Hopkins study finds brain cells that orient us in relation to gravity

Neural pathway makes us "intuitive physicists," Johns Hopkins study finds

You're playing tennis with a friend, volleying back and forth, when he suddenly smacks a rocket to your backhand. You lunge, extend your racket and hit a good return.

You've almost certainly moved your head, body and line of sight in ways that radically changed their orientation to the ground. But you never forgot which way was up and which was down, and you knew where the ball would be at the optimal swinging moment.

How did your brain and body perform such a complex feat of perceptual adjustment?

The calculation takes place in a portion of the brain scientists once thought was responsible for something completely different, and the location of the cells in question could help explain what makes human begins so good at tracking and interpreting motion while also remaining aligned and balanced.

So says a team of neuroscience researchers at the Zanvyl Krieger Mind/Brain Institute at the Johns Hopkins University.

The group's most recent findings, published last month in the journal Current Biology, suggest that a set of specialized cells in what's known as the ventral visual pathway — part of the brain believed to help us identify objects — also aligns our perception of those objects with signals about the direction of gravity.

"Our results show how the brain deduces the direction of gravity from visual cues, providing us with critical information about object physics — as well as additional cues for maintaining posture and balance," said Charles E. Connor, a neuroscience professor at Johns Hopkins and the study's senior author.

This multifaceted skill is so complicated, Connor said, that neuroscientific researchers are only now developing the mathematical models that can help them measure it.

The ventral visual pathway — a narrow channel in the lower brain also called the ventral stream that carries visual information from the occipital lobe into the temporal lobe — has long been recognized as the area that handles "object vision," allowing human beings to identify and categorize visual objects based on their physical properties.

That's why scientists nicknamed it "the what stream" — the one that determines what an object is, not how it moves or what it does.

Over the past decade or so, researchers have learned more about this function, showing, for example, that certain parts of the ventral stream are sensitive to specific categories of objects (faces or body parts, for example) and other parts respond to specific properties (shape, color or texture).

Connor, the director of the Zanvyl Krieger Institute, and Siavash Vaziri, a former Johns Hopkins postdoctoral fellow, began building on that knowledge two years ago when they identified cells in the ventral streams of rhesus monkeys that responded not just to small, simple objects but also to large-scale, three-dimensional shapes.

Scientists have long known that the brains of most primates, including rhesus monkeys, are remarkably similar to those of humans in the way they're structured to process visual cues.

The findings appeared in the neuroscientific journal Neuron in 2014.

Their more recent study took a further step. They showed the same monkeys a carefully developed range of such large-scale shapes, monitoring the electrical responses within the relevant brain cells.

Connor and Vaziri kept an eye on which shapes sparked the strongest reactions, searching for patterns in the data. The ones they found told an interesting tale.

The cells, or neurons, responded to a variety of shapes, but they generated especially strong current during exposure to large horizontal planes, to sharp edges that extended in space, and to lines that were vertical, or at least within 20 to 30 degrees of verticality.

What do such shapes have in common?

They're all useful, the team said, in defining our physical orientation with respect to the world — a place our brains usually perceive as having a horizontal bottom plane (the floor or the ground) and strong verticals (trees, cliffs, walls, the direction of gravity).

"We've thought long and hard about what that kind of information would provide to the brain," Connor said. "The most obvious is that it provides information about gravity, which all these things are aligned with, and conversely, about the overall orientation of the world relative to the head."

These cells, in other words, may well be designed to help us know which way is up, even when we're tilted, tumbling or in motion.

The findings suggest that the perception of spatial properties — of objects as they move in space — is, in fact, a function of the ventral stream, not the nearby (and parallel) dorsal stream, as was long believed.

Scientists have long called the dorsal stream "the how stream" or "the where stream" because they believed it was where the brain recognized and processed the location of objects in space. The dorsal stream moves visual information into the upper brain's parietal lobe.

Robert W. Kentridge, a neuroscientist at Durham University in Britain, said the Hopkins study shows things in the temporal lobe aren't what they seemed.

"It has often been assumed that spatial properties are not involved in object identification — after all, a chair is still a chair if you move it from one place to another — so spatial information must be handled in the dorsal stream, not the ventral stream," he wrote in an email to The Baltimore Sun. "The interesting thing about Connor's gravity result is that it is an example of coding spatial properties in the ventral stream."

Some of the hardest work for the study began several years ago, when Vaziri, a biomedical engineer who now works at the National Institutes of Health, started creating the software to generate the large-scale images to show the monkeys.

They not only had to be "convincing in 3-D way," Connor said, but Vaziri had to create subtle variations from shape to shape, creating a smooth continuum "all the way to landscapes," providing "an infinite variety of shapes that would happen in the natural world."

That provided more than enough sample shapes to guarantee statistically significant results.

The team then spent two years showing them to the monkeys and monitoring their neural responses via electrodes.

The results, Connor said, offer new clarity on what allows human beings to be such "intuitive physicists," not only when we play sports but in ways large and small throughout the day, and without even knowing we're doing it.

It's amazing we can do it at all, and Connor said it's still exciting to be learning how it works.

"This is easily one of the more mysterious, complex things the brain does," he said. "We're good at it because, unbeknownst to us, the dorsal and ventral pathways are processing information in computationally intense ways we still don't understand. But we're working on it."


Copyright © 2018, The Baltimore Sun, a Baltimore Sun Media Group publication | Place an Ad