Subscribe to eTOC

The Thalamus, Not the Sensory Cortices, Regulates Selective Attention, Optogenetic Experiments Show



THE RESEARCHERS STIMULATED the visual thalamus (lateral geniculate nucleus, or LGN), recording spiking in the LGN in response to well-controlled, head-mounted visual stimuli. They found that when the animal attends to the light, the same visual stimulation evokes more LGN spikes than when the animal is ignoring vision.

Using optogenetic techniques, researchers were able to show that the thalamus plays a central role in selective attention, which was previously thought to be an exclusively cortical phenomenon.

In a discovery that experts say will require a rewriting of the textbooks, researchers have shown that the thalamus plays a central role in selective attention, which was previously thought to be an exclusively cortical phenomenon.

The finding that the prefrontal cortex relies on the thalamus, not the sensory cortices, to modulate the strength of competing sensory inputs may have implications for developing treatments for autism, schizophrenia, and attention deficit disorder, conditions in which sensory overload is thought to contribute significantly to disability.

The standard model relegates thalamic function to serving as a passive relay point for incoming sensory signals. According to this model, cognition, including the sensory selection at the root of attention, is strictly a cortical phenomenon. To put that model to a rigorous test, Michael Halassa, MD, PhD, an assistant professor of psychiatry, neuroscience, and physiology at New York University and the lead author of the study published in the October 29 issue of Nature, turned to optogenetics.


DR. MICHAEL HALASSA said his experiments confirmed that prefrontal cortex activity influences thalamic sensory processing to increase attention to the relevant stimulus and decrease attention to the irrelevant one.


CHANGES IN SENSITIVITY (gain) of the visual system (A) when attention is directed towards vision (top) versus hearing (bottom). Cross-modal task design (B): Context is provided by binaurally emitted blue or brown noise, informing the mouse which stimulus is the appropriate target to attend to; when the conflicting visual and auditory target stimuli are presented, the mouse has to choose based on context. Mice perform similarly on visual and auditory stimuli (C). Psychometric performance curves on many experiments (D-G), collectively showing that animals split their attention between vision and audition in the cross-modal task, and that this division of attention is a top-down process.

Optogenetics takes advantage of light-sensitive ion channels isolated from marine organisms to explore the function of individual neurons and entire circuits. In a typical experiment, mice are genetically engineered to express the ion channel in neurons in the brain. An optical fiber is implanted into the brain region of interest and used to deliver a light pulse to the neurons expressing the channels. The light causes the channels to open, allowing an influx of ions that hyperpolarize the neurons, suppressing their firing.

The ability to temporarily suppress specific brain regions gives researchers the ability to dissect brain circuits with unprecedented spatial and temporal precision. “There would have been no way for us to do this study without optogenetics,” Dr. Halassa said.


In his experiments, Dr. Halassa expressed the channel protein throughout the mouse brain, then placed the optical fibers into either the prefrontal cortex (PFC), the visual cortex, the auditory cortex, or the visual thalamus (lateral geniculate nucleus, or LGN).

Since the focus of the study was selective attention, Dr. Halassa designed an experiment in which a mouse had to choose whether to pay attention to either a visual stimulus or an auditory stimulus when presented with both in order to get a reward. The mouse first heard either a low tone or a high tone, which signaled not only that a trial was about to commence, but also which type of stimulus would be relevant in that trial. Then the mouse was presented with both the visual and auditory stimuli. In a visual trial, a light on one side of the cage signaled that a reward was located there, while the sound signaled nothing. Conversely, in an auditory trial, it was the nature of the sound, not the position of the light, that revealed the position of the reward.

The pitch of the first tone “is telling the animal which of the upcoming sensory stimuli it should pay attention to,” Dr. Halassa explained.

The standard model of attention circuitry gives the most prominent role to the PFC. Indeed, when Dr. Halassa used optogenetics to suppress activity in the PFC after presentation of the first tone but before the reward-indicating stimulus, the ability of the mice to then selectively attend to the relevant stimulus was impaired, increasing their error rate by about 60 percent. But during that same critical period — after the animal had heard the tone and was anticipating the stimulus — suppressing activity in the visual or auditory cortices had no effect on the error rate. In contrast, suppressing thalamic activity caused almost as many errors as suppressing the prefrontal cortex.

These data do not support a causal role for the PFC's interactions with the visual or auditory cortex in this divided attention task, Dr. Halassa said. Instead, “they support a model in which PFC activity influences thalamic sensory processing” to increase attention to the relevant stimulus and decrease attention to the irrelevant one, he said.

Dr. Halassa then confirmed the new model in a second experiment where he directly measured the flow of chloride ions into the thalamic neurons using a chloride-sensitive fluorescent dye, whose intensity he could detect using fiber-optic recording. GABAergic neurons synapsing onto the visual thalamus increase chloride flow, hyperpolarizing them and suppressing their firing. In a mouse expecting an auditory cue, there should be an increase in suppression in the visual thalamus, and thus an increased flow of chloride ions there, Dr. Halassa explained. “And that is what we saw.”

The selective inhibition of sensory input provided by the thalamus is essential for allowing the brain to concentrate on what's important. “The normal mode of operation for a healthy brain is to pretty much suppress everything except the information you care about,” Dr. Halassa said. “I think one of the most exciting discoveries over the past few years has been that neurodevelopmental disorders such as autism, schizophrenia, and attention deficit hyperactivity disorder share quite a bit of genetic architecture. All these diseases have common molecular deficits.”

Those deficits, evidence suggests, lead to phenotypic overlap in which sensory overload is a prominent and highly disabling feature, contributing to learning impairments and social withdrawal.


“Our ideas about the thalamus have [now] been transformed, since it was thought to be just a passive relay,” said Earl Miller, PhD, a professor of neuroscience at the Massachusetts Institute of Technology and an expert on the neural circuits underlying attention.

“There has been a lot of circumstantial evidence that ‘top-down’ attention selection depends on neurons in the prefrontal cortex, and this paper provides direct demonstration of that. But we have assumed the entire process was taking place in the cortex. Dr. Halassa and colleagues showed that the signal by which the prefrontal cortex can cause attentional modulation of sensory cortex depends on the thalamus.”

“The conceptual advance described here is profound,” said Sabine Kastner, MD, PhD, a professor of psychology at Princeton University whose work has revealed the importance of another thalamic structure, the pulvinar, in temporally aligning the computational efforts of disparate cortical regions.

“That the prefrontal cortex controls the gain, or signal amplification, through its interaction with the thalamus, is the key data point,” she said, “and [it is] a paradigm shift for what the prefrontal cortex is doing. That is a really radical divergence from what is in all textbooks at this time.”

The elucidation of the essential role of the thalamus in attention lays the groundwork for targeted therapy, she said. “You can envision depth electrodes providing electrical stimulation to help regulate the gain” in thalamic structures. “I don't think we are far enough along in our understanding, but this is a great beginning. If you'd asked me 10 years ago, I would have thought this was science fiction, but not anymore. It may be a meaningful way to help people with severe attentional deficit disorders. That's not out of our reach.”


•. Wimmer RD, Schmitt LI, Davidson TJ, et al. Thalamic control of sensory selection in divided attention Nature 2015;526(7575):705–709.
    •. Saalmann YB, Pinsk MA, Wang L, et al. The pulvinar regulates information transmission between cortical areas based on attention demands Science 2012;337(6095):753–756.