Photos by Chris Meyer Technician Jon Hobbs pulls out a drawer containing rat brain tissue used to study neuronal avalanches. Clay Haldeman, a senior undergraduate physics student, hooks up a machine used to study rat brain tissue for neuronal avalanches. IU physics graduate students (from left) Shaojie Wang, Clay Haldeman, technician Jon Hobbs and Aonan Tang meet with assistant professor John Beggs in the lab.

What’s the difference between a huge cascade of snow, ice and debris that could potentially bury you alive and that song you heard on the radio that’s been replaying itself mercilessly in your head?

At one level, not much.

Replace the snow, ice and debris with neural impulses, and you have a “brain avalanche,” an event scientists say may be the key to helping you store maddening ear worms, along with other memories. But the two are only alike in their underlying physics: in a brain avalanche, cells activate each other in stable patterns that scientists suspect may provide the right environment for memory storage, no icy destruction involved.  

Building on previous experimental studies, IU researchers recently designed a computer model that re-creates avalanche patterns found in rat brain slices in order to determine whether that busy state is best for socking away information. Although the model created was a very simple one, further investigation in this area could eventually lead to helping people with memory problems.

Animals often provide a place to start when looking at the brain, and IUB assistant physics professor John Beggs of the Biocomplexity Institute, one of two IU authors of a paper on the study, credits data collected from animals years ago with the inspiration for his work on brain avalanches. Studies conducted on birds, for example, have shown that when they sing, certain brain cells are activated in conjunction with certain syllables in the song.

“There’s a mapping between some behavior that an animal does and a sequence of brain cells going off—a pattern of activity,” said Beggs. “Now here’s the cool part: when the animal goes to sleep and starts to dream, it replays those patterns. In other words, it seems like the animal’s dreaming about what it did the previous day.”

Science has long told us that if some disturbance interrupts REM sleep, memory is affected when the subject wakes. “When you go to sleep, you rehearse these patterns that you experienced during the day, and that has something to do with storing them into long-term memory,” said Beggs.

The physicist’s recent history includes two studies demonstrating that slices of rat brain kept alive under the right conditions exhibit very similar types of patterns found in those animal studies, which makes them a good candidate for more exploration.

Beggs isn’t the only one who thinks the connection between patterns, rats and memory is cool. Clayton Haldeman, a senior physics major at IU, co-authored the modeling paper, which was published in Physical Review Letters on Feb. 7.

Haldeman and Beggs teamed up last summer to check out the data. “We saw that we had these avalanches, and we saw that we had these repeating patterns,” Beggs said. “The question is: what happens if you move the little network of the brain away from the state where it’s producing avalanches? How does that affect the number of patterns that you get?”

The scientists wrote a computer program that allowed them to mirror a neural system that they saw in the rat brain slices, then manipulate it. They found that when the system was in an avalanche state, a significantly greater number of stable patterns appeared. So, if the current ideas about the patterns are correct, it looks like the brain must be in this avalanche state for optimal information storage. In other words, more patterns may equal better memory.

“When you think of an avalanche, you think of something wiping everything away. But when you’re at this point, you’re actually producing repeatable activity,” said Beggs.

In the lab, researchers found they could ease a network into an avalanche state by applying certain neurochemicals. This aspect of the study smacks of possibility–could we someday apply these same chemicals to the brains of people with memory problems? How about college students who don’t get enough sleep before an exam?

The scientists agreed that the most exciting part of the study was seeing how well the data from the computer model matched up with data taken right from the rat brain slices, not any “eureka” discovery that might revolutionize modern medicine. Beggs emphasized the basic nature of the research, warning against reading too much into data too soon, but still managed to seem optimistic that what’s science fiction now might some day become fact.

“If the brain is at this avalanche point, and if that really does create a condition where the largest number of stable patterns are stored, and the stable patterns really are important for memory, and this really does apply to the human brain,” he said, “Then yeah, what you’d want to do is be in that stable state.”

Scientists may be years away from being able to get “Mmm-Bop”-esque songs out of your head, but next time you’re lying awake at night, tortured by a memory that won’t go away, you can thank your brain and its stable states for putting you there.

Mind and brain lecture Barbara Landau of Johns Hopkins University is the next speaker in the IUB Department of Speech and Hearing Sciences Language Acquisition Colloquium Series Monday, April 18, at 4 p.m. at the School of Fine Arts 015.

The topic of the Horizons of Knowledge Lecture will be “Specialization and breakdown of spatial representations: Mind, brain and development.”

“Our experience of the spatial world is a unitary one—we perceive objects and layouts, we remember them and act on them, and we can even talk about them with ease,” Landau wrote in a synopsis of her presentation. “Despite this impression of seamlessness, spatial representations in human adults appear to be specialized in domain-dependent manner, engaging different properties and computational mechanisms for different functions. In this talk, I will present evidence that this domain-specific specialization emerges early in development and is reflected in patterns of breakdown that occur under genetic defect.

“I will offer evidence from Williams syndrome—a rare genetic syndrome that gives rise to an unusual profile of severely impaired spatial representation together with spared language.”