In a tiny beige room in Sidney Frank Hall, the doors to decision-making are being unlocked. Monitors hum and chairs swivel as the next study participant prepares to take her turn, leaving her backpack by the door and signing off on a consent form.
A bucket of salty electrolyte solution sits on the table. What appears to be a complicated swimming cap made of white mesh and dotted with blue and red electrodes floats in the liquid. This apparatus measures neural activity for an EEG, or electroencephalogram, which enables researchers to monitor the brain's activity during a task and record its levels in specific areas.
Jim Cavanagh, a postdoctoral research assistant in the Laboratory of Neural Computation and Cognition, extracts the dripping apparatus and brings it over to rest on the participant's head. This new kind of recorder no longer requires the use of messy conductive gel, he explains, as he massages the dots into the participant's scalp. The device can gather readings using nothing but water, and it straps on easily, so the process of preparation is relatively short.
"It takes a lot of processing to pull out exactly what we were doing, but until then, it just looks like this ocean of activity," Cavanagh explains, pointing to a landscape of squiggly black lines etched across the screen of a large monitor.
These imaging methods, along with behavioral analysis, can validate researchers' hypotheses about which regions of the brain are involved in cognitive processes, adds Anne Collins, the postdoc who developed the task, from her perch on a black chair.
The researchers tap some secret commands onto a keyboard, and off the participant goes. She is presented with a series of simple exercises. Shapes and colors flash before her eyes, but as fingers try to move quickly, her eyelids become heavy.
Meanwhile, the cap on her head takes careful measurements of her neural activity, tracking valuable data that will help to untangle the complicated processes behind decision-making. It is so accurate that it is possible to discern the blink of an eye in the resulting graph.
In this experiment, the cap specifically measures interactions between the prefrontal cortex and the basal ganglia, two sections of the brain involved in decision-making.
"It's like a game of telephone," explains Christina Figueroa, a lab manager. "While your prefrontal cortex is telling a different part of your brain to do something or perform a certain action, that other area will loop back around and tell the prefrontal cortex to do something or not do something."
Figueroa goes on to describe the basal ganglia as a "dam," with lots of different paths and options coming in from other parts of the brain. She says it acts as a "gatekeeper," driven by certain neurotransmitters to select what behaviors are ultimately executed.
Figueroa's research team looks to combine the two aspects of cognition that tend to be studied separately — the learning and the structural — in order to gain a fuller picture of the system. Much of the lab's work centers on using computational models to predict human behavior.
"A Parkinson's patient versus a healthy senior can perform differently on a task, and that can tell us something about how the brain works and build a computational model to perform the same way," Collins says.
The lab can have 10 to 12 projects running at once, all of which investigate the learning and decision-making process.
"When you actually break it down to your daily activities, you make decisions every moment of every day, whether it's to get up in the morning, where you're going to go," Figueroa says. "You can see how vital that process is to understanding ourselves."
The decision-making process can be influenced by everything from environment to genetics, and experiments require careful tinkering to tease apart the influence of different factors and hone in on fundamental processes. "In order to tap into those, we need to create tasks simple enough to be able to pinpoint certain functions," Figueroa says. "It can't be a videogame."
Figueroa admits that psychological testing is "somewhat notorious" for its uneventful experiments — the research team has had participants fall asleep at the monitor, or click the same button disinterestedly throughout the session. "We can't use that data, because it's not measuring the learning and decision making process," Figueroa says. "It's not giving us a good baseline or a good understanding. … It's not comparable to any other participants."
It can be difficult to understand the significance of an experiment as a participant. Access to information on a study is limited prior to the experiment because this knowledge can affect the way participants perform on a task.
"Only afterwards can you find out how cool our research really is," Figueroa says, adding that she enjoys the debriefing process because it gives subjects an idea of the importance of what they are doing beyond the compensation they receive.
"Imagine if you were really, really curious about evolution and you had no Darwin and you had no book," Figueroa says. "Without participants, it's like having no books in the library, nowhere to go to get that information."
