baciu, M., Juphard, A., Cousin, E., & Le Bas, J. F. (2005). Evaluating fMRI methods for assessing hemispheric language dominance in healthy subjects. European Journal of Radiology, 55, 209–218. Baciu, M., Kahane, P., Minotti, L., Charnallet, A., David, D., Le Bas, J. F., et al. (2001). Functional MRI assessment of the hemispheric predominance for language in epileptic patients using a simple rhyme detection task. Epileptic Disorder, 3, 117–124.

E. Cousin et al. / Brain and Cognition 69 (2009) 465–471 Bourne, V. J. (2006). The divided visual field paradigm: Methodological considerations. Laterality, 11, 373–393. Bradshaw, J. L., Gates, E. A., & Nettleton, N. C. (1977). Bihemispheric involvement in lexical decisions: Handedness and a possible sex difference. Neuropsychologia, 15, 277–286. Bradshaw, J. L., & Taylor, M. J. (1979). A word-naming deficit in nonfamilial sinistrals? Laterality effects of vocal responses to tachistoscopically presented letter strings. Neuropsychologia, 17, 21–32. Cappa, S. F., Perani, D., Schnur, T., Tettamanti, M., & Fazio, F. (1998). The effects of semantic category and knowledge type on lexical-semantic access: A PET study. Neuroimage, 8, 350–359. Chiarello, C., Kacinik, N., Manowitz, B., Otto, R., & Leonard, C. (2004). Cerebral asymmetries for language: Evidence for structural–behavioral correlations. Neuropsychology, 18, 219–231. Chiarello, C., Liu, S., & Shears, C. (2001). Does global context modulate cerebral asymmetries? A review and new evidence on word imageability effects. Brain and Language, 79, 360–378. Chiarello, C., Shears, C., Liu, S., & Kacinik, N. (2005). Influence of word class proportion on cerebral asymmetries for high- and low-imagery words. Brain and Cognition Tennet XV, 57, 35–38. Clements, A. M., Rimrodt, S. L., Abel, J. R., Blankner, J. G., Mostofsky, S. H., Pekar, J. J., et al. (2006). Sex differences in cerebral laterality of language and visuospatial processing. Brain and Language, 98, 150–158. Cohen, L., Martinaud, O., Lemer, C., Lehericy, S., Samson, Y., Obadia, M., et al. (2003). Visual word recognition in the left and right hemispheres: Anatomical and functional correlates of peripheral alexias. Cerebral Cortex, 13(12), 1313–1333. Coney, J. (2002). Lateral asymmetry in phonological processing: Relating behavioral measures to neuroimaged structures. Brain and Language, 80, 355–365. Cousin, E., Peyrin, C., & Baciu, M. (2006). Hemispheric predominance assessment of phonology and semantics: A divided visual field experiment. Brain and Cognition, 61, 298–304. Crossman, D. L., & Polich, J. (1988). Hemispheric differences for orthographic and phonological processing. Brain and Language, 35, 301–312. D’Hondt, M., & Leybaert, J. (2003). Lateralization effects during semantic and rhyme judgement tasks in deaf and hearing subjects. Brain and Language, 87, 227–240. Eviatar, Z., Hellige, J. B., & Zaidel, E. (1997). Individual differences in lateralization: Effects of gender and handedness. Neuropsychology, 11, 562–576. Eviatar, Z., Ibrahim, R., & Ganayim, D. (2004). Orthography and the hemispheres: Visual and linguistic aspects of letter processing. Neuropsychology, 18(1), 174–184. Frost, J., Binder, J., Springer, J., Hammeke, T., Bellgowan, P., Rao, S., et al. (1999). Language processing is strongly left lateralized in both sexes. Brain, 122, 199–208. Frost, R., & Katz, L. (1989). Orthographic depth and the interaction of visual and auditory processing in word recognition. Memory and Cognition, 17(3), 302–310. Halderman, L. K., & Chiarello, C. (2005). Cerebral asymmetries in early orthographic and phonological reading processes: Evidence from backward masking. Brain and Language, 95(2), 342–352. Hatta, T. (1977). Recognition of Japanese kanji in the left and right visual fields. Neuropsychologia, 15, 685–688. Hunter, Z. R., & Brysbaert, M. (2008). Visual half-field experiments are a good measure of cerebral language dominance if used properly: Evidence from fMRI. Neuropsychologia, 46, 316–325. Kahlaoui, K., & Joanette, Y. (2008). Semantic processing and right hemisphere. Medical Science, 24, 72–76. Kareken, D. A., Lowe, M., Chen, S. H., Lurito, J., & Mathews, V. (2000). Word rhyming as a probe of hemispheric language dominance with functional magnetic resonance imaging. Neuropsychiatry Neuropsychology and Behavioral Neurology, 13, 264–270. Khateb, A., Pegna, A. J., Michel, C. M., Custodi, M. C., Landis, T., & Annoni, J. M. (2000). Semantic category and rhyming processing in the left and right cerebral hemisphere. Laterality, 5, 35–53.

471

Lavidor, M., Johnston, R., & Snowling, M. J. (2006). When phonology fails: Orthographic neighbourhood effects in dyslexia. Brain and Language, 96, 318–329. Lukatela, G., Carello, C., Savic, M., & Turvey, M. T. (1986). Hemispheric asymmetries in phonological processing. Neuropsychologia, 24, 341–350. Lurito, J. T., Kareken, D. A., Lowe, M. J., Chen, S. H., & Mathews, V. P. (2000). Comparison of rhyming and word generation with FMRI. Human Brain Mapping, 10, 99–106. Majeres, R. L. (1999). Sex differences in phonological processes: Speeded matching and word reading. Memory and Cognition, 27, 246–253. Nakamura, K., Oga, T., Okada, T., Sadato, N., Takayama, Y., Wydell, T., et al. (2005). Hemispheric asymmetry emerges at distinct parts of the occipitotemporal cortex for objects, logograms and phonograms: A functional MRI study. Neuroimage, 28, 521–528. Oldfield, R. C. (1971). The assessment and analysis of handedness: The Edinburgh inventory. Neuropsychologia, 9, 97–113. Piazza, D. M. (1980). The influence of sex and handedness in the hemispheric specialization of verbal and nonverbal tasks. Neuropsychologia, 18, 163–176. Proverbio, A. M., Vecchi, L., & Zani, A. (2004). From orthography to phonetics: ERP measures of grapheme-to-phoneme conversion mechanisms in reading. Journal of Cognitive Neuroscience, 16(2), 301–317. Pugh, K. R., Shaywitz, B. A., Shaywitz, S. E., Constable, R. T., Skudlarski, P., Fulbright, R. K., et al. (1996). Cerebral organization of component processes in reading. Brain, 119, 1221–1238. Sasanuma, S., Itoh, M., Mori, K., & Kobayashi, Y. (1977). Tachistoscopic recognition of kana and kanji words. Neuropsychologia, 15, 547–553. Schmidt, G. L., DeBuse, C. J., & Seger, C. A. (2007). Right hemisphere metaphor processing? Characterizing the lateralization of semantic processes. Brain and Language, 100, 127–141. Shaywitz, B. A., Shaywitz, S. E., Pugh, K. R., Constable, R. T., Skudlarski, P., Fulbright, R. K., et al. (1995). Sex differences in the functional organization of the brain for language. Nature, 373, 607–609. Simon, G., Bernard, C., Lalonde, R., & Rebai, M. (2006). Orthographic transparency and grapheme-phoneme conversion: An ERP study in Arabic and French readers. Brain Research, 1104, 141–152. Smolka, E., & Eviatar, Z. (2006). Phonological and orthographic visual word recognition in the two cerebral hemispheres: Evidence from Hebrew. Cognitive Neuropsychology, 23, 972–989. Springer, S. P., & Deutsch, G. (2000). Cerveau gauche, cerveau droit. A la lumière des neurosciences. Bruxelles: De Boeck Université. Tremblay, T., Ansado, J., Walter, N., & Joanette, Y. (2007). Phonological and semantic processing of words: Laterality changes according to gender in right- and lefthanders. Laterality, 12, 332–346. Tremblay, T., Monetta, L., & Joanette, Y. (2004). Phonological processing of words in right- and left-handers. Brain and Cognition, 55, 427–432. Tremblay, T., Monetta, L., & Joanette, Y. (2007). Hemispheric dynamics during easy and complex phonological processing: An ERP study. Brain and Language, 103, 8–249. Voyer, D. (1996). On the magnitude of laterality effects and sex differences in functional lateralities. Laterality, 1, 51–83. Waldie, K. E., & Mosley, J. L. (2000). Hemispheric specialization for reading. Brain and Language, 75, 108–122. Walter, N., Cliche, S., Joubert, S., Beauregard, M., & Joanette, Y. (2001). Graphemephoneme processing of single words. Brain and Cognition, 46, 295–299. Yvon, F., Boula de Mareüil, P., d0 Alessandro, C., Aubergé, V., Bagein, M., Bailly, G., et al. (1998). Objective evaluation of grapheme-to-phoneme conversion for text-to-speech synthesis in French. Computer Speech and Language, 12(4), 393–410. Zaidel, E., & Peters, A. M. (1981). Phonological encoding and ideographic reading by the disconnected right hemisphere: Two case studies. Brain and Language, 14, 205–234.