Guest Blog

Guest Blog

Commentary invited by editors of Scientific American

How the Brain Does "Attention" Is Still Unknown


Humans can focus on one thing amidst many. “Searchlight of attention” is the metaphor. You recall a childhood friend’s face one moment, then perhaps the dog you loved back then, and then…what you will. Your son’s face on stage rivets your attention; the rest of the cast is unseen.

No “ghost” in the brain aims that searchlight. What does? Neurons do, somehow, but how is a mystery that new research actually deepened.

The experiment used monkeys. They can focus attention like people do. They can zero in on a red square on a screen full of distractions, for instance. When the square moves, a trained monkey will press a button. Electrodes inserted in a monkey neuron will reveal “firing” (minuscule electrical ripples) simultaneous with attention. This may locate brain areas by which the monkey watched that red square.

It’s not only the explosive firing in neurons that instruments detect. They also spot the milder priming to fire, when the monkey expects (from training) that neurons are about to be stimulated. Neurons in a one area of the cortex fire when an object moves (but not, for instance, if it gets brighter but stays still.) If a monkey learns that an onscreen cue (a blip of light) signals that the red square is about to move, the cue alone primes the motion-sensing neurons. They also synchronize more tightly (i.e. reduce random noise among them.) Cues cock neurons, like a gun. It’s like Pavlov’s dogs salivating at the bell that preceded feeding.

Potassium-gated ion channel: a valve of intricate proteins, embedded in neuronal membranes. A potassium ion (purple) is in the center. The searchlight of attention opened these valves when the volunteer remembered The Simpson’s. The neuronal “firing” that the electrode detected ensued.

Potassium-gated ion channel: a valve of intricate proteins, embedded in neuronal membranes. A potassium ion (purple) is in the center. The searchlight of attention opened these valves when the volunteer remembered The Simpson’s. The neuronal “firing” that the electrode detected ensued.

Because of this Pavlov effect, experiments can inspect the sort of attention that follows overt cues. That’s what today’s experiment did. Earlier work had identified three brain areas that seemed to work in sequence to produce visual attention. Today’s post will simply call them A, B, and C, instead of their cumbersome anatomical names. (“A”, for instance, substitutes for “superior colliculus”).

The machinery of attention seemed to be that A would fire, which primed B and C, which then fired when the stimulus – the red square moving -- occurred. But no, it has turned out. When an experiment inactivated A temporarily (with drugs) in two monkeys, the result was a surprise: yes, the monkey’s visual attention was crippled, but neurons in B and C still became primed (and less noisy too). Those parts of the attention machine still seemed to function as before.

What to make of this? That unknown brain areas besides A, B, and C are needed for monkey visual attention after a cue. Research will continue.

But cued attention isn’t the only sort of attention, of course. In fact, most choices in humans seem to have no cues at all. You choose to remember the face of your childhood friend but not, say, Abe Lincoln’s. You could remember Lincoln’s, but you choose not to. That sort of attention isn’t examined with monkeys and onscreen cues.

But it has been, in human volunteers (epileptics about to undergo brain surgery), and nothing so simple as an A, B, C sequence was present. One man agreed to view 5-10 second video clips of famous people and scenes, one after another. There were 48 of them. (Tom Cruise, Baywatch, the Pyramids…) Electrodes in his brain recorded which neurons fired when he saw clip 1, clip 2 and so on. Then he was asked to choose to remember one of them. A neuron fired that had fired when a cartoon clip of The Simpsons had played on the screen. Was it The Simpsons that, silently and unprompted, he had chosen to recall? Yes, it was, he told the researchers.

What in his brain had triggered firing in that neuron? Its membrane’s electrical potential had changed. Ion channels had opened. An electrical rippled had flowed to the electrode. What was the cause?

No one knows. The searchlight is one of neuroscience’s great mysteries.

What phenomenon in nature?

Monkeys and people can focus attention on one object and ignore distractions.

What did this discovery show?

That the brain sequences that “handle” attention in response to a cue must involve brain areas besides those previously thought sufficient.

What was known before?

What was thought before was that three brain areas in sequence were sufficient to produce visual attention in monkeys.

What remains unknown?

First, what other brain areas handle visual attention that is triggered by an overt cue. Second, when there’s no cue, what aims the “searchlight of attention” that humans seem to control freely?


A. Zenon and R. Krauzlis, Attention deficits without cortical neuronal deficits, 489 Nature 434-437, 20 September 2012; A. Smolyanskaya and R. Born, Attention is more than meets the eye, 489 Nature 371-371, 20 September 2012.

The Simpson’s experiment is described in: H. Gelbard-Sagiv, et al. Internally Generated Reactivation of Single Neurons in Human Hippocampus during Free Recall, 322 Science 96-101 (October 3, 2008)

Image: Potassium channel by Hannes Röst at Wikimedia Commons.

The views expressed are those of the author and are not necessarily those of Scientific American.

Share this Article:


You must sign in or register as a member to submit a comment.

Starting Thanksgiving

Enter code: HOLIDAY 2015
at checkout

Get 20% off now! >


Email this Article