Abstract
Objects that move independently in the world alter the optic flow field and can induce errors in perceiving the direction of self-motion (heading). We show that adding vestibular (inertial) self-motion signals to optic flow almost completely eliminates the errors in perceived heading induced by an independently moving object. Furthermore, this increased accuracy occurs without a substantial loss in the precision. Our results thus demonstrate that vestibular signals play a critical role in dissociating self-motion from object motion.
Introduction
Navigating a busy world requires the brain to distinguish between our own movement and the movement of objects around us. When moving through an environment, stationary objects create a pattern of visual motion on the retina known as optic flow, which the brain uses to calculate "heading"—the direction of self-motion. However, if an object in the field of view moves independently (e.g., a passing car), it distorts the flow field. If the visual system cannot parse the object's motion from the global flow, it results in a heading bias. This study investigates whether the vestibular system (the inner ear's motion sensors) provides the "ground truth" necessary to dissociate self-motion from object motion.
Methods
The researchers trained rhesus macaque monkeys to perform a heading discrimination task while seated on a 6-degree-of-freedom motion platform. In each trial, monkeys experienced a 1-second movement and reported if their heading was slightly left or right of "straight ahead."
The experiment compared three conditions:
- Vestibular: Physical motion in the dark.
- Visual: Simulated movement via a 3D "star cloud" on a projection screen.
- Combined: Simultaneous physical and simulated visual motion.
In many trials, a spherical "distractor" object moved independently across the screen. The researchers also varied the "coherence" (signal-to-noise ratio) of the visual stars to test how the brain weighs different cues based on their reliability.
Theoretical Framework: Vector Summation
The motion of an object on the retina is a vector sum of its actual world motion and the observer's self-motion. To perceive heading accurately, the brain must solve an inverse problem to recover the self-motion vector from this combined signal.
The authors suggest this behavior is driven by specific neurons in brain area MSTd. "Congruent" neurons prefer the same direction for visual and vestibular cues, while "opposite" neurons prefer conflicting directions. Computational models suggest these "opposite" cells are key to filtering out the "nuisance" of object motion.
Experimental Results
The most striking finding is that while an independently moving object (IMO) creates significant errors in single-cue conditions, the brain uses the combination of senses to cancel out the error.
- Bias Reduction: In the visual-only condition, a moving object induced heading biases of ~5° to 6°. In the combined multisensory condition, this bias was reduced to nearly zero (often <1°).
- Precision Gains: Adding a moving object actually improved the precision (lowered thresholds) of heading judgments. This is because the object's motion relative to the eye still contains implicit information about forward movement (looming), even if it biases lateral perception.
- Optimal Integration: The monkeys integrated senses in a "near-optimal" way, weighing each sense by its reliability. When background flow was "noisy" (low coherence), the moving object had a much stronger biasing effect on visual-only trials, but the combined condition remained robust.
Discussion and Limitations
While the study demonstrates that vestibular signals are critical for dissociating self-motion from object motion, it used simplified side-to-side object motion. Real-world 3D motion creates more complex flow distortions. Additionally, the monkeys' extensive training might have influenced the small biases seen in vestibular-only conditions. Future work involving direct neural recordings during this specific "flow parsing" task is necessary to confirm the "opposite cell" hypothesis in area MSTd.
Summary updated