A new global research effort to study the brain’s decision-making process includes one UT-Austin neuroscience lab.
The International Brain Laboratory, IBL, is a virtual laboratory launched in September composed of 21 separate labs around the world. The group of labs will conduct a joint experiment measuring brain activity in mice by giving them a simple decision-making task which involves presenting a visual stimulus that appears on either the left or right of a screen to the mouse. The mouse is tasked with turning a wheel to align that image with the center of the screen.
The experimental labs will conduct these experiments, said Alexandre Pouget, a neuroscience professor at the University of Geneva and IBL group leader. The theoretical labs, such as that of UT neuroscience associate professor Ila Fiete, will create the initial mathematical models that the experiment tests.
Although it is made up of multiple labs, the IBL functions as a single unit in the sense that it focuses on researching a single behavior and integrates theoretical and experimental approaches of understanding, Fiete said.
“These (labs) were chosen because of (their) proven track record in solving major problems within neuroscience,” said Anne Churchland, principal investigator of the Churchland Lab in New York.
Just three scientists started this project, Pouget said. Through their connections in the field, they were able to identify the best labs in the world that participated in this type of work and begin collaboration.
“We picked up our phones — the old fashioned way — and talked to those guys,” Pouget said.
The IBL is unique because of its size and its goal in standardizing experimental procedures, Churchland said. Each of the labs will run the same experiments with standardized experimental conditions so that researchers can pool data from the different experimental sites.
Standardization — keeping all experimental variables consistent — is important to ensure reproducibility across labs, an existing problem in science, Fiete said. The IBL will ensure that all labs use shared instruments and a standardized experimental protocol to maximize efficiency.
The IBL will also attempt frictionless, barrier-free data sharing and analysis, which has historically been difficult in neuroscience, Fiete said. Neurophysiological data is highly multidimensional because it involves many different kinds of signals, which makes sharing and standardization difficult.
“Nobody has ever attempted this — to have 21 labs focus on the same task,” Pouget said. “Instead, it’s been the opposite. There are strong incentives for labs to develop new tasks.”
This will be the first organization in which experimentalists and theorists collaborate in a neuroscience project to a such a high degree, Pouget added.
“That’s a bit unusual in neuroscience. It’s common in physics, where theory plays a huge role, but neuroscience is a very new science — it’s barely 50 years old,” Pouget said. “Like any young science, it’s currently dominated by people measuring things. But now we’re approaching a time where the theorists are bound to play a really critical role to start making sense of all this data that is being collected.”
The ultimate goal is integrating the data into a grand theory of decision-making to explain the contributions of all the different parts of the brain.
“(Integrating data) is a formidable challenge, in my opinion,” Pouget said. “And it’s far from clear whether or not we’ll succeed there.”
The IBL is also hoping to change the dominating mentality of competition in neuroscience.
“Here, we’re going to work together instead of competing against each other,” Pouget said. “(That) hasn’t really been done yet in neuroscience, so we’re hoping that we can convince people that it’s really worth giving it a shot.”
