Biology subjects

Wang, D.-H.

Publications and source records attributed to Wang, D.-H..

5 recordsLinked to original sources

bioRxiv2025

A tale of three types of internal object representations in the human brain

Perception and mental imagery are widely thought to rely on overlapping neural representations, giving rise to perception-like visual experiences during imagery generation. However, such overlap, by itself, is agnostic to the nature of the underlying neural geometry for both perception and imagery. Here we addressed this question by comparing visualizers and individuals with congenital aphantasia using fMRI, representational similarity analysis, and multimodal deep neural network models to probe different types of neural geometry. Despite comparable behavioral performance and perception-imagery generalization of neural object representations in the ventral occipitotemporal cortex, the two groups exhibited systematically different object representational geometries already in the perceptual encoding phase. Visual-similarity-aligned geometry was selectively reduced in the left lateral occipitotemporal cortex of individuals with aphantasia during perceptual encoding, before imagery generation, and its expression during imagery covaried with imagery vividness. In contrast, language-relation-aligned geometry was preserved across groups during both perception and imagery. Together, these findings suggest that subjective imagery experience is associated with the geometry of neural object representations during both perception and imagery, and that representations within perception- imagery cross-decoding regions comprise heterogeneous components rather than a unitary shared neural code.

neuroscience↗

bioRxiv2024

Non-Negative Connectivity Causes Bow-Tie Architecture in Neural Circuits

Bow-tie or hourglass architecture is commonly found in biological neural networks. Recently, artificial neural networks with bow-tie architecture have been widely used in various machine-learning applications. However, it is unclear how bow-tie architecture in neural circuits can be formed. We address this by training multi-layer neural network models to perform classification tasks. We demonstrate that during network learning and structural changes, non-negative connections amplify error signals and quench neural activity particularly in the hidden layer, resulting in the emergence of the networks bow-tie architecture. We further show that such architecture has low wiring cost, robust to network size, and generalizable to different discrimination tasks. Overall, our work suggests a possible mechanism for the emergence of bow-tie neural architecture and its functional advantages.

neuroscience↗

bioRxiv2022

Plant-on-Chip: core morphogenesis processes in the tiny plant Wolffia australiana

A plant can be thought of as a colony comprising numerous growth buds, each developing to its own rhythm. Such lack of synchrony impedes efforts to describe core principles of plant morphogenesis, dissect the underlying mechanisms, and identify regulators. Here, we use the tiniest known angiosperm to overcome this challenge and provide an ideal model system for plant morphogenesis. We present a detailed morphological description of the monocot Wolffia australiana, as well as high-quality genome information. Further, we developed the Plant-on-Chip culture system and demonstrate the application of advanced technologies such as snRNA-seq, protein structure prediction, and gene editing. We provide proof-of-concept examples that illustrate how W. australiana can open a new horizon for deciphering the core regulatory mechanisms of plant morphogenesis. SignificanceWhat is the core morphogenetic process in angiosperms, a plant like a tree indeterminately growing, or a bud sequentially generating limited types of organs? Wolffia australiana, one of the smallest angiosperms in the world may help to make a distinction. Wolffia plantlet constitutes of only three organs that are indispensable to complete life cycle: one leaf, one stamen and one gynoecium. Before the growth tip is induced to flower, it keeps branching from the leaf axil and the branches separate from the main plantlet. Here we present a high-quality genome of W. australiana, detailed morphological description, a Plant-on-Chip cultural system, and some principle-proof experiments, demonstrating that W. australiana is a promising model system for deciphering core developmental program in angiosperms.

plant biology↗

bioRxiv2022

Functional autapses form in striatal parvalbumin interneurons but not medium spiny neurons

Autapses (or self-synapses) selectively form in specific cell types in many brain regions including the neocortex and the hippocampus, where they provide feedback control over self-spiking activities. Previous morphological studies also found putative autapses in medium spiny neurons (MSNs) of the striatum. However, it remains unclear whether striatal neurons indeed form physiologically functional autapses. We performed whole-cell recordings from striatal neurons in acute mouse brain slices, and identify autaptic neurons by the occurrence of prolonged asynchronous release (AR) of neurotransmitter after high-frequency burst of action potentials (APs) in the same cell. To our surprise, we found no autaptic release in all recorded MSNs after the AP burst, even in the presence of Sr2+ that should desynchronize and thus prolong synaptic vesicle release. In sharp contrast, we observed robust autaptic AR events in half of the recorded parvalbumin (PV)-positive neurons. Autaptic responses in PV cells were mediated by GABAA receptors, and the AR strength was dependent on the frequency and the number of APs during the burst. Further simulation results show that autapses regulate burst spiking in PV cells by providing self-inhibition and thus shape network oscillation at certain frequencies. Together, we reveal that, distinct from MSNs, striatal PV neurons form functional autapses, activation of which would regulate self-activities in PV cells, and thereby shape MSN firing and network oscillations. Author summarySynapses, which usually occur between two neurons, are key structures for signal communication in the nervous system. However, some types of neurons form autapses, where a neuron synapses onto itself. Autaptic transmission provides feedback signal regulating self-spiking activities. Neuronal and network activities in the striatum play critical roles in motor control and other brain functions. Previous studies suggest formation of autapses in striatal principal MSNs, but it remains unclear whether striatal neurons form functional autapses. We performed direct recordings from striatal neurons and examined the occurrence of autaptic transmission in acute brain slices. Surprisingly, we did not detect any autaptic responses in MSNs. A large proportion of striatal PV neurons, however, produced robust autaptic GABA release upon high-frequency stimulation, indicating selective formation of autapses in striatal PV cells. Our computation simulations suggest that autapses provide self-inhibition in PV cells and thereby shape activities in MSNs and striatal network, particularly when PV cells discharge at high frequencies corresponding to a high dopamine state. Together, our findings indicate that PV cells, but not MSNs, in the striatum form physiologically functional autapses. Autapses in PV cells could be essential circuit elements in the striatum and contribute to striatal functions, such as motor control.

neuroscience↗

bioRxiv2020

Degeneracy and stability in neural circuits of neuromodulators

Degenerate neural circuits perform the same function despite being structurally different. However, it is unclear whether neural circuits with interacting neuromodulator sources can themselves be degenerate while maintaining the same neuromodulatory function. Here, we address this by computationally modelling the neural circuits of neuromodulators serotonin and dopamine, local glutamatergic and GABAergic interneurons, and their possible interactions, under reward/punishment-based conditioning tasks. The neural modelling is constrained by relevant experimental studies of the VTA or DRN system using e.g. electrophysiology, optogenetics, and voltammetry. We first show that a single parsimonious, sparsely connected neural circuit model can recapitulate several separate experimental findings that indicated diverse, heterogeneous, distributed and mixed DRN-VTA neuronal signalling in reward and punishment tasks. The inability for this model to recapitulate all observed neuronal signalling suggests potentially multiple circuits acting in parallel. Then using computational simulations and dynamical systems analysis, we demonstrate that several different stable circuit architectures can produce the same observed network activity profile, hence demonstrating degeneracy. Due to the extensive D2-mediated connections in the investigated circuits, we simulate D2 receptor agonist by increasing the connection strengths emanating from the VTA DA neurons. We found that the simulated D2 agonist can distinguish among sub-groups of the degenerate neural circuits based on substantial deviations in specific neural populations activities in reward and punishment conditions. This forms a testable model prediction using pharmacological means. Overall, this theoretical work suggests the plausibility of degeneracy within neuromodulator circuitry and has important implication for the stable and robust maintenance of neuromodulatory functions.

neuroscience↗