Homonymous visual field defects (HVFDs) following acquired brain lesions impact the survivors’ functional recovery and independent living by hampering several activities of everyday living, such as reading, finding objects in the surrounding space, navigating the space (i.e., bumping into people or getting lost), and driving a car (1). These impairments can lead to anxiety and depressive symptoms, affecting patients’ quality of life (2) and that of their caregivers.

Rehabilitative treatments for VFDs primarily involve restitutive and compensatory approaches (3, 4). Restitutive trainings include a variety of techniques aimed at reactivating lesioned, hypoactive visual networks, in order to restore visual functions, such as the enlargement of the defective visual field. Visual restoration therapy [VRT; (5, 6)], one of the most studied restitutive approaches, is a computerized training whereby the border zone of the scotoma is repeatedly stimulated with luminous stimuli, with the purpose of shifting the border area and reducing the blind hemifield. Despite some methodological controversies, a study with a large sample demonstrated the efficacy of VRT in terms of an enlarged visual field, enhanced detection abilities, and subjective improvements in everyday life (6). Instead, compensatory approaches train the patient to compensate for visual loss by developing more efficient eye movements and improving the speed and the accuracy of saccades toward the defective side of the visual field (7). This is achieved with paper-and-pencil or computerized visual scanning training that may include visual search and reading tasks [e.g., (8–10)]. Compensatory trainings can enhance search times and reduce the impact of HVFDs in the activities of daily living (11), and the improvements can persist for several years (12), even in absence of visual field enlargement (10).

Among compensatory trainings, multisensory, audio-visual trainings (AVTs) have been successfully employed to improve visual field scanning in acquired HVFDs (13–16). A recent review of 16 studies (17) confirmed that AVTs can improve visual information processing, as indexed by reduced search times in visual exploration tasks and reduced disability in activities of daily living [e.g., (13)], as well as an increased number of larger saccades in the blind hemifield (18). The effects of AVT are thought to be mediated by the activity of retino-collicular-extrastriate pathways, with a pivotal role of the multisensory neurons of the superior colliculus (SC) that integrate and amplify cross-modal signals (19). Combined with saccadic tasks, a repetitive, spatially and temporally congruent presentation of audio-visual stimuli can induce plastic changes in visual functions mediated by collicular activity, as recently demonstrated by Stein and Roland (20) in an animal model.

Irrespective of the type of approach, effective visual trainings are usually hospital-based, long (i.e., two weeks or longer), and intensive (i.e., multiple hours a day). Even though some logistical problems can be overcome [e.g., home-based multisensory training in (16) and https://osf.io/jmq86/ for an ongoing clinical trial], the idea of boosting brain plasticity by means of non-invasive brain stimulation (NIBS) to make rehabilitation faster and/or more effective has attracted some attention. Accordingly, visual trainings have been combined with non-invasive transcranial electrical stimulation techniques (tES) such as transcranial Direct Current Stimulation [tDCS; (21, 22)] or repetitive transorbital alternating current stimulation [rtACS; (23, 24)].

tDCS, in particular, relies on the application of a continuous current on the scalp and has been shown to modulate the excitability of the stimulated area, mostly on the motor cortex (25, 26), as well as on visual areas (27, 28), through long-term potential/depression mechanisms [e.g., (29)], inducing long-range effects on functionally connected networks [e.g., (30)]. Not surprisingly, tDCS has also been investigated as a tool to promote brain plasticity and recovery in diverse neuropsychiatric conditions (31), including acquired HVFDs, although to a much lesser extent (32, 33).

With respect HVFDs, most of the studies have applied anodal tDCS over the occipital cortex (to enhance brain plasticity in an excitatory fashion) in synergy with restitutive approaches designed to reactivate hypofunctioning populations of the visual network [(21, 34, 35); Matteo et al. (36); however, see (24, 37) for cathodal tDCS on contralesional occipital cortex]. For example, Plow et al. (22) showed, in chronic patients, an expansion of visual field, increased detection accuracy, and improved vision-related ADLs following a 3-month VRT (“border training”) combined with occipital anodal tDCS (compared to sham tDCS). More recently, Alber et al. (21) applied occipital tDCS to seven sub-acute patients undergoing VRT and observed an improvement in visual field sensitivity, which was higher than that of a control group who received a compensatory exploration training without tDCS. However, despite these promising results, the recent review by Perin et al. (32) concluded with a Level C recommendation for clinical application of tDCS for HVFDs.

To our knowledge, no studied attempted to combine tDCS with compensatory trainings in HVFDs patients, including those based on multisensory integration. As previously mentioned, these kind of approaches rely on the training of oculomotor functions in a context of audio-visual stimulation, mediated by cortico-subcortical circuits also involved in eye movements and attentional orienting. As such, tDCS could be applied to brain areas and networks beyond the occipital cortex, such as the temporo-parietal cortex, an important hub for visuo-spatial functions and a heteromodal region for visual and auditory processing [e.g., (38)]. For instance, we have investigated the effects of parietal neuromodulation on visuo-spatial orienting and visual search in multisensory contexts such as those used to rehabilitate HVFDs (39, 40). Notably, in the study of Bolognini et al. (40) on healthy individuals, the authors found that right anodal parietal tDCS (compared to sham or left parietal tDCS) increased the effects of a multisensory visual search training and these effects transferred to non-trained tasks of visual orienting and visuo-spatial attention. These results open the way to parietal tDCS to increase visuo-spatial functions also in HVFDs patients undergoing multisensory trainings.

Thus, in the present proof-of-principle study on brain-damaged patients with chronic HVFDs, we investigate the behavioral effects of a single application of anodal tDCS over the ipsilesional occipital cortex, the ipsilesional or the contralesional posterior parietal cortex (PPC), combined with multisensory detection task. The aims are multiple: first, the comparison of an occipital tDCS protocol (previously combined with restitutive approaches) with a parietal tDCS protocol (40) not yet tested in individuals with HVFDs. Moreover, in post-stroke rehabilitation NIBS often targets interhemispheric (im)balance, with the aim of reducing the hyperactivation of the spared, contralesional, hemisphere or of increasing the excitation of the ipsilesional one [e.g., (24, 37, 41)]. Nonetheless, a more complex model taking into account the amount of lesional load (42) highlights the central role of the spared hemisphere – whose plasticity should be promoted with excitatory protocols – when the ipsilesional one has sustained large structural damages. Accordingly, we also investigated the effects of contralesional parietal tDCS, characterizing the extension and location of grey and white matter, by means of lesion mapping. Finally, we explored clinical, lesional, or behavioral differences between left and right brain-damaged participants.

In the present study, we investigated the behavioral effects of a single application of anodal occipital and parietal tDCS applied during a multisensory audio-visual training aimed at promoting compensatory oculomotor scanning behavior in brain-damaged patients with chronic HVFDs. To our knowledge, this is the first time that tDCS has been combined with a compensatory training for chronic HVFDs. In previous clinical reports, occipital tDCS was combined with multiple-session restorative trainings [e.g., (35)]. Here, instead, we performed an explorative study to assess the potential benefit of combining tDCS during an audio-visual detection training involving multisensory and attentional orienting networks.

Overall, we found that both ipsilesional occipital and ipsilesional parietal anodal tDCS accelerated visual field scanning (EF and Triangle tests) immediately at the end of a single 30-min training, whereas sham tDCS did not improve the performance of brain-damaged patients in a chronic stage of illness.

Early tDCS studies [e.g., Antal et al. (60, 61)] demonstrated that the application of offline tDCS on primary visual areas can induce polarity-dependent, short-lived, changes in the excitability of the underlying cortex; in particular, anodal tDCS was associated with an increase in excitability of the primary visual cortex, as indexed by a transient reduction of TMS-induced phosphenes threshold [Antal et al. (60)], and an increase in the N70 component of visual evoked potentials (61). Accordingly, occipital tDCS has been combined with restorative trainings (21, 22, 34), to boost the excitability of the underlying, hypofunctioning, visual areas in synergy with the rationale of visual restitution therapies. In our investigation, however, tDCS was applied on-line during a multisensory training requiring participants to overtly orient their attention toward audio-visual stimuli. Therefore, considering that the effects of occipital tDCS were observed in visual search tasks and that the effects of on-line protocols are known modulate task-related networks [e.g., (62)], we argue that the effects of occipital anodal stimulation go beyond the excitation of the underlying primary visual areas, possibly modulating those circuits mediating eye movements, including cortical–subcortical structures of the extra-geniculate pathways projecting to temporo-parietal areas, part of multisensory circuits and crucial for the deployment of visuospatial attention.

In this regard, we also found that responders to occipital tDCS in the Triangle test had smaller lesions affecting the parietal cortex and two branches of the Superior Longitudinal Fasciculus, a white matter tract connecting posterior and frontal areas that plays a main role in attentional orienting [Corbetta and Shulman, (63, 64)]. With respect to this, Bueichekù et al. (65) demonstrated that the strength of functional connectivity (even “at rest”) between primary visual areas and the posterior parietal lobule is a predictor of visual search efficiency; thus, the neuromodulation of these functional networks may improve visuo-spatial abilities, in turn speeding up visual search performance after a posterior stroke.

Moreover, tDCS applied over the posterior parietal cortex, an important hub of the network for visuospatial attention [Corbetta and Shulmann, (63)], improved visual search in patients with chronic HVFDs. Indeed, a facilitatory effect on RTs emerged in the Triangles test after ipsilesional parietal stimulation, in line with previous evidence (40), which showed, in healthy participants, that anodal parietal tDCS over the right hemisphere enhanced the effects of an audiovisual training as indexed by decreased latencies in visual search tasks. Furthermore, the effects of parietal neuromodulation on visuospatial functions have been demonstrated by other works investigating attentional orienting [Sparing et al. (39, 43, 66, 67)].

The effects of parietal tDCS on attentional orienting and visuospatial attention should also be interpreted from the perspective of interhemispheric balance. Indeed, the (un)balanced excitability of intra- and interhemispheric circuits is the focus of NIBS interventions for the recovery of motor functions (68), language deficits (69) and HVFDs (37). Accordingly, increasing the excitability of the lesioned hemisphere or reducing the, presumably pathological excitation of the spared one would promote functional recovery. With regard to visuospatial attention, this model was shown to be effective in diminishing the attentional asymmetries of hemispatial neglect (70). Within this framework, the present findings suggest that even in chronic HVFDs, the increased excitability and strengthening of the affected parietal circuits brought about anodal parietal tDCS may favor a better deployment of attention to the contralateral, blind, hemifield, resulting in an overall improvement of the visual exploration of such contralesional sector of the space. Nonetheless, contralesional parietal tDCS seems to boost search times but selectively in those participants with larger lesions of the temporo-parietal cortex or of white matter tracts connecting the occipital cortex with temporal and frontal areas. These results are in line with the bimodal hypothesis (42, 71) of plasticity modulation in brain recovery: in those patients with larger brain lesions, the excitability of the contralesional hemisphere should be promoted (rather than inhibited) as it would compensate for functions previously mediated by the lesioned one. In our sample of chronic patients with larger lesions, contralesional tDCS may have boosted those circuits that adapted to brain lesions to subserve visual exploration.

Finally, in light of the well-known hemispheric differences in visuospatial processing, we further characterized the effects of ipsilesional and contralesional tDCS considering the possible role of the hemispheric side of the lesion. Overall, the explorative analyses showed that patients with right-sided hemispheric damage seemed to benefit more from the ipsilesional occipital (Triangles test) and parietal (EF test) tDCS than left-lesioned ones. These results depict a complex relationship between lesion side, functional lateralization, and inter-hemispheric balance asymmetries. As for parietal tDCS, hemispheric asymmetries of fronto-parietal attentional circuits have been largely investigated [Corbetta and Shulmann, (63)], with a certain specialization for the right hemisphere, especially, for a ventral circuit mediating attentional re-orienting. Moreover, Bolognini et al. (40) found that anodal tDCS of the right PPC improved visual search performance. Here, however, patients with a left-sided brain lesion did benefit from right parietal (i.e., contralesional) tDCS. Therefore, ipsilesional excitatory parietal stimulation seems to be more effective to reactivate core areas of right-sided injured parietal circuits. Although such inter-hemispheric asymmetries are well characterized for parietally-mediated visuospatial attention, evidence about structural or functional asymmetries in the occipital pathways of patients with HVFDs remains limited (37). A recent study by Pietrelli et al. (72) analyzed the oscillatory activity of the posterior cortices of hemianopic patients and found an imbalanced inter-hemispheric alpha activity, which could predict the performance in visuo-spatial tasks. Notably, patients with right-sided hemispheric lesions showed more alterations of the oscillatory activity and a stronger inter-hemispheric imbalance, possibly mediated by cortico-cortical connections of the more right-lateralized parietal networks. Overall, this intriguing account may explain the larger benefits brought about by ipsilesional occipital and parietal tDCS in the right-damaged participants in our study. However, this hypothesis needs further confirmation, given the small- and not completely balanced- sample tested in the present work. The complex picture emerging from these results highlights the need for an optimal characterization of intra and inter-hemispheric functional mechanisms in HVFD (e.g., by means of EEG or functional MRI), as a crucial step toward a more tailored and effective application of NIBS for visual rehabilitation (73). Moreover, the understanding of HVFDs through the lenses of interhemispheric, oscillatory-based models may pave the way for other tES protocols beyond tDCS, such as the multifocal, cross-frequency transcranial alternate current stimulation (tACS) approach proposed by Raffin et al. (2020), or tDCS-tACS integrated protocols (24, 37).

The results discussed so far are confined to a reduction of RTs in visual search; in fact, the ACC was not modulated by tDCS. This null result is not surprising: a single application of tDCS, as well as a single training session, cannot be sufficient to induce more robust changes at the level of accuracy in patients with chronic VFDs. In fact, conventional multisensory trainings require multiple days of training [usually 2–3 weeks; (17)], especially in chronic patients whose neuroplastic changes require longer training than in sub-acute ones (21). We believe the same reasoning applies to the null findings on the visual detection task where no gains in speed and accuracy were induced by the multisensory training, whether coupled with real or sham stimulation (40).

The present study has some limitations that need to be mentioned. First, we acknowledge that the RTs improvements discussed above are small; nonetheless, we believe such changes in visual search after a single tDCS session coupled with multisensory training represent a promising finding, considering that visual search by itself was not trained, suggesting a transfer of training that could be larger and stronger after multiple days of training. However, a future study should better characterize the impact of other factors that may affect behavioral performance and rehabilitation, such as the level of alertness of the participants. Finally, whereas the sample size appeared sufficient for repeated-measure analyses, it should be increased to observe reliable correlations with clinical and lesional variables and should be better balanced for the lesional side to draw solid conclusions about hemispheric asymmetries.

In conclusion, the present results suggest that occipital and parietal tDCS combined with an audiovisual training may improve visual search speed in HVFDs. Importantly, however, these effects have to be confirmed and extended by repeated sessions of tDCS combined with multisensory trainings, after which, we may expect faster behavioral improvement with a positive impact on the activities of daily living.

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found at DOI: 10.5281/zenodo.10041851.

The studies involving humans were approved by Ethics Committee of IRCCS Istituto Auxologico Italiano. The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.

LD: Conceptualization, Data curation, Formal analysis, Methodology, Visualization, Writing – original draft, Writing – review & editing. CC: Conceptualization, Data curation, Investigation, Methodology, Project administration, Writing – review & editing. LM: Investigation, Methodology, Writing – review & editing. SBM: Investigation, Methodology, Writing – review & editing. NB: Conceptualization, Methodology, Project administration, Resources, Supervision, Writing – review & editing.