Learning to see in 3-D
Until now, scientists believed that learning in the newborn brain was achieved by pruning an over-wired neural circuitry. The traditional view was that genetic mechanisms laid down cortical connections in the womb and then sensory experience refined these connections to establish the ability to detect various visual features.
Now UCLA researchers show that vision improvement is achieved by dismantling the brain's original wiring and building new circuitry shaped by early experience and visual feedback. The discovery offers insights into how newborns learn and suggest potential targets for treating neurodevelopmental disorders.
Newborns have poor depth perception until 5 to 8 months of age, when their eyes begin working together. Neurons in the young cerebral cortex also must develop binocularity—the ability to receive and process information from both eyes at once. Yet more than half of the binocular neurons present during early brain development stop responding to one eye, preventing depth perception, or the ability to see in three dimensions.
Babies who don't receive normal visual input can suffer permanent impairment. For example, infants born with misaligned eyes will fail to develop depth perception and full acuity if the problem remains corrected.
UCLA scientists studied a mouse model to examine the development of depth perception. The team measured changes in neural activity during early development—tracking which stimuli that particular cells prefer--to uncover how and why the young brain learns. To monitor the same neurons, the authors combined a non-invasive microscope with genetic manipulations to make the cells glow green when active.
The scientists discovered that the group of neurons encoding visual features changed in response to visual experience. Large-scale changes in neural circuitry were achieved by controlling a small set of visual inputs. The team demonstrated that the brain establishes new binocular neurons by converting the most attuned neurons from the available pool. Visual information conveying depth was also encoded by neurons with the best tuning properties.
By refuting the idea that all connections in the cortex are malleable, the UCLA discovery proposes that specific pathways can be targeted for future therapeutic approaches to neurodevelopmental disorders. And it raises the question of whether these particularly plastic pathways are impacted by aging.
Joshua Trachtenberg, professor of neurobiology at the David Geffen School of Medicine at UCLA; and first author Liming Tan, a researcher in biological chemistry, are available for comment.
The journal Neuron publishes the findings.
The work was funded by grants from the National Eye Institute, National Institute of Biomedical Imaging and Bioengineering, National Institute of Neurological Disorders and Stroke, the W.M. Keck Foundation and the Howard Hughes Medical Institute.