https://orcid.org/0000-0002-4969-6569
ABSTRACT
Working memory (WM) distinctions between right- and left-handers show inconsistent results in visual and verbal modalities. However, the tactual modality has not yet been assessed. The aim of the current study was to compare right- and left-handers across the three modalities. Thirty-two right handers and thirty left handers participants performed the Digit Span, Visuospatial Span, and Tactual Span tasks, encompassing forward and backward recalls. Results showed no difference between left- and right-handers on overall WM ability, and Tactual Span scores were lower compared to the other tasks. Left-handers performed better on the forward Visuospatial Span but worse than right-handers on the backward Tactual Span. Findings indicate that hemispheric lateralization patterns associated with handedness yield modality-specific advantages, highlighting the importance of considering both modality and handedness in interpreting WM performance, with implications for refining neuropsychological assessment.
INTRODUCTION
Roughly 10% of the human population is left-handed, and their cognitive characteristics have been extensively studied across domains such as language, learning, memory, and executive functioning (Chaudhary et al., 2009; Somers et al., 2015). However, findings regarding cognitive differences between left- and right-handers are mixed and generally fall into two patterns. The first, reflects a dichotomous approach, where studies report either the presence or absence of overall cognitive differences between groups. For example, Beratis et al. (2013) highlighted this binary trend, and a large meta-analysis concluded that there are no substantial differences in general cognitive abilities between left- and right-handers (Papadatou-Pastou, 2018).
By contrast, other research has suggested a more differentiated pattern, indicating that performance may vary by specific abilities. These studies propose that left-handers may outperform right-handers in certain domains, such as attention or auditory learning styles, while right-handers may excel in visual learning (Chaudhary et al., 2009). Somers et al. (2015) found a small but significant spatial advantage for right-handers, with no differences in verbal ability. Similarly, Li et al. (2024) reported that right-handers tend to perform better on general cognitive tests, showing stronger left-hemisphere dominance for language, whereas left-handers display advantages in creativity, divergent thinking, and interhemispheric cooperation. This variability underscores the need for further research focusing on specific domains and using well-validated, modality-sensitive measures (Li et al., 2024; Papadatou-Pastou, 2018).
One example of a specific cognitive ability is working memory (WM), defined as the temporary storage and manipulation of information in different modalities. Studies comparing WM between left- and right-handers have mostly examined verbal and visuospatial modalities, again yielding inconsistent findings. Some studies have reported better verbal WM among left-handers (Beratis et al., 2013), others found superior visuospatial WM among either right-handers (Hatta, 2018) or left-handers tasks (Piper et al., 2011), whereas some, found no differences (Axmacher et al., 2009; Corballis et al., 2008; Smilee et al., 2021; Tan et al., 2024). Imaging studies indicate that right-handers tend to rely more on right-hemisphere connectivity for visual WM, while left-handers show more bilateral engagement (Shirzadi et al., 2020), highlighting that neural variability does not always translate into performance differences (Axmacher et al., 2009).
Tactual WM, which refers to the storage and manipulation of tactual information (Levi & Heled, 2024), has received little attention in this context. Neuroimaging research has revealed it engages both sensory-specific and multisensory regions, with the left hemisphere more involved in encoding structural features and the right hemisphere supporting spatial processing (Gallace & Spence, 2009; Stoycheva & Tiippana, 2018). Furthermore, a recent study comparing left- and right-hand performance among right-handed individuals, found no laterality effect (Heled, 2025).
To date, no study has assessed tactual WM alongside verbal and visuospatial WM, in right- and left-handers. Therefore, the current study aims to test the effect of hand dominance on WM across these modalities, evaluating storage and manipulation separately. The findings may provide a more nuanced understanding of how handedness interacts with WM processes while also informing neuropsychological assessment and intervention strategies.
METHODS
Participants
A total of 62 healthy young adults participated in the study (29 women), who were divided into left- and right-dominant hand groups. Thirty-two right-handers (18 women), with a mean age of 24.81 (SD = 2.82) and an average of 14.09 (SD = 1.3) years of education, were matched with thirty left-handers (15 women) with a mean age of 25.97 (SD = 3.01) and an average of 13.63 (SD = 1.65) years of education. Hand dominance was determined by indicating the writing hand, while ensuring that the opposite hand could not also be used for this purpose. Participants were recruited from university classes, social media, and word of mouth. Inclusion criteria were ages 20–30 years and 12–16 years of education. Exclusion criteria included developmental or acquired neurological disorders, hearing or visual impairments that would interfere with computer use, and a psychiatric diagnosis. The study was approved by the ethics committee of Ariel University (AU-EH-20181220). A priori power analysis using G*Power 3.1 indicated that a minimum total sample of 34 would be sufficient to detect a medium-sized interaction effect (f = 0.25) in a mixed-design ANOVA, with α = 0.05, power = 0.80, and an assumed correlation of 0.40 among the variables.
Instruments
Tactual Span (Heled et al., 2021): Participants placed four fingers of each hand on keyboard keys and put on a blindfold. In forward recall, the examiner touched their fingers in a set order, and participants pressed the corresponding keys. Sequences started with two taps and increased by one, if at least one trial was correct. After failing all three trials of the same sequence length, participants proceeded to the backward recall, in which the procedure was the same as before, except that they had to press the keys in reverse order.
Digit Span (Wechsler, 1997): Numbers (1–9) were presented orally, each for one second. In the forward recall, participants repeated the numbers in the same order, with sequence length increasing after each correct response until two errors occurred at the same length. They then proceeded to the backward recall, where they repeated the sequences in reverse order.
Visuospatial Span (Corsi, 1972): Nine squares appeared in a random array on the screen and changed color sequentially. In the forward recall, participants clicked the squares in the same order, with sequence length increasing after each correct response until two errors occurred at the same length. They then proceeded to the backward recall, where the sequences were repeated in reverse order.
Procedure
Eligible participants provided informed consent, completed the demographic questionnaire, and performed the tasks in counterbalanced order.
Data Analysis
First, a chi-square test was used to examine sex distribution differences between the left- and right-handed groups. Group comparisons were then assessed with a 3 (modality: tactual, visuospatial, verbal) × 2 (handedness: left, right) repeated measures ANOVA, conducted separately for the forward and backward recalls. Bonferroni-adjusted post hoc tests followed, and in cases of significant interactions, independent samples t-tests were performed. The alpha level was set at p < .05.
RESULTS
Given potential gender differences in information processing, we compared gender distribution across handedness groups, which revealed no significant difference in the number of men and women between left- and right-handers, χ2(1) = 0.24, p = .622. Analysis of the forward recall showed a significant main effect for modality, F(2, 120) = 20.26, p < .001, η2p = 0.252. Pairwise comparisons indicated for the forward recall that the Tactual Span (M = 5.15, SD = 1.18) was significantly lower than the Digit (M = 6.23, SD = 1.29) and Visuospatial Spans (M = 6.23, SD = 1.06; p < .001). No difference was observed between the Digit and Visuospatial Spans (p = 1.00). Furthermore, no main effect for handedness was found, F(1, 60) = 0.015, p = .904, η2p = 0.001. However, there was a significant modality by handedness interaction, F(2, 120) = 3.16, p = .046, η2p = 0.05. Simple effects analyses revealed that left-handed participants outperformed right-handed participants on the Visuospatial Span task (p = .026), but no significant group differences were found for the Digit or Tactual Spans (Fig. 1).
Span tasks’ longest sequence scores (standard deviation in error bars) in the forward recall of the right- and left-handed groups.
Next, assessing the backward recall revealed a significant main effect for modality, F(2, 120) = 4.21, p = .017, η2p = 0.06, while Bonferroni-adjusted comparisons showed that the Tactual Span (M = 4.71, SD = 0.98) was significantly lower than the Visuospatial Span (M = 5.24, SD = 1.22; p = .007), whereas no other pairwise differences were significant. Additionally, the main effect of handedness was again non-significant, F(1, 60) = 0.73, p = .397, η2p = 0.01, but there was a significant modality by handedness interaction effect, F(2, 120) = 6.13, p = .003, η2p = 0.09 (Fig. 2). Follow-up tests showed no significant differences between the right- and left-handed participants in the Digit and Visuospatial Spans (p = .112, p = .069, respectively), while right-handed participants outperformed left-handed participants on the Tactual Span (p = .015).
Span tasks’ longest sequence scores (standard deviation in error bars) in the backward recall of the right- and left-handed groups.
DISCUSSION
Our findings showed that the Tactual Span was consistently lower than the Digit and Visuospatial Spans, in line with previous studies (Heled et al., 2021; Heled & Levi, 2024). This pattern may be explained by limited daily experience with tactual processing and difficulties in localizing touch, both of which increase cognitive load and reduce efficiency compared to verbal and visuospatial WM (Heled, 2025). Importantly, this pattern establishes a baseline difference across modalities, underscoring that tactual WM operates under distinct constraints. Clinically, this means that lower tactual WM scores should not be interpreted as global WM deficits, but rather assessed against modality-specific expectations.
The lack of WM differences between left- and right-handers corroborates other studies (Axmacher et al., 2009; Corballis et al., 2008; Smilee et al., 2021; Tan et al., 2024), suggesting that neural organization differences do not inherently impair task efficacy, thus highlighting the brain’s capacity for functional compensation (Axmacher et al., 2009). Corballis et al. (2008) further stated that handedness alone does not strongly influence WM abilities, but that the relationship between handedness and cognitive performance is nuanced. A view supported by our findings of significant modality-by-handedness interactions for both storage and manipulation.
Left-handers’ superior performance on the forward Visuospatial Span, with a marginal trend in the backward recall (p = .069), may reflect more bilateral or flexible hemispheric engagement, supporting spatial WM through enhanced interhemispheric cooperation (Powell et al., 2012; Smilee et al., 2021). Although some studies have reported a spatial advantage for right-handers (e.g., Somers et al., 2015), our findings suggest that left-handers may benefit in tasks emphasizing short-term sequential storage and manipulation, rather than spatial navigation or mental rotation. Additionally, variability in lateralization patterns among left-handers may confer advantages in WM tasks that rely on bilateral processing (Gerrits et al., 2020), which may not extend to all domains of spatial cognition.
Conversely, right-handed participants showed superior performance on the backward Tactual Span, possibly reflecting more efficient contralateral sensorimotor-cognitive coupling, whereby tactual input and executive processing tend to be streamlined within the dominant left hemisphere (Kaas et al., 2013). By contrast, left-handers’ bilateral and less predictable lateralization may result in less efficient neural routing, when tasks require precise sequential manipulation of somatosensory information, similar to visuospatial information (Gerrits et al., 2020). However, the lack of imaging data to support these explanations indicates that these conclusions should be taken with caution.
Our findings that modality- and handedness-specific differences influence cognitive performance underscore the importance for clinicians to consider these factors during neuropsychological evaluation. This is especially pertinent in neurological conditions with atypical lateralization patterns, such as temporal lobe epilepsy, where lesion side and handedness are associated with a higher degree of cognitive deterioration (Kalinin et al., 2014). Similar considerations apply to post-stroke aphasia, where lateralized damage in the dominant hemisphere not only affects verbal WM, but also produces inconsistent levels of impairment across modalities, with additional disruption of dominant-hand functions (Heled et al., 2025). In patients with focal brain tumors, cognitive outcomes also vary by modality depending on lesion laterality, while handedness may further modulate these effects (Połczyńska, 2021). Therefore, designing assessment batteries that incorporate multiple modalities and explicitly account for handedness, could enhance cognitive profiling, aid diagnosis, identify preserved processing routes, and guide assessment strategies.
Limitations of this study include the relatively small sample size, which, although sufficient to achieve power estimates of at least 0.8 as indicated by the a priori sample size estimation, may limit the validity and reliability of the research findings (Lakens, 2022). Additionally, the use of only one task per modality hinders the robustness of the assessments, and may not comprehensively capture the different underlying components of WM (Miyake et al., 2000). Finally, handedness was assessed via a self-report dichotomous question, rather than a standardized laterality measure, which may affect classification validity. Future research should employ larger samples and incorporate imaging methods to investigate the neurological underpinnings of the Tactual Span in left- and right-handers.
In conclusion, the results show that WM performance depends on task modality and its interaction with handedness. Across both stages, participants performed worse on the Tactual Span, reflecting the higher cognitive demands of tactual processing. No overall WM differences emerged between left- and right-handers, suggesting that handedness alone does not determine general performance. However, significant interactions indicated modality-specific advantages, showing that left-handers outperformed right-handers on forward visuospatial recall, likely due to more bilateral hemispheric engagement. In contrast, right-handers excelled in backward tactual tasks, possibly reflecting more efficient left-hemisphere processing of sequential somatosensory input. These findings highlight the importance of considering both modality and lateralization when assessing WM.
CONFLICT OF INTEREST
None declared.
