A patient with complete blindness caused by irreversible optic nerve damage partially recovered natural vision after participating in a clinical trial of electrical stimulation of the visual cortex conducted by researchers from the Universidad Miguel Hernández de Elche (UMH) and the CIBER in Bioengineering, Biomaterials and Nanomedicine (CIBER-BBN). The unexpected improvement was spontaneous, sustained over time, and independent of the implanted device.
The case was observed during a study designed to evaluate the safety and feasibility of a cortical visual prosthesis. Although the goal of the trial was to generate artificial visual percepts through direct brain stimulation, one participant experienced a progressive recovery of natural vision after more than three years of total blindness. The findings have been published in Brain Communications.
"To date, our Biomedical Neuroengineering Laboratory has conducted four clinical trials involving volunteers with severe blindness," explains Eduardo Fernández Jover, lead researcher of the study and director of UMH's Institute of Bioengineering. "As in all these trials, the aim was to generate artificial visual perceptions, not to restore natural vision. The fact that one participant showed a measurable and sustained visual improvement suggests the influence of individual factors that remain to be determined."
Neurologist Arantxa Alfaro Sáez, from Hospital de la Vega Baja de Orihuela and a member of UMH's NBio group, notes that "although some cases of visual recovery have been described in patients with severe optic nerve damage, these usually occur within the first months after injury. Observing recovery after such a long time is highly unusual."
The procedure involved the surgical implantation of an intracortical array of 100 microelectrodes into the primary visual cortex, the brain region responsible for processing visual information, Alfaro explains. Through this array, researchers applied controlled electrical stimulation patterns to elicit artificial visual percepts known as phosphenes.
Two days after surgery, while still hospitalized, the patient reported perceiving lights and movement in front of him. "We had barely begun stimulating the visual cortex to calibrate the system," Alfaro recalls, "but when we gestured, the patient was able to correctly describe the position of our arms and locate where people were around him." He described the experience as a moving shadow—his first natural visual perception years after becoming completely blind.
Visual training and sustained changes
Over the following months, the patient followed a daily visual training routine of at least 30 minutes, including standardized exercises of increasing complexity to assess light perception, spatial localization, motion detection, visual acuity, contrast sensitivity, and the search, identification, and tracking of objects, shapes, letters, and numbers.
Leili Soo, also a first author of the study, suggests that this training, together with the participant's motivation, may have played a relevant role in the partial recovery of natural vision. Notably, the improvement persisted even after the intracortical implant was surgically removed.
"Visual evoked potentials—electrical signals generated by the brain in response to visual stimuli and used to assess whether visual information reaches the cortex—were almost absent in this participant before the study," Soo explains. Over time, however, these signals gradually reappeared and improved, confirming a real and measurable recovery.
Overall, the participant showed a significant improvement in visual acuity and a marked increase in autonomy. He was able to consistently identify shapes and letters, improve coordination when grasping objects, and move with greater confidence in daily life. The patient reported that the recovered vision allowed him to function more safely in everyday activities.
Hypotheses on the origin of visual improvement
These findings may help inform new therapeutic approaches for visual rehabilitation in people with severe damage to visual pathways, or even in other types of brain injury, using non-invasive techniques such as transcranial electrical stimulation, Fernández Jover suggests. However, he emphasizes that "these results were observed in only one participant, which suggests that unique individual characteristics may have contributed to this outcome."
Many key questions remain unanswered, including how visual neural circuits function in detail, which stimulation parameters are optimal, and how the brain responds to long-term artificial stimulation. "Each brain is different, and responses can vary widely depending on pathology, duration of blindness, and residual vision," Fernández Jover cautions, adding that the observed recovery may not be reproducible in other patients. Future studies will determine whether this is an isolated case or a reproducible phenomenon.
"Precisely because of this diversity, we are deeply grateful to this patient and to all volunteers who have participated in these clinical trials," the researchers note. "None did so with the expectation of seeing again, but with the awareness that their contribution helps advance knowledge about how to restore the complex neural dialogue that makes vision possible."
The clinical study was conducted in close collaboration with IMED Elche Hospital. In 2021, UMH's Biomedical Neuroengineering Laboratory successfully implanted a device in the brain of a blind person that induced the perception of shapes and letters with higher resolution than previously achieved. The group has also developed technology capable of establishing bidirectional communication with the visual cortex, enabling more natural and functional artificial vision. As a result, implanted individuals have been able to recognize objects and letters and navigate complex environments.
The study is authored by Eduardo Fernández, Arantxa Alfaro, Leili Soo, Dorota Waclawczyk, Roberto Morollón, and Fabrizio Grani, from UMH's Institute of Bioengineering and the Bidons Egara Chair. Alfaro and Fernández Jover are also members of CIBER-BBN, funded by the Carlos III Health Institute.
Funding
This research was supported by the Spanish Ministry of Science, Innovation and Universities (PDC2022-133952-100), the European Union's Horizon 2020 program (grant agreement no. 899287, NeuraViPeR), and the Regional Government of Valencia's program for excellence research groups (PROMETEO CIPROM/2023/25).
