Looking for something? Start here!

To acquire wisdom, one must observe

Looking for something? Start here!

Guest lecturer speaks on neuroscience

Doris Tsao from the University of California Berkeley took part in a lecture series honoring the legacy of Dr. John Lisman at the university on Nov. 15. Tsao spoke on her career and research in neuroscience with community members, explaining the neural pathways involved with different stimuli including object and facial recognition. 

Tsao is known for her work on using fMRI techniques in combination with single electrodes to locate neural areas involved in face and object recognition in monkeys. Her research has been published in publications including Science and the Journal of Neuroscience. She has been rewarded for her work with numerous awards including the National Institutes of Health (NIH) Pioneer Award and the Alden Spencer Award from Columbia University.

Tsao introduced her research by explaining how her lab focuses on the concept of the method in which the brain makes sense of the visual world, specifically objects, and how they examine this in macaque monkeys. She explained how a lot of the research is focused on the intertemporal (IT) cortex which is responsible for more complex coding of object stimuli after it receives input from lower-level input in V1-V4 cortices. 

Within her research as a graduate student, she found regions responsive for face recognition within the IT cortex. She explained how she wanted to explore these regions because of the positive implications they could have for the field of neuroscience as they would allow us to “understand the detailed code for facial identity, understand the general principles of object representation, and understand fundamental principles of cortical communication.”

Tsao went on to explain in detail how we can know for sure that we have an accurate feature code for facial identity if we can create the expected face or stimulus from a map of neurons. Within her study, they used computer mechanisms in the form of a “Shape Appearance Model” where there were generated faces that were presented to the monkeys while their brain responses were measured. They were successfully able to accurately reconstruct faces from the mapped-out cells in the monkey’s response to what the monkeys actually saw. 

To expand their understanding of the IT cortex, Tsao and other researchers examined a region of the IT cortex with an unknown role, “Network X” (i.e. other neighboring regions were specifically defined as being responsible for faces, bodies, scenes, colors). They did this by combining simultaneous fMRI image studies with electrical stimulation. They presented the unknown region with a series of object stimuli with various orientations and mapped out the responses. Tsao explained how the behavior of the cells in Network X were found to be comparable to the behavior of face cells. By further comparing the response of cells in this region on four axes of stimuli—inanimate/animate and spikey/stubby—they found that Network X responded mostly to spikey stimuli regardless of if they were animate/inanimate. Because of this they predicted that there was another region that would prefer stubby stimuli regardless of if they were animate or inanimate. 

Through fMRI tests in response to stimuli, they were able to find the small regions responsible for stubby stimuli. In the same way, they were able to reconstruct faces. They examined if they would be able to reconstruct arbitrary objects from neural activity maps in the four axes regions of the IT cortex they discovered through a “Generative Adversarial Network.” Although the results weren’t as accurate as the face reconstructions, they bore enough resemblance to be distinctive and imply the distinctive role these regions play in object mapping. 

Tsao continued to examine the ways she imagined discussing these results with Lisman based on her understanding of his theories from his published papers. Tsao concluded the lecture with Lisman’s own words, “The philosopher John Locke defined consciousness as ‘the perception of what passes in a man’s own mind’ implying that consciousness depends on a brian architecture in which thought can activate perceptual processes (e.g., imagery or auditory sensation) … If Locke’s definition of consciousness is correct, understanding the top-down information flow that produces imagery will provide a mechanistic description of consciousness.”

The lecture series, explained by Stephen Van Hooser (BIOL), is an initiative brought on by Jay Pepose, the first undergraduate student to work in Lisman’s lab, and Pepose’s wife, Susan Feigenbaum.

Get Our Stories Sent To Your Inbox

Skip to content