Suppression shows up in space, time, and feature space
The most direct evidence comes from motion perception. When monkeys paid attention to a motion direction that matched a neuron's preferred direction, the neuron's response dropped sharply for directions 60–90 degrees away from that preferred direction—a clear dip in the tuning curve that researchers call an attentional suppressive surround [3]. In other words, attending to one direction didn't just enhance that direction; it actively quieted responses to neighboring directions.
The same pattern appears in human behavior. In a motion repulsion task, feature-based attention reduced motion repulsion near the attended direction, consistent with suppression of nearby feature values [2][3]. And suppression isn't limited to space or features: when people attended to a specific moment in a sequence of letters, accuracy dropped for targets appearing just after the attended time compared with the same time point under neutral conditions—suggesting a temporal suppressive surround as well [1].
How far does the suppressive surround reach?
In feature space, the inhibitory zone is narrow. One study describes attention as eliciting a narrow inhibitory surround in feature space, meaning suppression is strongest for features very similar to the attended one and falls off quickly [4]. The motion-direction work puts a number on this: the deepest suppression occurred 60–90 degrees away from the attended direction in monkeys [3].
In time, suppression appeared one or two frames (100–200 milliseconds) after the attended moment [1]. In space, the reach of suppression depends on how widely attention is spread. When participants attended only to the center of a motion stimulus, surround suppression was weaker; when they attended to both center and surround, suppression grew stronger, and the effect scaled with the size of the surround at low contrast [5].
Why would attention suppress its neighbors?
The leading explanation is that attention works like a spotlight with a dimming halo. The Selective Tuning model proposes that selecting an attended element includes suppressing the surrounding network, which sharpens the selected signal and reduces interference from nearby distractors [1][3]. This is modeled as a balance between narrowly tuned excitatory inputs and more broadly tuned inhibitory inputs, with attention mainly boosting the gain of the inhibitory side [2][3].
That architecture would explain why suppression is nonlinear—strongest at intermediate distances from the attended feature rather than simply falling off with distance—and why wider attention produces stronger suppression: a broader attentional focus recruits more of the inhibitory surround [3][5]. The model also suggests a possible function: precise timing and sharp discrimination in dynamic visual tasks may depend on this kind of surround suppression [1].
When does the suppressive surround get stronger or weaker?
Stimulus conditions matter. In the spatial attention study, the effect of attention on surround suppression grew with surround size when the stimulus had low contrast, but stayed flat at high contrast [5]. Suppression was also weaker when the center and surround gratings drifted in opposite directions [5], suggesting the surround is tuned to features similar to the attended one.
The breadth of attention is another lever: wider attention conditions produced stronger surround suppression than narrow ones [5]. And the effect isn't purely a laboratory curiosity—researchers suggest that reduced surround suppression seen in conditions such as schizophrenia and autism spectrum disorder could potentially stem from abnormal attention mechanisms, though that link remains a hypothesis rather than a demonstrated cause [5].
Sources used in this answer
Does attention to a point in time lead to temporal surround suppression?
Attending to a specific moment in a letter sequence improved accuracy at that time but reduced accuracy for targets appearing one or two frames later, indicating a temporal suppressive surround after the attended time.
Feature-based attention induces surround suppression during the perception of visual motion
Feature-based attention to a motion direction suppressed neuronal responses to nearby directions in monkey MT/MST and reduced motion repulsion near the attended direction in humans, consistent with a suppressive surround in feature space.
Attention to visual motion suppresses neuronal and behavioral sensitivity in nearby feature space
When monkeys attended to a neuron's preferred motion direction, tuning curves showed minimum responses 60–90 degrees away, an attentional suppressive surround modeled as stronger gain on broadly tuned inhibitory inputs, with a corresponding nonlinear effect on human motion discrimination.
Feature-based attention elicits surround suppression in feature space
Feature-selective attention elicits a narrow inhibitory surround in feature space, suppressing processing of features similar to the attended one.
Increasing the spatial extent of attention strengthens surround suppression
Wider spatial attention produced stronger surround suppression in human motion discrimination, with the effect scaling with surround size at low but not high contrast, and weaker when center and surround drifted in opposite directions.