Vision is one of the most fundamental senses, shaping how we perceive, navigate and interact with the world around us. Yet for more than 12 million Americans living with visual impairments, even small deficits can profoundly impact daily life, limiting independence and overall quality of life.

Researchers have long recognized the potential of visual perceptual learning (VPL) – a process by which the brain improves its ability to detect subtle differences in visual stimuli, such as fine patterns or orientations – to enhance vision. VPL is already being explored in professional contexts such as radiology, where precise detection of faint anomalies in medical images can save lives. However, a significant challenge remains: improvements in visual perception through VPL are typically confined to the exact part of the visual field that is trained, greatly restricting its broader clinical and practical applications.

To address this challenge, Luke Rosedahl, Ph.D., principal investigator, an assistant professor in the Department of Biomedical Engineering within the College of Engineering and Computer Science at Florida Atlantic University, and a Sensing Institute (I-SENSE) fellow, has received a $746,998, three-year grant from the National Eye Institute of the National Institutes of Health.

The grant will support his research aimed at uncovering the neural mechanisms that allow VPL to generalize beyond the specific conditions under which it is acquired. Rosedahl's work focuses on how different forms of attention – such as feature-based attention, which focuses on specific characteristics of visual stimuli, and spatial attention, which directs focus to particular locations in the visual field – interact to enable the transfer of visual learning to untrained areas.

Rosedahl's project will employ a combination of computational modeling, brain imaging, and neurochemical analysis to study how visual information is reorganized during learning. The research will specifically investigate a technique known as "double-training," in which training on a secondary, seemingly unrelated task at a new visual field location prompts the transfer of previously acquired visual skills to that location. By integrating data from behavioral performance measures, functional magnetic resonance imaging (fMRI), and changes in neurotransmitter concentrations measured through magnetic resonance spectroscopy, the project aims to develop a unified model of visual processing, VPL and attention.

"We expect this model to not only advance fundamental neuroscience but also serve as a valuable resource for the broader scientific community studying visual learning in real-world scenarios," said Rosedahl.

The implications of this research are far-reaching. For individuals with visual impairments, the ability to generalize visual learning across the visual field could dramatically improve the efficiency and effectiveness of rehabilitation programs. For professionals in fields such as radiology, surveillance or any domain that relies on nuanced visual discrimination, these insights could inform advanced training methods that enhance accuracy and performance. Furthermore, the principles uncovered in this research could inspire the design of AI systems that learn and adapt in ways more analogous to the human brain, particularly in tasks requiring sophisticated visual judgment.

Rosedahl's long-term research goal is to understand the interactions between multiple visual processes including category learning, visual perceptual learning, and attention, to optimize both training paradigms and rehabilitation strategies. His work builds on prior findings that VPL is typically highly location-specific, aiming to uncover how combinations of attention mechanisms can overcome this limitation.

Over the next three years, Rosedahl and his team will work to decode the neural processes underlying flexible visual learning, paving the way for more effective interventions for individuals with visual impairments and novel approaches to training in professional fields that demand high-level visual expertise.

"Professor Rosedahl's work addresses a fundamental challenge in vision science with enormous real-world impact," said Stella Batalama, Ph.D., dean of the College of Engineering and Computer Science. "By uncovering how the brain can generalize visual learning, this research could transform vision rehabilitation and professional training, redefining what is possible for individuals with visual impairments and advancing our understanding of human perception. It also has the potential to inspire innovative approaches to learning and adaptation in both humans and machines."