“People like to build rules, even when rules don’t necessarily apply,” said James Cavanagh, a former postdoctoral fellow at Brown and assistant professor of psychology at the University of New Mexico. Cavanagh and several colleagues recently published a study in the Journal of Neuroscience about how humans apply rules and strategies to tasks they have to complete, using new technology to analyze the data they collected.
The researchers were the first ever to be successful in creating a computer algorithm to recognize specific patterns in electroencephalography data, which was used to analyze neural activity as participants completed tasks, Cavanagh said.
According to electrophysiological data, the prefrontal cortex was activated when the participants were learning the initial task, wrote Michael Frank, associate professor of cognitive, linguistic and psychological sciences, in an email to The Herald.
The prefrontal cortex is “particularly involved in executive functions, or cognitive control, that allows us to make more complex choices by taking into account contexts, past events, goals, information in working memory, etc.,” wrote Anne Collins, a postdoctoral research associate in the Department of Cognitive, Linguistic and Psychological Sciences, in an email to The Herald.
“This study showed that people tend to construct abstract rules even when it is not needed in a task, but that this allowed them to then generalize the rules to new tasks,” Frank wrote.
In the study, participants were shown sets of shapes and colors on a computer and had to learn to press the correct button associated with the particular figure.
“There were just four different buttons on the keyboard, … and they aren’t told which is correct but have to learn by just trying out different buttons for each shape and getting feedback that tells them if they are right or wrong,” Frank wrote.
Next, the participants completed two more sets with different colored shapes. In the first additional set of tasks, participants could use rules they created to their advantage to successfully finish the tasks. But in the following set, these rules stymied participants’ efforts to select the right button.
EEGs connected to participants’ heads allowed researchers to collect data on what parts of the brain were activated during the experiment. The researchers collected data from 35 participants in the study.
Throughout the sets, “it’s very clear that people do develop rules to help them,” Cavanagh said, adding that participants naturally created rules even when they did not turn out to help.
“It’s probably because it’s most often a rewarding strategy: Structuring new information helps us to simplify it when possible, and to generalize it,” Collins wrote.
For the study, the team of researchers had to develop an algorithm to understand the EEG data, Cavanagh said. “We trained an algorithm to notice when people were attending to color and to shapes.”
The program was difficult to make because each person uses different neural networks in his brain when recognizing shapes or colors, he said. “It took quite a few months to get these algorithms to be bias-free.”
Since the algorithm is a new application of technology, researchers were surprised by how successful it turned out to be in recognizing patterns, Cavanagh said.
“What is particularly striking about this result is that (the researchers) were able to decode the rule structures that different people had constructed,” said David Badre, assistant professor of cognitive, linguistic and psychological sciences, who was not involved with the study. The researchers could even use this structure to “predict how well (participants) would generalize that rule to a new task,” he added.
This study may hold a lot of potential for future research and discovery in learning and memory.
“We will try to understand better what benefits come from structuring what we learn, to figure out what constrains how we structure it and to observe more precisely how the brain creates that structure,” Collins wrote.
Collins and her team began this study in 2011. “We have already conducted pilot experiments for future investigations, and are hoping to publish more results within a year,” she added.
The team’s research on how the brain creates a series of rules to respond to varied contexts could “eventually inform the types of strategies that are used when teaching new skills,” Frank wrote. It could even help “encourage those who are less likely to engage this system” to develop a systematic approach to learning.
The individual differences in rules participants made “may help us explain how people differ in their ability to successfully adapt to novelty in their everyday lives, as well as better understanding neurological and psychiatric disorders that compromise this ability,” Badre said.