Home | Pregnancy Timeline | News Alerts |News Archive Apr 14, 2015

The strength of brain areas were mapped during this study. Warm colors indicate high strength and cool colors indicate low strength. Image Credit: University of California Santa Barbara

When using our brain muddles solving a problem

Why do some people learn a new skill right away, while others only gradually improve? What is happening in their brains that creates this variation.

Researchers have taken a network science approach to answer this question and began by measuring the connections between different brain regions as participants learned to play a simple game.

The difference in neural activity between quick and slow learners provides new insight into what is happening in the brain during the learning process. Findings suggest that recruiting unnecessary parts of the brain for a task — or "over-thinking" — is critical in the difference between quick and slow processing.

The results are published in Nature Neuroscience.

Scott Grafton of the Brain Imaging Center at the University of California Santa Barbara (UCSB), and Nicholas Wymbs of the Human Brain Physiology and Stimulation Laboratory at Johns Hopkins University had study participants at UCSB play a simple game while research staff scanned their brains with fMRI [Functional magnetic resonance imaging or functional MRI] — to track blood flow through regions of the brain active while participants learned a given task. The game learned involved responding to a sequence of color-coded notes by pressing the corresponding button on a hand-held controller. There were six pre-determined sequences of 10 notes each, played multiple times during the scanning sessions. Researchers instructed participants to play the musical sequences as quickly and as accurately as possible, responding to cues given on a screen.

Participants were also required to practice the sequences at home while researchers remotely monitored these sessions as well. Participants were brain scanned at two, four and six weeks, to see how well home practice sessions helped them master the game skills. All participants' completion times dropped over the course of the study, but at different rates. Some picked up sequences immediately, while others gradually improved over the six-week period.

Danielle Bassett, a Skirkanich Assistant Professor of Innovation in the University of Pennsylvania School of Engineering and Applied Science's departments of Bioengineering and of Electrical and Systems Engineering, is an expert in network science and developed novel analysis methods to determine what was happening in the participants' brains to explain differences in processing times. But rather than measuring a single spot in the brain as more or less active, the researchers investigated the learning process as a complex, dynamic network. "We weren't using a traditional fMRI approach," Bassett added, "where you pick a region of interest and see if it lights up. We looked at the whole brain at once and saw which parts were communicating with each other the most."

Researchers compared 112 anatomical brain regions to measure how well they mirrored one another. The more two regional patterns matched, the more they were considered to be communicating. By graphing these connections, highly interconnected regions emerged.

"When a network scientist looks at these graphs, they see what is known as community structure," Bassett said. "There are sets of nodes in a network that are really densely interconnected to each other. Everything else is either independent or very loosely connected with only a few lines."

Bassett and her colleagues used algorithms to determine which brain nodes were incorporated into clusters and how cluster interaction changed over time. They went on to measure how long any two brain nodes remained in the same cluster while game participants played the same sequence repeatedly. By comparing these regional results, researchers identified where different brain functions worked together.

"If we look just at the visual and motor blocks, they have a lot of connectivity between them during the first few trials. But, as the experiment progressed, these areas essentially become autonomous. The part of the brain that controls movement of your fingers and the part of your brain that processes visual stimulus didn't really interact at all by the end of the trial."

Danielle Bassett, Skirkanich Assistant Professor of Innovation, departments of Bioengineering, Electrical and Systems Engineering, the University of Pennsylvania, School of Engineering and Applied Science's.

In some ways, this trend was unsurprising. Researchers observed the learning process only at a neurological level as participants' brains reorganized the flow of activity while picking up this new skill. Bassett: "What we think is happening, is that they see the first few elements of a sequence and realize which one it is. Then they can play it from motor memory. There no longer needs to be constant communication between the visual stream and their motor control."

With the neurological correlations of the learning process being identified, researchers explored the differences between participants, that might explain why some sequences were learned faster than others.

Counterintuitively, participants who showed decreased neural activity learned the fastest. A critical distinction was seen in areas not directly related to seeing the cues or playing the notes: the frontal cortex and the anterior cingulate cortex.

Bassett: "The reason this is interesting is that those two areas are hubs of the cognitive control network. It's the people who can turn off the communication between these two parts of their brain the quickest, who have the steepest drop-off in their completion times."

These cognitive control centers are thought to be most responsible for what is known as "executive function." This neurological trait is associated with making and following through with plans, spotting and avoiding errors and other higher-order types of thinking. Good executive function is necessary for complex tasks but might actually be a hindrance to mastering simple ones.

"It seems like those other parts are getting in the way for the slower learners. It's almost like they're trying too hard and overthinking it," Bassett said.

The frontal cortex and the anterior cingulate cortex are some of the last regions of the brain to fully develop in humans. This may explain why children are able to acquire new skills so quickly as compared to adults.

Further research will look into why some people are better than others at shutting down the connections in these two parts of the brain.

Abstract
Distributed networks of brain areas interact with one another in a time-varying fashion to enable complex cognitive and sensorimotor functions. Here we used new network-analysis algorithms to test the recruitment and integration of large-scale functional neural circuitry during learning. Using functional magnetic resonance imaging data acquired from healthy human participants, we investigated changes in the architecture of functional connectivity patterns that promote learning from initial training through mastery of a simple motor skill. Our results show that learning induces an autonomy of sensorimotor systems and that the release of cognitive control hubs in frontal and cingulate cortices predicts individual differences in the rate of learning on other days of practice. Our general statistical approach is applicable across other cognitive domains and provides a key to understanding time-resolved interactions between distributed neural circuits that enable task performance.

The research was supported by the John D. and Catherine T. MacArthur Foundation, Alfred P. Sloan Foundation, Army Research Laboratory, Institute for Translational Medicine and Therapeutics, National Science Foundation, National Institutes of Health and U.S. Army Research Office.

The study was conducted by Danielle Bassett, the Skirkanich Assistant Professor of Innovation in Penn's School of Engineering and Applied Science's departments of Bioengineering and of Electrical and Systems Engineering; Muzhi Yang, a graduate student in Penn Arts & Science's Applied Mathematics and Computational Science program; Nicholas Wymbs of the Human Brain Physiology and Stimulation Laboratory at Johns Hopkins; and Scott Grafton of the UCSB Brain Imaging Center.

Return to top of page