Attention and Interhemispheric Communication: Implications for Language Dominance

—Dominance of the left hemisphere for language processing is a prominent feature of brain organisation. Whereas structural models clarify the functional asymmetry due to direct access to local language circuits, dynamic models propose functional states of intrahemispheric activation and interhemispheric inhibition that are coupled with attentional processes. Real word settings often require modulations of lateralised neural processing and further express individual heterogeneity. In this research, we tested left- and right-handers, and used a behavioural paradigm with presentation of lateralised cue-target pairs to the same or opposite visual ﬁeld. We observed that handedness distinctly aﬀected word processing in the left hemisphere following contralateral cueing. Moreover, left-hemispheric dominance strengthened for right-handers vs abolished for left-handers, inﬂuencing behavioural eﬃciency. In combination with eye dominance recordings, these data suggest that attentional biases guided the processing strategies of both groups and in turn their achievements. Therefore, hand and eye dominance are both essential factors with a functional role in directing the communication of visual information between both hemispheres. Overall, the ﬁndings underline the importance of interacting hand-eye control systems in contributing to interhemispheric patterns in the context of language processing. (cid:1) 2022 The Author(s).


INTRODUCTION
The cerebral hemispheres have unique functional properties that shape the lateralisation patterns of many cognitive systems. Functional lateralisation relies, however, on both intrahemispheric and interhemispheric connections (Tzourio-Mazoyer, 2016). Whereas intrahemispheric circuits facilitate local activation within one hemisphere, interhemispheric circuits underlie communication between hemispheres that can involve cooperation or competition of information exchanges, including excitatory and inhibitory functions (Bloom and Hynd, 2005;van der Knaap and van der Ham, 2011). In particular, the excitatory model proposes that the corpus callosum assists cross-hemispheric sharing of information (Ringo et al., 1994), which decreases hemispheric lateralisation. Conversely, the inhibitory model suggests that the corpus callosum offers pathway by which one hemisphere can inhibit the contralateral side (Kinsbourne, 1974;Cook, 1986), which strengthens independent processing and increases hemispheric lateralisation. A central feature of the corpus callosum is its connection of homotopic areas between hemispheres (Chao et al., 2009). Evolutionary and developmental influences likely supported an increased intrahemispheric and reduced interhemispheric connectivity, favouring patterns of asymmetry and thus lateralisation (Nowicka and Tacikowski, 2011).
As a key cognitive function, language generally relies more heavily on the left than right hemisphere. It is argued that the corpus callosum plays an essential role in establishing the functional asymmetry of language during development, especially by suppressing cortical activity in the competing hemisphere (Jeeves and Temple, 1987). This is further supported by data from individuals with agenesis of the corpus callosum who show an increased distribution of language processing across hemispheres in addition to right-hemispheric dominance (Hinkley et al., 2016). The left-sided superiority such as for the ability to recognise written words or to extract meaning has been demonstrated in neuroimaging work (Price, 2000;Josse and Tzourio-Mazoyer, 2004), and behavioural research that has revealed performance advantages when words are presented to the right as compared to left visual field (Hunter and Brysbaert, 2008;Serrien and O'Regan, 2022). Two influential models are proposed in the literature to explain these lateralised effects. Whereas structural models clarify that hemispheric asymmetries arise due to differences of dealing with language content, dynamic models propose that https://doi.org/10.1016/j.neuroscience.2022.12.006 0306-4522/Ó 2022 The Author(s). Published by Elsevier Ltd on behalf of IBRO. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). each hemisphere directs attention to its contralateral visual field and simultaneously inhibits the other hemisphere, guided by intra-and interhemispheric mechanisms (i.e., opponent-processor theory; Kinsbourne, 1974). Thus dynamic models consider that both hemispheres work together in a complementary manner (Querne´et al., 2000). In case of a lateralised visual input, an activation imbalance is triggered in favour of the stimulated hemisphere. That is, presenting a stimulus to the right visual field produces an attentional shift to the opposite left hemisphere, regulated through reciprocal inhibition. Each hemisphere allocates this contralateral attentional bias, although its strength differs with a stronger bias from the left hemisphere than from the right hemisphere (Kinsbourne et al., 1977;Reuter-Lorenz et al., 1990).
Neuroanatomical models preserve the concept of interhemispheric inhibition, and refer to two attentional systems that rely on fronto-parietal circuitry (Corbetta and Shulman, 2011). First, the bilateral-organised dorsal network regulates shifts of spatial attention in a topdown manner, albeit according to asymmetries that are distinct for parietal and frontal regions. In particular, parietal nodes are characterised by a contralateral bias of attention and interact through interhemispheric inhibition as opposed to frontal nodes that are driven by righthemispheric dominance (Duecker and Sack, 2015). Second, the ventral network that is strongly right-hemispheric dominant is involved in detecting behaviourally relevant stimuli and can bias the dorsal network (Corbetta and Shulman, 2011). When active, both systems display distinct recruitment and switch patterns for flexible use of attention (Bartolomeo and Seidel Malkinson, 2019;Mengotti et al., 2020).
A dynamic model of functional lateralisation further addresses individual variation due to differences and responsiveness of information processing mechanisms (Hutchinson et al., 2003). One such identified factor is handedness, which represents an expression of hemispheric asymmetry for hand movement control. At the population-level, there is a strong 90:10% prevalence of right vs left handedness in humans (Coren and Porac, 1977). Both groups show distinct functionality across hemispheres for language processing (Tzourio et al., 1998;Tussis et al., 2016) with the handednesslanguage association specifying that more right-than left-handers have typical left-hemispheric lateralisation (Pujol et al., 1999;Knecht et al., 2000;Flo¨el et al., 2005). In particular, about 95% of right-handers and 75% of left-handers express left-sided dominance for language with the remaining minority demonstrating bilateral or right-hemispheric profiles (Mazoyer et al., 2014Tzourio-Mazoyer et al., 2016Vingerhoets, 2019). With respect to attention, evidence illustrates that the ventral system is predominantly right-hemispheric lateralised in right-as opposed to left-handers who show bilateral or left-hemispheric lateralisation (Bareham et al., 2015) whereas the dorsal network is more pronounced in the right hemisphere for left-handers (Petit et al., 2015). Together, these data express that handedness associates with modifications of language and attention pro-cessing in the brain, suggesting differences for the integration of distributed processing and information exchange between hemispheres.
The aim of this research is to investigate modulations of the activation-inhibition states during language processing and its relation with handedness. Here, we use an experimental behavioural approach of lateralised cue-target pairs that are guided by attentional shifts and implemented within a divided visual field paradigm. In particular, we examine how word processing is affected by non-instructive cues presented in the same vs opposite visual field, i.e., when cues and targets are processed by a single hemisphere vs both hemispheres. Here, we use brief cue-target intervals that support automatic spread of activation (Chiarello, 1985;Koivisto, 1998;Korsnes and Magnussen, 2007). The hypothesis is that processing of cue-target pairs will be distinct in the same hemisphere (intrahemispheric) vs opposite hemispheres (interhemispheric) due to attention shifts and altered interactions when both sides are involved. In addition, differences between both hemispheres are expected to occur as right-hemispheric regions tend to be more bilaterally interconnected whereas lefthemispheric regions interact more locally (Gotts et al., 2013). We further study left-and right-handers as well as their eye dominance that provides a foundation for attentional biases (Schintu et al., 2020), and investigate differences between groups and individuals. We hypothesise that cue-target processing will be differently regulated by left-than right-handed individuals due to differentiation of attentional influences that interact with interhemispheric mechanisms. Overall, we argue that studying the dynamic involvement of interhemispheric interactions will provide increased insights into the regulation of hemispheric lateralisation.

EXPERIMENTAL PROCEDURES Participants
A total of 41 participants was included in the study. Their mean age (M AGE ) and standard error (SE) of the mean was M AGE = 21.2 ± 0.5 years. The participants had normal or corrected-to-normal vision and no history of a neurological or psychological condition as assessed by a health history questionnaire. We recruited individuals who self-identified as right-handed and non-righthanded. Participants were all informed about the procedures of the study and provided written informed consent prior to the start of the experiment. The research was approved by the School of Psychology Ethics Committee (reference: 604R) and was conducted in accordance with the Declaration of Helsinki.

Handedness
Handedness consists of different dimensions and is typically studied by means of preference and performance assessments. Whereas preference is generally captured by self-reports, performance is measured through manual proficiency tests (Corey et al., 2001). Here, we include an assessment of both dimensions.
Handedness questionnaire. Participants completed a handedness questionnaire that consisted of 15 items (i.e., write, hold toothbrush, use scissors, throw ball, hold racquet, use spoon to stir, open lid from drinks can, use computer mouse, peel apple, use comb, hold knife to cut, use needle, deal cards, use eraser, broom useupper hand). The handedness questionnaire used a 5point Likert scale that varied between always left (0), usually left (1), equal (2), usually right (3) and always right (4). Accordingly, the scores of the items were added for each participant, and divided by the maximum score of the questionnaire, and multiplied by 100. This gave a handedness score that ranged from 0 (extreme left-handedness) to 100 (extreme right-handedness), (Serrien and O'Regan, 2019). The handedness scores were used to categorise the participants as 20 lefthanders (M AGE = 22.4 ± 0.8 years; M HAND = 22.2 ± 3. 3, 16 females) and 21 right-handers (M AGE = 20.1 ± 0. 4 years; M HAND = 92.4 ± 6.5, 15 females). The writing hand was included as an additional condition as most individuals categorise their handedness on the basis of their writing hand (Perelle and Ehrman, 2005).
Manual proficiency. Participants' performance was measured using a computerised finger tapping test. This required participants to tap on the spacebar of a computer keypad with the left or right index finger as quickly as possible for 10 s. There were three trials for each hand. The participants scores were calculated by obtaining the average score from the trials performed with each hand. We also calculated the difference between the right and left hand performances (DIF TAP ). Positive and negative scores associated with right and left hand benefits for tapping, respectively.

Eye dominance
Eye sighting dominance represents the behavioural preference for one eye over the other under monocular viewing conditions (analogous to hand dominance), and has been associated with activation patterns of attentional systems (Roth et al., 2002). It has a population bias like handedness with the majority of people being right eye dominant, i.e., 65% (Porac et al., 1976) albeit less pronounced (Bourassa et al., 1996).
Eye dominance assessment. Participants were asked their preference to look through a key hole. The scores involved a 5-point Likert scale that varied between always left (0), usually left (1), equal (2), usually right (3) and always right (4), divided by the maximum score of the questionnaire, and then multiplied by 100. This provided an eye dominance score that ranged from 0 (extreme left-eyed) to 100 (extreme right-eyed). The scores were calculated separately for left-handers (M EYE = 31.3 ± 5.9) and right-handers (M EYE = 73.8 ± 5.6). In our sample, 70% of left-handers showed left eye dominance whereas 76% of right-handers had right eye dominance. There were four participants with no eye dominance (two left-and two right-handers).

Experimental task
Divided visual field paradigm. We used a divided visual field paradigm that involves presenting information to the visual fields on the basis of the neuroanatomical organisation of the visual system. Thus, stimuli projected to the right visual field are processed initially by visual cortical regions in the left hemisphere and vice versa. Subsequent to this projection, communication between hemispheres permits transfer to the other side through direct connections, mainly via the corpus callosum. Although providing behavioural measures, the technique can be used to obtain reliable predictors of lateralisation patterns (Hunter and Brysbaert, 2008;Gerrits et al., 2020).
Experimental paradigm. The experimental paradigm includes cue-target pairs presented in close succession to the same (intra) or opposite (inter) visual field for investigating the contribution of attentional mechanisms to word recognition. We use non-instructive cues that are shown briefly and the cue-target presentations occur with a short stimulus onset asynchrony followed with backward masking. These experimental features support cueing due to automatic spread of activation and ensure that use of conscious strategies is limited (e.g., Chiarello, 1985;Koivisto, 1998;Korsnes and Magnussen, 2007), although the cue still captures attentional resources (Eimer, 1997;Kiefer and Brendel, 2006). It is argued that such biases before target onset can interact with interhemispheric inhibition. The stimuli are action words associated with use of the hands (i.e., draw) and abstract words referring to internal states (i.e., hope), (O'Regan and Serrien, 2018) based on the idea that their meaning is acquired through interactions with the action or process that is captured by the word (Hauk et al., 2004). Finally, the cue-target pairs comprise congruent and incongruent associations (combinations of the same or different word categories) with participants being asked to respond to the target only, as indicated by a central arrow at fixation. The participants received no instructions with respect to the stimuli combinations.
Procedure. Participants were seated at a viewing distance of 70 cm from a computer monitor, with their head rested on a chinrest. A trial sequence is illustrated in Fig. 1 and was presented using PsychoPy (Peirce and MacAskill, 2018). All stimuli subtended 1.1°of visual angle in height and were presented in white Arial font on a black background. The trial sequence started with a central fixation cross for 500 ms followed by the presentation of a cue word for 150 ms alongside a filler stimulus ($$$$) in the opposite visual field. Cue and filler stimuli were matched for length, presented at ±2°visual angle of central fixation, and followed by a backward mask for 30 ms. Thereafter, there was a blank screen for 50 ms, followed by the presentation of a target word for 150 ms to the same (intrahemispheric) or opposite (interhemispheric) visual field.
The fixation cross was replaced by an arrowhead pointing to the left or right side in order to indicate the target word. There was an inter-stimulus interval of 1000 ms. Participants were instructed to specify the category of the target word i.e. whether it was an action or an abstract word, and to respond as accurately and as quickly as possible. Bimanual responses were used in order to avoid an effect of stimulus-response compatibility, i.e., keypress responses to stimuli presented in right and left visual fields are faster for hand ipsilateral than contralateral to stimuli (Berlucchi et al., 1977). Participants pressed two keys simultaneously using their index fingers for one category and their middle fingers for the other category. There were 40 observations per participant and per condition of the intra-and interhemispheric trials. There were four blocks of 80 trials, resulting in a total of 320 trials. An equal number of intrahemispheric and interhemispheric trials were presented in each block. Half of the trials were congruent (i.e. cue-target were action words or abstract words) and the other half were incongruent (i.e. action cue with abstract target or vice versa). Trials were randomised within blocks of trials. Participants completed 16 practice trials to familiarise themselves with the task demands. Participants were offered the opportunity to take a break in between each of the blocks of trials.
Target word performance was measured using reaction time (i.e., averaged bimanual responses) and response accuracy (i.e., percentage of correct responses). We further computed a laterality index for reaction time (LI RT ) according to [L À R]/[L + R] Â 100 and for response accuracy (LI ACC ) according to [R À L]/[ R + L] Â 100 where R and L stimuli presented in the right and left visual field, respectively. The LI quantifies the relative strength of lateralisation with positive scores of LI RT and LI ACC representing a right visual field advantage, whereas negative scores of LI RT and LI ACC indicate a left visual field advantage (Serrien and O'Regan, 2022). LI RT and LI ACC were calculated separately for intra-and interhemispheric trials. A LI cut-off score was computed to categorise individuals as left-, right-or non-lateralised. Based on the group sample across intra-and interhemispheric conditions, the cut-off score was calculated according to mean-SE if mean > 0, and mean + SE if mean < 0 and was set to 0.2. Thus, scores at +0.2 and 0.2 were used to divide right visual field (LI > +0.2) from left visual field (LI < -0.2) and non-lateralised (À0. 2 LI +0.2) dominance.

Statistical analyses
Effect sizes were calculated by means of partial eta squared g 2 p and Cohen's d as provided by SPSS statistics (version 27). A pvalue of <0.05 was used as statistically significant. Means alongside the SE are reported throughout the paper.
The reaction times of the language task were analysed using mixed 2 Â 2 Â 2 Â 2 Â 2 ANOVAs. There were within-subjects factors of Stimulus Presentation (intrahemispheric vs interhemispheric), Visual Field (left vs right), Target Stimulus (abstract vs action), Cuetarget Congruency (congruent vs incongruent) and a between-subject factor of Handedness Group (lefthander vs right-hander). The LI RT was analysed using mixed 2 Â 2 Â 2 Â 2 ANOVA, including within-subjects factors of Stimulus Presentation, Target Stimulus, Cue-Target Congruency and a between-subjects factor of Handedness Group. Mixed ANOVAs were followed by post-hoc analyses (two-tailed paired t-tests and independent t-tests) with Bonferroni correction for multiple comparisons where appropriate. In addition, correlations were calculated between the LI RT of the target words in the intra-and interhemispheric conditions alongside the handedness scores. At the individual-level, we assessed the number of individuals who demonstrated either a left-or right-hemispheric profile as well as those who lacked a clear pattern (nonlateralised). Chi-square tests were conducted to assess the LI RT frequency counts of the handedness groups.
Hemispheric functioning was further detailed by assessing the complementary organisation of the activation-inhibition states. Complementary organisation would indicate functional biases at the group-and/or individual-level. Here, we examined participants' hemispheric associations of intra-and interhemispheric conditions by correlating their LI RT scores and by assessing the number of individuals with a dominant profile. In particular, right visual field superiority for intraand interhemispheric conditions would reflect dominance of the left hemisphere whereas right visual field superiority for intrahemispheric condition alongside left visual field superiority for interhemispheric condition would reflect a shift of hemispheric dominance to the right hemisphere. Prevalence of the dominant pattern was evaluated against the null-hypothesis of five combinations (four lateralisation and one nonlateralisation pattern).
An unpaired t-test was used to assess eye dominance differences between both handedness groups. Correlation analyses were further conducted between eye and hand dominance as well as between eye dominance across handedness groups and the LI RT of the intra-and interhemispheric conditions.

RESULTS
We report on the reaction time data of the language task as the accuracy data which were above 87% did not show significant effects, p > 0.05. In this experimental context, we argue that the timing measurement is most sensitive to the small effects associated with visual half field processing. We study group-and individual-level data on the basis that as an evolutionarily stable strategy, two levels of lateralisation are observed. Whereas lateralisation at the individual level implies a pronounced bias in single individuals, lateralisation at the populationlevel indicates that the majority of individuals demonstrate the same bias (Vallortigara, 2006). The data from the manual proficiency (tapping) task can be found in the supplementary materials section.
LI RT and handedness. For the intrahemispheric condition, a correlation analysis showed no significant Fig. 2. Reaction times for the target words in both handedness groups as a function of the stimulus presentation (interhemispheric vs intrahemispheric condition), and visual field of presentation (right vs left visual field). Box-and-whiskers plots with the box representing the median with 25th and 75th percentiles whereas the whiskers represent the 5th and 95th percentiles. There was a right visual field preference across handedness groups in the intrahemispheric condition. However, in the interhemispheric condition, right-handers showed a right visual field advantage whereas left-handers tended towards a non-significant left visual field advantage. Significant effects (**p 0.01; *p < 0.05). association, p > 0.05, R 2 = 0.04 (Fig. 4, left panel). The scatter plot indicates that right-handed individuals demonstrated dominance of the right visual field (N = 10, 48%) and left visual field (N = 9, 43%) vs non-lateralisation (N = 2, 9%). A large number of lefthanded individuals showed right visual field (N = 14, 70%) as opposed to left visual field (N = 6, 30%) dominance. For the interhemispheric condition, a correlation analysis provided a positive association. r (39) = 0.52, p < 0.001, R 2 = 0.28 (Fig. 4, right panel), revealing that increased right-handedness associated with stronger dominance of the right visual field. The scatter plot illustrates that the majority of right-handed individuals demonstrated superiority of the right visual field (N = 15, 71%) whereas a smaller number obtained left visual field dominance (N = 6, 29%). Conversely, a large number of left-handed individuals had superiority of the left visual field (N = 12, 60%) as opposed to right visual field (N = 6, 30%) and non-lateralisation (N = 2, 10%).
These results at an individual-level organisation illustrate that more left-than right-handed individuals had right visual field dominance for the intrahemispheric condition, v 2 1 = 10.00, p < 0.002 whereas this was reversed for the interhemispheric condition with more right-than left-handed individuals demonstrating right visual field dominance, v 2 1 = 33.82, p < 0.0001. Also, the number of right-handed individuals with right visual field dominance was lower for the intra-than interhemispheric condition, v 2 1 = 4.07, p = 0.04, whereas this was reversed for the number of left-handed individuals with right visual field dominance being lower for the inter-than intrahemispheric condition, v 2 1 = 15.21, p = 0.0001.
LI RT intra-and interhemispheric. The correlation analysis between the LI RT scores did not show a significant association, p > 0.05, R 2 = 0.003 (Fig. 5). The scatter plot illustrates that various combinations were observed across participants, but many righthanders (N = 8, 38%) showed right visual field dominance for both intra-and interhemispheric conditions. Moreover, there was prevalence of this combination against the null-hypothesis with 20% according to five categories, z = 2.06, p < 0.04, CI: 18.03-61.47%. In contrast, many left-handers (N = 8, 40%) demonstrated right visual field dominance for intrahemispheric condition and left visual field dominance for interhemispheric condition. Moreover, there was prevalence of this combination against the null-hypothesis with 20% according to five categories, z = 2.24, p < 0.003, CI: 19.12-63.95%. No other combinations obtained significance, p > 0.05.
Eye and hand dominance. An unpaired t-test revealed a significant difference for eye dominance between both handedness groups, t(39) = -5.21, p < 0.0001, d = -1.63. A correlation analysis between eye and hand dominance was significant, r(39) = 0.68, p < 0.001, R 2 = 0.46, indicating a positive association between the preference of eye and hand. However, of note is that both handedness groups showed variation, i.e., 70% of left-handers had left eye dominance whereas 76% of right-handers demonstrated right eye dominance.
Eye dominance and LI RT intra-and interhemispheric. Correlation analyses between the eye dominance score across handedness groups and the LI RT of the intrahemispheric condition was not significant, p > 0.05, whereas there was a significant association with the LI RT of the interhemispheric condition, r(39) = 0.56, p < 0.001, R 2 = 0.31 (Fig. 6). Correlation analyses that included only the individuals with typical eye dominance for each handedness group revealed no significance with the LI RT of the intrahemispheric condition, p > 0.05, whereas there was a significant association with the LI RT of the interhemispheric condition, r(28) = 0.66, p < 0.0001, R 2 = 0.45.

DISCUSSION
Hemispheric lateralisation underlines that cognitive processes such as language are lateralised in the brain, and relies on specialisation of the hemispheres alongside interhemispheric communication pathways Fig. 4. Scatter plots of the laterality index with handedness scores for the intrahemispheric (left panel) and interhemispheric (right panel) conditions. Language responses with positive score represent superiority of the right visual field, left-handed <50 and right-handed >50. (Hellige, 1993;Gazzaniga, 2000;Bloom and Hynd, 2005). In this work, we adopt a dynamic model of functional lateralisation that underlines that hemispheric asymmetries are sensitive to attentional processes guided by an activation-inhibition synergy that acts as a regulatory mechanism. We address influences of task parameters and individual variation by means of contralateral vs ipsilateral cueing together with the study of individuals with different hand and eye dominance.

Language processing and activation-inhibition processes
Language processing strongly relies on the left hemisphere as evidenced from behavioural, imaging and clinical research (Springer et al., 1999;Knecht et al., 2000;Szaflarski et al., 2002;Hunter and Brysbaert, 2008;Skeide and Friederici, 2016). Under normal circumstances, word and meaning processing are optimised through facilitation of the left hemisphere alongside inhibition of the right hemisphere (Kinsbourne, 1974;Chiarello and Maxfield, 1996). However, left-hemispheric dominance is a dynamic feature and can vary as a result of factors that modulate the interhemispheric balance (Smith-Conway et al., 2012;Hartwigsen et al., 2013). Using resting state measurements, it has been shown that differences in language lateralisation associate with those in homotopic interhemispheric connectivity . We used an experimental design that included lateralised non-instructive cues with brief stimulus duration that further minimise eye movements to spatial locations (Landis and Regard, 1988;Hunter and Brysbaert, 2008). We observed no effect of cueing congruency on target processing, suggesting that changes of attentional asymmetries did not significantly influence task performance. There was further no impact of word type (abstract vs action), which underlines dominance of left-hemispheric processing and is in line with previous work (O'Regan and Serrien, 2018).
The reaction time data showed different effects on intra-and interhemispheric regulation. In the intrahemispheric condition, word targets presented in the right visual field were processed quicker than those presented in the left visual field, underlining the functional superiority of the left hemisphere. That word processing is less efficient in the non-dominant vs dominant hemisphere has been addressed in the literature by two main models. Whereas the callosal relay model suggests that information is transferred to the dominant side via the corpus callosum, the direct access model proposes that information is processed by the right hemisphere only (Iacoboni and Zaidel, 1996;Olk and Hartje, 2001). These models imply reduced processing in the non-dominant hemisphere which leads to inferior performance, and there is research that supports both viewpoints (Chu et al., 2018).
In the interhemispheric condition, the changes indicated that transfer between both hemispheres is a fundamental mechanism of information integration that steers neural processes, particularly for tasks that support sensorimotor and perceptual control (Koch et al., 2011;Chaumillon et al., 2018;Schintu et al., 2021). Yet, the reaction times of the target responses differed as a function of the visual field of presentation and further interacted with handedness. That is, whereas target processing in the right hemisphere was similar for both groups, this was not the case for the left hemisphere. In particular, targets presented to the right visual field were affected by cues initially presented to the opposite visual field, suggesting modulation of the interhemispheric balance by attentional resources, and in turn lefthemispheric functioning. Previous work has shown that interhemispheric transfer of visual information is fastest from the right to left side, and thus links with the specialisation of the right hemisphere for visuospatial attention (Marzi et al., 1991). However, the left-hemispheric reaction times in our study revealed characteristic effects as a function of handedness. In particular, right-handers showed faster reaction times whereas left-handers  tended to slow down their reaction times. This suggests that initial handling in the right hemisphere triggered release of interhemispheric inhibition, influencing efficiency of processing. The laterality index further detailed these handedness-related influences, with right-handers strengthening left-hemispheric dominance as opposed to left-handers who demonstrated elimination of its superiority. Therefore, the switch between right-and left-sided processing elicited distinct interactions between hemispheres and rebalancing of resources.
The association of handedness with the inter-but not intrahemispheric condition points to modulations of attentional resources. Moreover, both handedness groups have different neural circuits that underlie attentional control with the ventral network showing right-hemispheric lateralisation in right-handers vs bilateral or left-hemispheric lateralisation in left-handers (Bareham et al., 2015) whereas the bilateral-organised dorsal network is more pronounced in the right hemisphere in left-handers (Petit et al., 2015). A critical component is the dorsal system with its distinct parietal nodes (but not frontal nodes) for which attentional allocation occurs towards the contralateral side while both hemispheres maintain mutual inhibition, suggesting regional differences of hemispheric asymmetries as proposed by a hybrid model of attentional control (Duecker and Sack, 2015). In addition, the ventral system can bias the dorsal system, for example due to a significant stimulus that triggers a reorientation of attention (Corbetta and Shulman, 2011), illustrating that both circuits cooperate for flexible use of attention as a result of the task or contextual requirements (Vossel et al., 2014;Bartolomeo and Seidel Malkinson, 2019;Mengotti et al., 2020). The foundation of these attentional networks is facilitated by the superior longitudinal fasciculus (SLF), a major bidirectional association tract that connects various premotor and parietal areas, with branches that are closely linked with their functional role; dorsal (SLF I, partly SLF II) and ventral (SLF III, partly SLF II), (Koch et al., 2011;Thiebaut de Schotten et al., 2011;Budisavljevic et al., 2021). In particular, SLF I and SLF II are involved in top-down modulation of selective attention and online movement control (Budisavljevic et al., 2021) with the anatomical characteristics of SLF I prominently associating with handedness (Howells et al., 2018). Clinical work has further revealed that the resection of SLF II in surgical patients results in postoperative changes of selective attention and hand preference for goal-directed tasks, suggesting a relationship between attentional processing and handedness (Howells et al., 2020).
Besides handedness, eye dominance as an index of attentional systems has a relevant role in interhemispheric transfer of visual information at the level of posterior parietal circuitry. Moreover, righthanders (with dominant right eye) show faster right-toleft transfer whereas left-handers (with dominant left eye) lack a clear asymmetry (Chaumillon et al., 2018). These effects would thus lead to a strong right-to-left drive for right-handers vs an unbalanced drive for left-handers. Our results revealed that left-and right-handers had pronounced left and right eye dominance, respectively. That the majority of both handedness groups expressed typical eye dominance suggests that the release of interhemispheric inhibition likely boosted left-sided processing and facilitation of target identification for right-handers. In contrast, unstructured interhemispheric influences would weaken left-sided processing and delay target identification for left-handers.
Studies about the neural basis of eye dominance have shown that stimulation of each eye reaches both cerebral hemispheres. However, due to anatomo-functional association between the dominant eye and ipsilateral primary visual cortex (Erdogan et al., 2002;Rombouts et al., 1996;Seyal et al., 1981;Shima et al., 2010), visual information is processed more efficiently in the contralateral vs ipsilateral visual hemifield to dominant eye (Chaumillon et al., 2014(Chaumillon et al., , 2017Tagu et al., 2016). Furthermore, the asymmetric activation and priority of visual processing captured by the dominant eye may modulate visuospatial attention circuity (Shneor and Hochstein, 2008). This proposes that for participants with right eye dominance, the relationship with its ipsilateral hemisphere (right hemisphere; Shima et al., 2010) is consistent with a right-sided asymmetry of attentional networks. In contrast, in participants with left eye dominance, the relationship with its ipsilateral hemisphere (left hemisphere) neutralises the right-hemispheric dominance for visuospatial attention (Chaumillon et al., 2018;Schintu et al., 2020). Combined, the data illustrate that eye and hand dominance are central factors that associate with attentional control and interhemispheric communication, influencing functional lateralisation properties. Their interaction may be linked through an evolutionary origin that steers visually-guided activities to spatial settings through an exploration of the environment with hand and eyes (Petit et al., 2015).

Individual handedness-language and complementary intra-interhemispheric associations
Handedness is an essential trait of motor control, and represents the dominance to use one hand over the other for skilled manual activities. Throughout history, a preferential bias of right-vs left-handers has been noted in humans at the population level according to a 90-10% ratio (Coren and Porac, 1977). Both groups demonstrate variation of intrahemispheric sensorimotor representations as well as interhemispheric interactions that are particularly critical through inhibitory regulation, with right-handers demonstrating modulations that tend to support control of the dominant (left) hemisphere whereas both hemispheres show more equal capabilities in lefthanders (Reid and Serrien, 2012;Tazoe and Perez, 2013;Pool et al., 2014).
Left-hemispheric lateralisation for language is prominent in about 95% of right-handers and 75% of left-handers, indicating that atypical language superiority is more frequent in individuals characterised by nonright-handedness (Pujol et al., 1999;Knecht et al., 2000). Two additional language-related variants have been confirmed through neuroimaging work: no hemispheric dominance with bilateral representations, a pattern that is rather equally present in both handedness groups, in addition to right-hemispheric dominance that is usually only observed in a small subgroup of lefthanders Vingerhoets, 2019). Taking these differences into account, it is therefore relevant to consider the profiles that characterise their behaviour (Serrien and O'Regan, 2022). Our data revealed that the participants from both handedness groups had distinct sensitivity during intra-and interhemispheric conditions. In particular, left-hemispheric dominance of target words was strongest during the interhemispheric condition for right-handed individuals as opposed to the intrahemispheric condition for lefthanded individuals. This confirms heterogeneity as a function of the hemispheric processing demands.
The concept of complementary organisation has especially been addressed in the literature for assessing whether lateralisation of one cognitive function can predict the asymmetric processing of another one (Cai et al., 2013;Brederoo et al., 2020;Serrien and O'Regan, 2022). We used the premise of complementarity to examine more closely the activation-inhibition synergy. In particular, theoretical viewpoints propose that intrahemispheric activation and interhemispheric inhibition are two separate mechanisms (Kinsbourne, 1974) or are components of the same mechanism that support processing priorities within and between hemispheres (Querne´et al., 2000). In examining complementarity, our results revealed a range of combinations, indicating flexibility of information processing. However, we noted a characteristic pattern for both groups. Whereas many right-handed individuals demonstrated left-sided dominance across conditions, left-handed individuals revealed left-sided dominance for intrahemispheric condition versus right-sided dominance for interhemispheric condition. This suggests that right-handed individuals tended to process information more robustly within unilateral left circuitry whereas left-handed individuals are more guided by bilateral circuits. Thus, there is pronounced individual variation that associates with distinctive hemispheric mechanisms as a function of handedness. In conclusion, our data showed that handedness distinctly affected word processing in the left hemisphere following contralateral cueing. This result was further supported by a prominent role of eye dominance, suggesting that characteristic attentional biases guided the processing strategies of both groups. Therefore, hand and eye dominance are both key factors with a functional role in directing transfer of information between both hemispheres alongside an impact on processing resources. Overall, the findings underline the importance of interacting hand-eye control systems in contributing to interhemispheric patterns in the context of language processing.

FUNDING
This work was supported by a research grant from the BIAL foundation to DJS (no. 376/14).