Ian Antón Oldenburg is a postdoctoral researcher in the department of Molecular and Cell Biology at the University of California, Berkeley, working in the lab of Hillel Adesnik. He received a B.S. in biology from Carnegie Mellon University. In 2014, he completed his Ph.D. in neuroscience at Harvard University under the supervision of Bernardo Sabatini for work investigating the interactions between the basal ganglia and several cortical areas. In general, Oldenburg is interested in understanding how neurons interact with each other to change behavior. However, these interactions can be diverse, depending heavily on the state of the animal and the neurons’ identities. In his current research, he is developing and implementing new technology that allow for the specific perturbation of many individual neurons in order to dissect these interactions at a previously intractable scale.
“Editing of the Neural Representation of Perception”
The perception of sensory stimuli, such as sights or sounds, depends on the activity of large networks of neurons working together. For years, neuroscientists studied sensory systems one neuron at a time by observing how those neurons responded to simple stimuli such as vertical, horizontal, or tilted black bars. Neurons seem to prefer one orientation to another, and theories were constructed to infer how vision arose from the collective activity of these neurons. More recently, neuroscientists have developed much more powerful tools to directly measure and manipulate far greater number of neurons and gain much richer insights into how groups of neurons represent and process sensory information. Yet these powerful new techniques have their limitations too. For instance, they are limited to manipulating neurons that are either very close to each other or neurons that are far apart but are the exact same genetic type. In reality, neuronal networks are densely interwoven with each other and consist of cells nearby and far apart, and genetically related and genetically distant. In fact, the only feature that may link a network together is how those neurons respond to stimuli. Until now, however, there has been no way to identify and manipulate the activity of networks of neurons based on that straightforward criterion. We have developed, for the first time, a technology that can do just this called “multi-photon structured light.” With this, we can identify and suppress the activity of specific neurons based on their activity patterns alone. We call this ability “editing the neural representation of perception,” and have recently piloted it in mice. These mice are trained to identify the orientation of a bar by moving their whiskers. We hypothesize that neural networks amplify the activity of neurons that respond similarly to the bar’s orientation. This integration across many neurons makes for more stable representations of perceptions that are immune to the variable activity of any individual neuron. By overcoming the spatial or genetic limitations of previous techniques, we can ask new questions about how groups of neurons act in concert. This work has implications beyond the mouse’s sensory systems, shedding light onto how information is processed in any brain region.