Recent research published in nature communications We provide evidence that an evolutionarily ancient group of brainstem neurons is required for the brain to filter out distractions and focus on important spatial information. This finding indicates that these particular inhibitory cells are specialized in helping the animal select the correct target of attention, without affecting basic perception or body movements. The discovery of these neurons in mice could be the first step toward developing targeted therapies for attention disorders.
To navigate complex environments, animals must constantly filter sensory input to select the most important information. The importance of a stimulus, known as priority, is a combination of two factors. One factor is physical salience. This is a bottom-up signal driven by how conspicuous an object is, like a bright flash of light. Another factor is behavioral relevance. This is a top-down signal driven by the animal’s current goal, such as searching for a specific shape that leads to a food reward.
Historically, the dominant view in neuroscience proposed that high-level spatial attention is primarily managed by sophisticated networks in the prefrontal cortex, a highly developed region in humans and other primates. However, animals with less developed external brains, such as birds, fish, and rodents, still show an excellent ability to focus attention. This observation suggests that ancient deep brain structures may be responsible for this basic cognitive skill across a variety of vertebrate species.
“If you actually go back in evolution, over hundreds of millions of years, birds have had this ability, fish have had this ability, and they typically don’t have highly developed prefrontal cortex. So how does the brain solve this problem?” said lead author Ninad Kothari, a postdoctoral fellow in the Johns Hopkins Department of Psychological and Brain Sciences. “We were able to identify an evolutionarily ancient region of the brainstem that enables this ability.”
The inspiration to study these neurons in mammals stems from previous work in birds, frogs, and turtles conducted by Johns Hopkins neuroscientist Shreesh Mysore and other scientists studying neural circuits related to behavior. Previous studies have shown that a midbrain region called the superior colliculus is involved in processing spatial information. Because the superior colliculus is a major center for both sensation and movement, destroying the superior colliculus tends to impair basic vision and body coordination.
This makes it difficult to determine whether the superior colliculus itself is calculating attention, or whether another specialized brain module is handling distracting competition filtering. The authors of the new study focused on an ancient group of brain cells called the parabigeminolateral tegmental inhibitory complex (PLTi). These particular brainstem neurons produce a chemical messenger called GABA that tends to reduce the electrical activity of other nearby neurons.
The researchers first mapped the anatomical connections of PLTi neurons in adult mice. Using fluorescent tracers, they discovered that these cells receive highly organized input from the superior colliculus and send long-range projections directly back to the superior colliculus. The authors demonstrated that activation of PLTi neurons directly inhibits the superior colliculus by using a technique called chemical genetics, which allows scientists to selectively activate or silence specific cells using custom drugs.
To test spatial attention, the researchers trained freely moving mice in a touchscreen test known as the flanker task. The mouse had to interact with the screen through a custom mask with three holes. They learned how to touch the screen with their nose to report the orientation of the central target image, such as vertical or horizontal stripes.
At the same time, distracting peripheral images called flankers were displayed on the screen, competing for the animal’s attention. Flankers can be matched, meaning they share the direction of the target, or mismatched, meaning they point in the opposite direction. The authors systematically varied the visual contrast of the flankers to alter their physical saliency.
The researchers then used chemical genetics to bilaterally silence PLTi neurons in six genetically modified mice. The researchers administered a specific drug that turned off the function of these targeted brain cells while the mice performed a flanker task. When PLTi neurons were inactivated, mice performed significantly worse on incongruent trials, indicating a significant increase in distraction. “When you inactivate these neurons, the mice become very distracted,” Kothari says.
“The hallmark of ADHD is that even subtle distractions distract you, and that’s exactly what we see here when these neurons are silenced,” Mysore said. “But the next day, when the neurons are turned on again, the same animal is able to ignore the distraction again, even if it’s very strong.”
Remarkably, the mouse still performed accurately when the flankers matched the target, or even when the target appeared alone. To see whether PLTi neurons simply responded to raw visual intensity, the researchers replaced the task-relevant flankers with simple blocks of light that contained no task information. Under these conditions, silencing PLTi neurons had no negative effect on animal performance. This provides evidence that PLTi neurons evaluate both physical intensity and goal-directed relevance to guide behavior.
Because attention is closely tied to basic sensory processing and movement, the researchers wanted to see whether silencing PLTi neurons simply impaired the animals’ ability to see and move properly. They analyzed data from four mice that performed a basic single-target vision test and found no changes in vision. The authors used a calibrated 3D video camera to track the mouse’s physical head movements and selection of upper and lower response ports.
The physical movement trajectory and motor selection did not change at all. The only physical difference was that mice responded slightly faster across all task conditions when PLTi neurons were silenced.
“The only thing that was compromised was the ability to retrieve competing information, compare it, and pay attention to where the most important information was,” Mysore said. “This part of the brain is like an attentional selection engine. It helps us answer the question, ‘What is the most important information to pay attention to right now?'”
Using mathematical models and brain recordings, the researchers determined that this faster reaction time was due to global hyperactivity of the superior colliculus without constant sedative inhibition by PLTi neurons. This supports the idea that PLTi acts as a specialized module specifically designed to remove distractions.
In natural environments, the brain must make rigorous choices about which stimuli are most important. This process relies on subjective decision boundaries. This is the exact point at which the distraction suddenly disables the target. The authors analyzed the accuracy of this boundary in mice and found that healthy animals exhibit sharp, highly precise transition points that act like a winner-take-all filter.
When PLTi neurons were silenced, this decision boundary changed significantly, allowing much weaker distractors to inappropriately capture the animal’s attention. The boundaries are also wider and less accurate. This means that the choice of animal is less obvious.
To understand the underlying brain mechanisms, the scientists directly recorded electrical activity from 16 individual neurons in the superior colliculus of head-fixed mice. They presented competing visual stimuli of magnifying black dots on a screen and observed how neurons processed the competing information. Normal superior colliculus neurons showed sharp and precise changes in firing rate when distracting stimuli outweighed central targets.
When the researchers silenced the PLTi neurons during these recordings, the neural signals in the superior colliculus lost their sharp precision. Neural boundaries changed in exactly the same way as behavioral boundaries, with weaker distractors significantly suppressing the main target signal. This suggests that PLTi orchestrates competitive interactions within the superior colliculus to generate precise signals for selective spatial attention.
Although this study provides detailed evidence for the role of PLTi in attention, there are potential limitations and areas for future exploration. The authors note that even when PLTi is completely silenced, mice can still perform slightly above random chance. This suggests that other brain regions likely contribute to the comparison of competing stimuli even when operating at lower resolution without PLTi.
Another potential misconception is assuming that PLTi acts as a passive relay station rather than an active computing center. The detailed structural logic of how mammalian PLTi and superior colliculus circuits communicate in a stepwise manner remains unclear. Future studies will need to map these precise local connections to fully understand the underlying neural wiring.
Moreover, scientists still do not know how this deep brainstem network interacts with the highly evolved networks in the cortex. Building a complete picture of sensory processing requires understanding how these different brain circuits work together to drive spatial attention. Addressing these questions may help explain attention deficits observed in atypical cognitive conditions such as schizophrenia, autism, and ADHD.
The researchers plan to investigate the extent to which these particular neurons are involved in human attention. If their function is affected in neurodivergent states, it may lead to the development of novel targeted drugs.
“All evidence so far suggests that these neurons also exist in humans,” Mysore said. “But are they involved in selective spatial attention in humans? An interesting hypothesis is that they play an important role.”
The study, “Control of selective spatial attention requires evolutionarily ancient brainstem neurons,” was authored by Ninad B. Kothari, Arunima Banerjee, Qingcheng Zhang, Wen-Kai You, and Shreesh P. Mysore.
