Biology subjects

Si, B.

Publications and source records attributed to Si, B..

4 recordsLinked to original sources

bioRxiv2026

Sparse mixed codes on shared manifolds for human-like spatial attention in artificial neural networks

Spatial attention is often partitioned into endogenous, exogenous, and social forms, yet it remains unclear whether a single neural circuit can support all three and how their population codes are organized. Here we trained recurrent artificial neural networks (ANNs) with convolutional sensory front-ends on three classic cueing paradigms (central, peripheral, and gaze cues) to reproduce human reaction time (RT) profiles across cue-target onset asynchronies. Despite differences in sensory architecture and visual experience, all ANNs captured the full facilitation-inhibition time course for all three attention types. Model-based targeted dimensionality reduction (mTDR) revealed that cue- and choice-related activity in the advanced cognitive module evolved as rotations within a shared low-dimensional manifold, with angular deflections that mirrored the distinct temporal dynamics of endogenous, exogenous, and social attention. Attentional signals were encoded by highly sparse, distributed population activity: a small subset of recurrent units explained most task-related variance, was sufficient to recover human-like RT patterns after virtual lesioning, and became progressively sparser as training improved performance. At the same time, single unit responses displayed pervasive mixed selectivity, dominated by nonlinear conjunctions of cue type, cue direction, and validity, whose strength and heterogeneity robustly predicted model performance. Together, these results identified low-dimensional geometric rotations, sparse coding, and nonlinear mixed selectivity as core computational principles through which a single recurrent circuit could generate human-like temporal dynamics across endogenous, exogenous, and social orienting, and provided testable predictions for population-level mechanisms of spatial attention in the brain.

neuroscience↗

bioRxiv2024

Precision-dependent modulation of social attention

Social attention, guided by cues like gaze direction, is crucial for effective social interactions. However, how dynamic environmental context modulates this process remains unclear. Integrating a hierarchical Bayesian model with fMRI, this study investigated how individuals adjusted attention based on the predictions about cue validity (CV). Thirty-three participants performed a modified Posner location-cueing task with varying CV. Behaviorally, individuals allocation of social attention was finely tuned to the precision (inverse variance) of CV predictions, with the predictions being updated by precision-weighted prediction errors (PEs) about the occurrence of target locations. Neuroimaging results revealed that the interaction between allocation of social attention and CV influenced activity in regions involved in spatial attention and/or social perception, such as the temporoparietal junction (TPJ), frontal eye field (FEF), superior temporal sulcus (STS), and inferior parietal sulcus (IPS). Precision-weighted PEs about target locations specifically modulated activity in the TPJ, STS, and primary visual cortex (V1), underscoring their roles in refining attentional predictions. Dynamic causal modeling (DCM) further demonstrated that enhanced absolute precision-weighted PEs about target locations strengthened the effective connectivity from V1 and STS to TPJ, emphasizing their roles in conveying residual error signals from low-level sensory areas to high-level critical attention areas. These findings elucidated how the precision of contextual predictions dynamically modulated social attention, offering insights into the computational and neurocognitive mechanisms of context-dependent social attention.

neuroscience↗

bioRxiv2022

Causal contributions to sensory-based decision-making by cell-type specific circuits in the tail striatum

The striatum comprises distinct types of neurons giving rise to the direct and indirect basal ganglia pathways and local circuits. A large amount of work has been focusing on cell-type specific striatal circuits in the context of movement control, proposing several models on their functional roles. But it remains to be elucidated how the cell-type specific striatal circuits contribute to decision-making behavior and whether the existing models apply. Here, we investigate the causal roles of the cell-type specific circuits in the posterior tail of the dorsal striatum (TS) of mice in an auditory-guided decision-making behavior. Transient unilateral activation of the direct- or indirect-pathway striatal spiny projection neurons (dSPNs or iSPNs) both biased decisions in opposite directions. These effects, however, were not due to a direct influence on movement, but was specific to the decision period preceding action execution. Optogenetic inactivation of dSPNs and iSPNs revealed their opposing causal contributions to decisions. At the local circuit level, simutaneous optical recording and manipulation of dSPNs and iSPNs revealed their antagnizing interactions. Inactivation of PV interneurons, a common inhibitory input to both dSPNs and iSPNs, facilitated contraversive choices, supporting a causal contribution of coordinated striatal circuits. Using a neural circuit model, we further demonstrated the computational implemenation of the causal circuit mechanism. Our results indicate that while the causal roles of the cell-type specific striatal circuits in decision-making largely agree with classic models in movement control, they show decision task-related specificity involving local circuit coordination.

neuroscience↗

bioRxiv2021

Asymmetrical choice-related ensemble activity in direct and indirect-pathway striatal neurons drives perceptual decisions

Sensory-guided decision-making is a vital brain function critically depending on the striatum, a key brain structure transforming sensorimotor information into actions. However, how the two opposing striatal pathways work in concert to select actions during decision-making remains controversial. Here, using cell-type specific two-photon imaging and optogenetic perturbations from the posterior dorsal striatum during decision-making behavior in mice, we uncover the population coding and causal mechanisms of the direct- and indirect-pathway spiny projection neurons (dSPNs and iSPNs) in decision-related action selection. Unexpected from prevailing models, we found that both dSPNs and iSPNs contain divergent subpopulations representing competing choices, and exhibit ensemble-level asymmetry: stronger contralateral dominance in dSPNs than in iSPNs. Such multi-ensemble competition/cooperation causally contributes to decision-related action selection, as supported by systematic optogenetic manipulations and verified by computational modeling. Our results unravel a multi-ensemble coordination mechanism in the striatum for action selection during decision-making.

neuroscience↗