Summary: A new UC Davis study shows that the brain’s attention system prepares in two rapid stages: it first establishes a broad focus on a general feature, then quickly narrows to the specific attribute. Using EEG combined with machine-learning decoding, researchers measured how people prepared to attend to either the color or the motion direction of moving dots before those stimuli appeared.
The team found that category-level attention (color versus motion) was detectable in brain activity in about 240 milliseconds, while attention to a specific attribute (blue versus green or up versus down) emerged around 400 milliseconds. This hierarchical timing sheds light on how attention is organized in the human brain and offers a framework for studying attention differences in clinical populations.
Key Facts
- Rapid hierarchy: The brain first biases broad feature categories, then refines attention to precise attributes within fractions of a second.
- Methods: EEG alpha-band recordings plus machine-learning decoding separated dimension-level and attribute-level attention signals.
- Clinical implications: The timing of these attention stages may help explain or identify delays in disorders such as ADHD or autism.
Source: UC Davis
How we set up attention before a sensory event affects how we perceive it. For example, expecting something moving in the sky prompts a different preparatory focus if you anticipate a bird gliding past versus a baseball coming toward you. The new research asks whether the brain prepares first for a broad feature, like motion versus color, and only afterward for the specific attribute, such as motion direction or a particular color.
Researchers at the UC Davis Center for Mind and Brain recorded electrical brain activity while volunteers prepared to view colored dots moving on a screen. By applying multivariate machine-learning decoding to the EEG alpha-band signals, the team could distinguish preparatory neural patterns that reflected either broad feature categories (color vs. motion) or the precise attributes of those features (blue vs. green; up vs. down).
The study was conducted in 2024 with 25 participants aged 19 to 39. On each trial, participants received a cue instructing them to attend either to a color (blue or green) or to a motion direction (up or down), then waited through a blank cue-target interval before the moving dots appeared. The researchers used EEG and eye tracking to ensure that the anticipatory signals reflected internal attention rather than overt eye movements.
Lead author Sreenivasan Meyyappan explained that directing attention to one feature dimension suppresses processing of the other: when participants prepared to attend color, neural markers related to motion were reduced, and vice versa. The machine-learning analysis separated these overlapping signals and revealed their distinct time courses.
On average, dimension-level attention (deciding to focus on color versus motion) emerged about 240 milliseconds into the cue-target interval. Attribute-level attention, the neural signature for selecting a particular color or a specific motion direction, became evident later — around 400 milliseconds. The pattern supports a hierarchical model in which top-down control first biases broad channels of processing and then sharpens toward the relevant detail.
George R. Mangun, Distinguished Professor of Psychology and Neurology and co-director of the UC Davis Center for Mind and Brain, summarized the finding: the brain narrows its attentional focus over time, shifting from a wide preparatory bias to a targeted selection of task-relevant features. He compared the process to a pilot heading first toward a continent and then zeroing in on a particular city.
Understanding this temporal hierarchy has potential clinical value. If people with attention disorders show slower or disrupted narrowing from dimension-level to attribute-level attention, that could help explain certain perceptual or behavioral symptoms and point toward tailored interventions.
Mingzhou Ding, Distinguished Professor of Biomedical Engineering at the University of Florida, is an additional co-author on the study. Funding for the research was provided by the National Institutes of Health and the National Science Foundation.
About this attention and visual neuroscience research news
Author: Karen Nikos
Source: UC Davis
Contact: Karen Nikos – UC Davis
Image: Image credited to Neuroscience News
Original Research: Closed access. “Hierarchical Organization of Human Visual Feature Attention Control” by George R. Mangun et al., Journal of Neuroscience.
Abstract
Hierarchical Organization of Human Visual Feature Attention Control
Anticipatory feature-based attention allows the brain to prepare in advance for visual properties such as color or motion direction. Top-down control signals from frontoparietal networks bias sensory cortex activity to enhance relevant information and suppress distractors. This mechanism supports selecting stimuli by a particular feature while ignoring competing attributes.
Anticipatory attention can operate at different levels of specificity. It may initially bias broad feature dimensions (for example, color versus motion) and then narrow to select specific attributes within that dimension (for example, blue versus green). To test this sequence, EEG was recorded while participants were cued on a trial-by-trial basis to attend either to motion direction (up or down) or to color (blue or green).
Multivariate decoding applied to EEG alpha-band (8–12 Hz) activity during the cue-target interval showed reliable decoding for both the attended dimension and the specific attribute. Critically, decoding for the dimension-level bias (motion vs. color) consistently emerged earlier than decoding for the attribute-level bias (up vs. down or blue vs. green). These results indicate that top-down control of feature-based attention unfolds hierarchically: the brain first biases a broad feature channel and then refines processing to the precise attribute.