Abstract
Multisensory plasticity enables our senses to dynamically adapt to each other and the external environment, a fundamental operation that our brain performs continuously. We searched for neural correlates of adult multisensory plasticity in the dorsal medial superior temporal area (MSTd) and the ventral intraparietal area (VIP) in 2 male rhesus macaques using a paradigm of supervised calibration. We report little plasticity in neural responses in the relatively low-level multisensory cortical area MSTd. In contrast, neural correlates of plasticity are found in higher-level multisensory VIP, an area with strong decision-related activity. Accordingly, we observed systematic shifts of VIP tuning curves, which were reflected in the choice-related component of the population response. This is the first demonstration of neuronal calibration, together with behavioral calibration, in single sessions. These results lay the foundation for understanding multisensory neural plasticity, applicable broadly to maintaining accuracy for sensorimotor tasks.
Introduction
To interact effectively with the world, our brains must be both precise (consistent) and accurate (unbiased). However, our sensory organs—like our eyes and our inner ears—can drift or provide conflicting information. When sensory feedback indicates an error, the brain must "recalibrate" to maintain internal consistency. While behavioral evidence for this calibration is well-documented, the underlying neural mechanisms remain elusive.
A key phenomenon in multisensory processing is "cue yoking": correcting one sense (e.g., vision) often leads to a corresponding shift in another sense (e.g., vestibular) to preserve their alignment. This study investigates whether such multisensory recalibration occurs in early multisensory processing areas, such as the dorsal medial superior temporal area (MSTd), or in higher-order regions more closely associated with perceptual decision-making, like the ventral intraparietal area (VIP).
Methods
Two rhesus macaques were trained on a heading discrimination task, requiring them to determine if their movement was left or right of center. Physical motion (vestibular) and moving star fields (visual optic flow) were provided using a 6-degree-of-freedom motion platform.
Researchers introduced a $10^\circ$ discrepancy between visual and vestibular headings. Calibration was "supervised" by rewarding choices consistent with only one of the cues (the "accurate" cue). Neural activity was recorded from 223 stable cells in VIP and 147 stable cells in MSTd. Targeted Dimensionality Reduction (TDR) was employed to decompose the neural population response into stimulus-related (heading) and decision-related (choice) components.
Results
Behaviorally, both monkeys exhibited significant shifts in heading perception for both visual and vestibular cues, even when only one cue was explicitly rewarded. This confirms that supervised calibration of one sense induces a yoked shift in the other.
The neural data revealed a striking divergence between the two brain regions. In VIP, neuronal tuning curve shifts were highly correlated with behavioral changes ($R = 0.45$ to $0.88$). In contrast, MSTd showed almost no plasticity, with correlations near zero ($R \approx 0.04$ to $0.10$) and statistically insignificant.
TDR analysis further clarified the nature of the plasticity in VIP. The observed calibration did not stem from a re-tuning of how neurons represented the physical heading stimulus. Instead, it was a shift in the choice-related component of the population response. Essentially, VIP neurons changed how they "voted" for a decision based on the same sensory input. These shifts were evident throughout the stimulus duration, peaking alongside maximum stimulus velocity.
Discussion
These findings suggest a functional hierarchy in multisensory processing. While MSTd provides a stable, relatively fixed representation of sensory input, VIP adapts this input based on task demands and feedback. The rapid, single-session plasticity observed in VIP highlights its role in evidence accumulation and decision-making.
While the correlation in VIP is robust, further research using causal interventions like optogenetics is needed to definitively establish VIP's role in driving behavioral shifts. Additionally, while MSTd appeared rigid in these single-session experiments, slower "unsupervised" forms of plasticity might still occur over longer timescales. This study provides a foundational understanding of how higher-level cortical areas manage multisensory calibration to maintain sensorimotor accuracy.
Reference: Zaidel, A., Laurens, J., DeAngelis, G. C., & Angelaki, D. E. (2021). Supervised Multisensory Calibration Signals Are Evident in VIP But Not MSTd. The Journal of Neuroscience, 41(49), 10108–10119.
Summary updated