How does the brain switch between different ways of seeing?
Spontaneous flips in how we perceive ambiguous images are driven by bidirectional communication between visual and parietal brain regions, and the strength of bottom-up signals determines how long each view lasts.
Source
Brain activity dynamics in human parietal regions during spontaneous switches in bistable perception
What they did
The researchers scanned eighteen healthy participants using fMRI while they viewed a computer-generated spinning sphere. Participants reported whenever their subjective perception of the sphere's rotation direction flipped spontaneously, or changed because of actual depth cues added to the display. They then used a mathematical framework to test how information flowed between three specific right-hemisphere brain areas during these perceptual flips.
What they found
They discovered that a model featuring reciprocal communication between the visual motion area and both subregions of the parietal cortex best explained the brain activity during perceptual switches. When looking at individual differences, they found that stronger bottom-up signaling from the visual motion area to the posterior parietal region, and less inhibitory feedback from the posterior to the anterior parietal region, predicted a longer time holding onto a single interpretation of the image. This pattern of brain connectivity successfully accounted for a portion of the variance in how long participants stayed locked into a single perspective, yielding an R-squared value of 0.77.
The limits
What it doesn't show
First, because the analysis of individual differences relied on a small sample of participants after removing three outliers, these brain-behavior correlations should be replicated with a larger group. Second, while the model estimates directed connectivity, fMRI is fundamentally correlational, meaning we cannot prove that these specific pathways causally trigger the perceptual switch. Additionally, the study focused on specific parietal regions based on a prior hypothesis, meaning it cannot rule out the contribution of a broader brain network in driving these perceptual changes.
Key terms
- bistable perception
- A phenomenon where an ambiguous sensory stimulus continuously flips between alternative interpretations while the physical stimulus itself remains completely unchanged.
- dynamic causal modeling
- A statistical method used to analyze brain imaging data that estimates how active brain regions influence one another and how this communication changes under different experimental conditions.
- predictive coding
- A theory of brain function suggesting that the brain continuously generates models of the world to predict sensory inputs, updating these models based on the errors between predictions and actual sensory data.
- structure-from-motion
- A visual effect where a collection of moving flat dots on a screen creates the vivid illusion of a stereoscopic object in motion.
- blood oxygenation level dependent (BOLD)
- The primary signal measured by functional MRI, which reflects changes in blood oxygen levels that occur in response to localized neural activity.
- parietal lobule
- A region of the brain's parietal lobe involved in processing spatial awareness, attention, and sensory integration.
Flashcards
Want these cards to stick?
Save the deck to NoteFren and study it with spaced repetition.
Quiz yourself
What specific visual stimulus did the researchers use to induce spontaneous alternations between perceptual states?
Common questions
Why did the researchers include a control condition alongside the spontaneous condition?
The control condition used unambiguous physical depth cues to force the sphere to change direction. Comparing this with the spontaneous condition allowed the researchers to isolate brain activity unique to internal, self-generated perceptual flips rather than simple reactions to physical changes on the screen.
How does predictive coding explain the findings about dominance duration?
According to this theory, the anterior parietal region generates a prediction of what the sphere is doing, while the posterior region calculates the error between that prediction and actual visual data. The researchers think that weaker prediction error signals traveling up from the visual area mean the brain is satisfied with its current interpretation, resulting in a longer dominance duration.
Why were some participants' data removed from the final analysis?
The researchers excluded three participants from the individual differences regression because their data points had a Cook's distance value greater than one. In statistics, this means these individuals were extreme outliers whose unique data would have disproportionately warped the overall results of the mathematical model.
Did the brain areas analyzed in this study work in a simple top-down or bottom-up hierarchy?
No. While the researchers tested models with only unidirectional pathways, the winning model was bidirectional. This means the sensory and higher-level parietal areas continuously send signals back and forth to resolve the visual ambiguity, rather than a single region simply dictating to the other.
More on Perceptual decision