According to a review highlighted by EurekAlert, intracortical microstimulation is rapidly becoming more than a laboratory curiosity — it is beginning to reshape how researchers build artificial sensory channels, engage cortical plasticity, and design the next generation of brain–computer interfaces.
From rigid probes to flexible, biohybrid interfaces
The review walks through a familiar translational dilemma. Conventional microwire and silicon electrodes are substantially stiffer than the soft cortical tissue they penetrate, a mechanical mismatch that drives micromotion-related tissue damage, inflammation, and glial scar formation over time. The shift underway — toward flexible, biomimetic, and biohybrid interfaces — is not simply a materials story; it is a clinical one. This shift allows us to imagine implants that behave like guests rather than intruders in cortical tissue, where long-term stability of recording and stimulation depends on the device integrating with, rather than fighting, the surrounding neuropil. For the neuroimaging and device-engineering community, the trajectory is worth tracking closely, because chronic stimulation and recording stability ultimately hinge on how the surrounding tissue responds to the implant.
Artificial perception, pattern by pattern
Consider the implications of patterned ICMS in the primary somatosensory cortex. Where single-electrode stimulation produces a diffuse tingling, multi-electrode arrays driven by carefully designed spatiotemporal pulse trains can evoke richer percepts — tactile edges, curvature, even apparent motion — improving the controllability and structure of artificial touch, though the review notes that current patterns still do not fully reproduce the complexity of natural tactile signals. In the visual cortex, coordinated stimulation produces phosphenes — perceived spots or lines of light — that researchers have assembled into simple two-dimensional patterns and used for object-localization tasks in blind participants. The picture emerging is one of ICMS moving beyond producing a single artificial sensation toward constructing structured sensory information, yet the reviewers are careful to flag that current visual prosthesis studies remain limited to relatively simple shapes and letters, and that predicting phosphene responses, identifying effective electrode combinations, and maintaining stable stimulation across long timescales remain major open challenges.
Plasticity as the quiet substrate
Perhaps the most striking thread in the review is the one about learning. Animals can be trained to interpret artificial stimulation patterns and use them to guide behavior, a finding that suggests the cortex may not require a signal to exactly mimic natural sensory input — it can learn the meaning of an unfamiliar one. Such a strategy offers flexibility for future bidirectional brain–computer interfaces, where electronic systems continuously exchange information with the brain. The review also examines a more ambitious possibility: using ICMS not simply to produce an immediate percept but to gradually reshape how cortical circuits function. Paired and activity-dependent stimulation protocols have been shown to alter functional connectivity between cortical regions. In one closed-loop paradigm, spontaneous neural activity in the motor cortex was used to trigger stimulation of the somatosensory cortex with a controlled delay, opening a window onto how timing, not just location, shapes plastic remodeling. For translational researchers, this is the direction worth watching — ICMS as a tool for reconfiguring the networks that interpret stimulation, not only for delivering it.