Preclinical study explores contact lens–based neuromodulation for depression

Researchers have developed a soft contact lens embedded with electrical stimulation electrodes that, in a mouse model of depression, reduced depression-like behaviors and altered brain activity associated with the disorder.

The device, described in a new study, uses a technique called temporal interference-based transcorneal electrical stimulation, or TI-TES. Two high-frequency electrical signals are delivered from electrodes on the lens surface and are designed to intersect at the retina, generating a lower-frequency field intended to activate retinal neurons and their downstream connections to brain regions involved in mood regulation, including the hippocampus and prefrontal cortex.

“This approach preserves vision and ensures conformal ocular coupling,” the study authors, led by Wonjung Park of the Department of Materials Science and Engineering, Yonsei University in Republic of Korea, stated.

“In a preclinical stress-induced mouse model that recapitulates key behavioral and biological features associated with depression, TI-TES enhanced behavioral resilience, restored prefrontal-hippocampal oscillatory synchrony, and normalized depression-related biomarkers. Machine-learning integration of behavioral, electrophysiological, and biological data consistently distinguished treated from control groups. These preclinical findings establish TI-TES as a non-invasive bioelectronic strategy that overcomes current limitations, enabling selective eye-brain pathway modulation for depression and potentially other brain disorders.”

The device

The lens integrates ultrathin electrode layers—a 3-nanometer gallium oxide base topped with a one-nanometer platinum layer—thin enough to maintain more than 80% optical transparency across the visible light spectrum. Stimulation electrodes were additionally coated with platinum nanoclusters, which reduced electrical impedance from 20.3 kilohms to 2.8 kilohms and roughly doubled charge storage capacity. Accelerated aging tests simulating 12 months of use showed minimal impedance change, from 2.78 to 3.02 kilohms.

Animal model and treatment

Experiments were conducted primarily in rd1 mice, a strain with inherited photoreceptor degeneration. Researchers chose this model specifically to eliminate normal visual input as a potential confounding factor, ensuring that any observed effects could be attributed to direct electrical activation of retinal output neurons rather than light-mediated signals.

Depression-like states were induced using chronic corticosterone injections, a method that replicates certain behavioral and biological features associated with depression. The optimized stimulation protocol—20 Hz envelope frequency, 200-millivolt peak-to-peak amplitude, 30 minutes per day—was applied for three weeks.

Behavioral results

Compared to untreated depressed mice, TI-TES-treated animals showed a 76% increase in locomotor activity and a 132% increase in time spent in the center zone during an open-field test, a measure of anxiety-like behavior. Immobility in the tail-suspension and forced swim tests—standard measures of behavioral despair—was reduced by 48% and 45%, respectively. Social preference and novelty scores in a three-chamber test also improved toward levels seen in non-depressed control animals.

In a separate comparison, the behavioral improvements produced by TI-TES were comparable in magnitude to those seen with fluoxetine, a widely used antidepressant drug, across three behavioral tests.

When researchers chemically destroyed retinal ganglion cells using intravitreal NMDA injection before applying TI-TES, the behavioral benefits disappeared, indicating that intact retinal output pathways are required for the treatment's effects.

Brain recordings

Neural probes implanted in the hippocampus and prefrontal cortex showed that TI-TES treatment restored synchronization between these regions. Phase-locking values in the theta band—a frequency range associated with hippocampal-prefrontal communication—were 163.6% higher in treated mice compared to depressed controls in hippocampal-prefrontal measurements and 192.6% higher within the hippocampus itself. Phase coherence between regions was 77.8% greater in the treated group than in untreated depressed animals.

Biological markers

Hippocampal tissue analysis showed that dendritic spine density—reduced in depressed mice—recovered by 48.6% with TI-TES treatment. Levels of mature brain-derived neurotrophic factor, a protein involved in synaptic plasticity, were reduced by 64.2% in depressed mice and recovered by 131.1% relative to that depressed baseline following treatment. Inflammatory markers IL-6 and TNF-α in the hippocampus were reduced by 65.8% and 56.8%, respectively. Plasma corticosterone levels fell by 48.1%, and serotonin levels recovered by 46.9% relative to depressed controls.

Machine learning classification

A support vector machine classifier trained on combined behavioral, electrophysiological, and biological data correctly separated depressed animals from controls and consistently grouped TI-TES-treated mice closer to the non-depressed control group than to the depressed group in multi-dimensional feature space. A permutation test with 1,000 label shuffles confirmed the separation was unlikely to result from chance.

Safety

Ocular safety testing in mice with intact retinas showed no detectable corneal or retinal tissue damage, cell death, inflammation, or change in intraocular pressure following three weeks of daily stimulation. Human corneal epithelial cells exposed to lens-conditioned medium showed 99.25% viability.

Limitations and next steps

The authors note several barriers to human use. The current system uses a wired configuration unsuitable for everyday wear and will require wireless integration. Individual differences in eye anatomy and tissue impedance may affect stimulation delivery. Long-term biocompatibility beyond the study's three-week window also remains to be established.

The researchers describe the findings as preclinical and call for further work in larger animal models before human trials, along with studies examining the effects of TI-TES in non-depressed subjects and broader investigation of which retinal cell types and pathways are most responsive to the stimulation.

Reference:

1. Park W, Paek J, Seo H, et al. Contact lens bioelectronic platform for non-invasive depression treatment with machine-learning-based evaluation. Cell Reports Physical Science. 2026;7(5)103303. doi:10.1016/j.xcrp.2026.103303