Answers are grounded in this paper and cite their sources.
Examines neural signals for motion perception in monkeys, revealing cortical specialization in decision-making.
Cortical areas involved in self-motion perception process cues from multiple sensory modalities, primarily visual and vestibular, alongside choice-related activity. Neurons in these regions often respond to combinations of these factors – demonstrating a high degree of mixed selectivity. Moreover, self-motion is inherently dynamic, with each cue – visual and vestibular – comprising a mix of distinct motion signals, such as velocity and acceleration, that evolve over time. Here, we developed a time-varying targeted dimensionality reduction (tTDR) method to expose these dynamic motion and choice-related signals in neuronal populations. We applied this approach to recordings from the ventral intraparietal cortex (VIP), dorsal medial superior temporal cortex (MSTd), and parietoinsular vestibular cortex (PIVC), in six male rhesus macaques, performing a task of heading discrimination. Distinct configurations of dynamic signals were seen in the population activity across the different cortical areas. VIP exhibited all components – namely, velocity and acceleration, for visual and vestibular cues, along with strong choice-related signals that rose early and remained elevated throughout the stimulus period, well before the choice was reported. By contrast, MSTd and PIVC had weaker and later choice-related activity. MSTd primarily exhibited visual signals (velocity more than acceleration) with weak vestibular signals. PIVC predominantly exhibited vestibular signals (velocity and acceleration), yet visual signals were nonetheless present. The distinct configurations of time-varying parameters across VIP, MSTd, and PIVC, likely reflect different functional specializations in heading discrimination. Significance statement This study extends the targeted dimensionality reduction method to incorporate time-varying stimulus components. Applied to neuronal data recorded during heading discrimination, this enabled dissociation of velocity, acceleration, and choice signals in population activity, revealing clear differences between cortical areas: VIP, MSTd, and PIVC. VIP reflects motion parameters across modalities, with early and robust supramodal choice signals; MSTd predominantly encodes visual velocity with weaker vestibular and choice-related activity; and PIVC specializes in vestibular motion, with reliable encoding of both velocity and acceleration, with modest visual and late-onset choice activity. These findings reveal complementary specializations across multisensory cortical areas that together support robust heading perception and flexible decision-making.
No takes yet. Share an insight, caveat, or question.
Zeng et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: