Hemispheric Asymmetry in TMS-Induced Effects on Spatial Attention: A Meta-Analysis

Hemispheric asymmetry is a fundamental principle in the functional architecture of the brain. It plays an important role in attention research where right hemisphere dominance is core to many attention theories. Lesion studies seem to confirm such hemispheric dominance with patients being more likely to develop left hemineglect after right hemispheric stroke than vice versa. However, the underlying concept of hemispheric dominance is still not entirely clear. Brain stimulation studies using transcranial magnetic stimulation (TMS) might be able to illuminate this concept. To examine the putative hemispheric asymmetry in spatial attention, we conducted a meta-analysis of studies applying inhibitory TMS protocols to the left or right posterior parietal cortices (PPC), assessing effects on attention biases with the landmark and line bisection task. A total of 18 studies including 222 participants from 1994 to February 2022 were identified. The analysis revealed a significant shift of the perceived midpoint towards the ipsilateral hemifield after right PPC suppression (Cohen's d = 0.52), but no significant effect after left PPC suppression (Cohen's d = 0.26), suggesting a hemispheric asymmetry even though the subgroup difference does not reach significance (p = .06). A complementary Bayesian meta-analysis revealed a high probability of at least a medium effect size after right PPC disruption versus a low probability after left PPC disruption. This is the first quantitative meta-analysis supporting right hemisphere-specific TMS-induced spatial attention deficits, mimicking hemineglect in healthy participants. We discuss the result in the light of prominent attention theories, ultimately concluding how difficult it remains to differentiate between these theories based on attentional bias scores alone.

Investigating the role of task relevance during rhythmic sampling of spatial locations

Recently it has been discovered that visuospatial attention operates rhythmically, rather than being stably employed over time. A low-frequency 7-8 Hz rhythmic mechanism coordinates periodic windows to sample relevant locations and to shift towards other, less relevant locations in a visual scene. Rhythmic sampling theories would predict that when two locations are relevant 8 Hz sampling mechanisms split into two, effectively resulting in a 4 Hz sampling frequency at each location. Therefore, it is expected that rhythmic sampling is influenced by the relative importance of locations for the task at hand. To test this, we employed an orienting task with an arrow cue, where participants were asked to respond to a target presented in one visual field. The cue-to-target interval was systematically varied, allowing us to assess whether performance follows a rhythmic pattern across cue-to-target delays. We manipulated a location's task relevance by altering the validity of the cue, thereby predicting the correct location in 60%, 80% or 100% of trials. Results revealed significant 4 Hz performance fluctuations at cued right visual field targets with low cue validity (60%), suggesting regular sampling of both locations. With high cue validity (80%), we observed a peak at 8 Hz towards non-cued targets, although not significant. These results were in line with our hypothesis suggesting a goal-directed balancing of attentional sampling (cued location) and shifting (non-cued location) depending on the relevance of locations in a visual scene. However, considering the hemifield specificity of the effect together with the absence of expected effects for cued trials in the high valid conditions we further discuss the interpretation of the data.

On the feasibility of concurrent human TMS-EEG-fMRI measurements

Simultaneously combining the complementary assets of EEG, functional MRI (fMRI), and transcranial magnetic stimulation (TMS) within one experimental session provides synergetic results, offering insights into brain function that go beyond the scope of each method when used in isolation. The steady increase of concurrent EEG-fMRI, TMS-EEG, and TMS-fMRI studies further underlines the added value of such multimodal imaging approaches. Whereas concurrent EEG-fMRI enables monitoring of brain-wide network dynamics with high temporal and spatial resolution, the combination with TMS provides insights in causal interactions within these networks. Thus the simultaneous use of all three methods would allow studying fast, spatially accurate, and distributed causal interactions in the perturbed system and its functional relevance for intact behavior. Concurrent EEG-fMRI, TMS-EEG, and TMS-fMRI experiments are already technically challenging, and the three-way combination of TMS-EEG-fMRI might yield additional difficulties in terms of hardware strain or signal quality. The present study explored the feasibility of concurrent TMS-EEG-fMRI studies by performing safety and quality assurance tests based on phantom and human data combining existing commercially available hardware. Results revealed that combined TMS-EEG-fMRI measurements were technically feasible, safe in terms of induced temperature changes, allowed functional MRI acquisition with comparable image quality as during concurrent EEG-fMRI or TMS-fMRI, and provided artifact-free EEG before and from 300 ms after TMS pulse application. Based on these empirical findings, we discuss the conceptual benefits of this novel complementary approach to investigate the working human brain and list a number of precautions and caveats to be heeded when setting up such multimodal imaging facilities with current hardware.