A Trimodal Approach to Brain Function: Combining tDCS, fMRI, and fNIRS for Mechanistic Insight

Advancing the study of neural networks depends on the ability to map where, when, and how brain activity unfolds in response to both natural cognitive processes and neurological dysfunctions. Yet each neuroimaging modality imposes inherent trade-offs that can limit the completeness and interpretability of the information obtained. These constraints have made multimodal approaches that combine neural stimulation with complementary imaging techniques increasingly necessary. Integrating transcranial direct current stimulation (tDCS), functional magnetic resonance imaging (fMRI), and functional near-infrared spectroscopy (fNIRS) into a single trimodal framework represents a promising strategy for overcoming the limitations of single-modality designs. Such a system offers whole-brain spatial coverage, surface-level temporal sensitivity, and direct causal manipulation of neural activity, enabling a more comprehensive characterization of large-scale brain function.

Functional magnetic resonance imaging (fMRI) provides detailed spatial maps but exhibits slow temporal dynamics due to the nature of the blood-oxygenation-level-dependent signal (Cui et al., 2010). Methods with stronger temporal resolution often fail to reach deeper brain structures and are highly susceptible to noise, which complicates interpretation. Additional concerns involve ecological validity and the difficulty of translating laboratory findings into real-world clinical settings. Integrating imaging tools that compensate for one another's weaknesses has therefore become an increasingly important strategy. Transcranial direct current stimulation (tDCS), a noninvasive method that modulates cortical excitability, is frequently paired with neuroimaging to investigate its mechanisms of action (Filmer et al., 2014). With the combination of tDCS, fMRI, and fNIRS, researchers can move beyond correlational interpretations and gain direct insight into how neural networks reorganize during and after stimulation.

The rationale for combining tDCS, fMRI, and fNIRS is grounded in the idea that each modality compensates for the limitations of the others. Together, these tools address the core constraints of human neuroscience, including resolution, depth, and causal manipulation. fMRI contributes macroscale spatial resolution across the entire brain, including subcortical structures, and therefore provides the anatomical foundation of the system (Novi et al., 2023). The primary weakness of fMRI is the slow and indirect nature of the BOLD response, which limits temporal interpretation. fNIRS complements fMRI by offering fast measurements of changes in oxyhemoglobin and deoxyhemoglobin at the cortical surface with relatively high temporal precision and strong tolerance to motion artifacts (Strangman et al., 2002). Although restricted to superficial cortex, fNIRS can refine or validate BOLD signals in corresponding cortical areas, improving interpretability (Cui et al., 2010). tDCS introduces a causal perturbation to neural circuits by altering cortical excitability. Because tDCS produces diffuse current flow, concurrent imaging is essential to understand how stimulation affects both targeted and distant brain regions (Saiote et al., 2013).

Implementing tDCS, fMRI, and fNIRS in a single experimental environment introduces significant engineering and methodological challenges. The MRI environment imposes strict safety and artifact-reduction requirements. tDCS electrodes must be fabricated from nonferrous materials, such as carbon fiber, and connected via shielded cables that safely exit the scanner room. These precautions reduce the risk of radiofrequency-induced heating and minimize distortion in the fMRI signal (Saiote et al., 2013). fNIRS hardware must also be fully compatible with MRI. Light sources and detectors require nonmetallic construction, and optical fibers must be routed so that they do not interfere with the MRI coil or produce optical noise (Plichta et al., 2006). Achieving temporal alignment across all systems requires synchronized timing signals to ensure coordination among stimulation onset, fMRI slice acquisition, and fNIRS sampling.

Because fMRI typically samples every two or three seconds while fNIRS samples multiple times per second, temporal fusion demands deconvolution or cross-modal regression techniques to match signal time courses (Cui et al., 2010). Spatial integration also requires substantial computational work. Structural MRI provides an anatomical reference for coregistering fNIRS optodes and tDCS electrode locations to the cortical surface. Finite element modeling of the tDCS montage can then be aligned with fMRI data to compare predicted current distribution with measured hemodynamic changes (Bhutta et al., 2016). A unified spatial framework is essential for coherent interpretation.

Despite its advantages, the trimodal approach faces notable practical and methodological challenges. The computational load is substantial because high-resolution fMRI combined with high-frequency fNIRS results in sizeable datasets that require specialized software and significant processing resources (Patel et al., 2020). Financial constraints also restrict widespread adoption because implementing a full trimodal system requires access to a high-field MRI scanner, MR-compatible fNIRS equipment, and MR-safe tDCS devices. Practical challenges include lengthy setup procedures, demanding coregistration steps, and patient discomfort. Individuals must remain as still as possible in the scanner while experiencing tDCS sensations and wearing multiple devices simultaneously. Continuous monitoring is required to prevent skin irritation, overheating, or hardware displacement. Despite these challenges, the trimodal approach is well-suited for research questions that require both whole-brain imaging and causal inference. Usually, the system can reveal how tDCS alters connectivity within memory or attention networks while fNIRS simultaneously monitors prefrontal hemodynamic responses (Di Rosa et al., 2019). In clinical contexts, such as stroke rehabilitation, fMRI can identify preserved and damaged pathways, while fNIRS verifies immediate cortical responses to stimulation. tDCS parameters may then be individualized in real time to optimize effective connectivity (Qi et al., 2024). For mood disorders, researchers can evaluate whether stimulation enhances connectivity between prefrontal regulatory regions and limbic structures, which may inform targeted neuromodulation strategies.

The integration of tDCS, fMRI, and fNIRS represents meaningful progress toward achieving causal and spatiotemporally resolved neuroimaging. Although considerable engineering, computational, and financial obstacles remain, the benefits of this approach are significant. By combining neural stimulation with multiple complementary imaging techniques, researchers can move beyond descriptive models and obtain mechanistic insight into how the brain reorganizes in response to targeted stimuli or intervention, thereby more accurately reflecting real-life situations. The trimodal framework has strong potential to advance both experimental research and clinical practice by providing comprehensive, causal, and individualized information about human function.