Quasi‐Homogeneous Integrated Strain Vector Sensors for Natural Human–Machine Interaction

Advanced Functional Materials, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 2, 2025

Abstract Body motion capture is a vital approach that underpins natural human–machine interaction. Strain sensors can detect both amplitude and direction are the basis for discerning interactive intent. However, most strain with heterogeneous architecture fail to perceive accurately. Here, quasi‐homogeneous integrated vector constructed by vertically stacking elastomer meshes different fiber orientations in order. By varying orientation, proportion of intrinsic elastic structural deformation mesh be adjusted, achieving effective regulation its electromechanical properties. A highly anisotropic unidirectional fibers customized, enabling stretch simultaneously. Notably, demonstrate minimum directional resolution 2° linear working range up 100% strain. The tough bonding between interlayers ensures high robustness even after 10 000 loading cycles. Moreover, wearable 3D system built acquire comprehensive data regarding motions, including amplitudes, directions, modes, visually synchronize them virtual reality. Natural interaction achieved intuitively effortlessly altering amplitudes directions. This work provides an alternative insight immersive future.

Language: Английский

Dissociable neural networks for processing fearful bodily expressions at different spatial frequencies

Cerebral Cortex, Journal Year: 2025, Volume and Issue: 35(4)

Published: April 1, 2025

Abstract The human brain processes visual input across various spatial frequency (SF) ranges to extract emotional cues. Prior studies have extensively explored SF processing in facial expressions, yielding partly conflicting results. However, bodily which provide complementary and survival-relevant cues, remain unexplored. We investigated the neural mechanisms underlying of low (LSF), high (HSF), broad (BSF) components fearful versus neutral postures. Using functional Magnetic Resonance Imaging, we examined activity 20 participants viewing SF-filtered images expressions a semi-passive task. A multivariate “searchlight” analysis based on Multi-Voxel Pattern Analysis was employed decode non-linear activation patterns associated with each band. Our findings reveal that engages distinct networks response expressions. BSF stimuli activated widespread network, including amygdala, pulvinar, frontal, temporal cortices. These suggest general threat-detection system integrating information all SFs. HSF engaged cortical regions detailed evaluation motor planning, such as orbitofrontal cortex, anterior cingulate premotor areas, suggesting fine-grained fear cues involves computationally demanding related resonance action preparation. In contrast, LSF primarily motor-preparatory linked rapid, action-oriented responses, highlighting prioritization quick readiness low-detail threats. Notably, amygdala showed no selectivity, supporting its role generalized “relevance detector” processing. present study demonstrates flexibly adapts strategy details available underscoring complexity adaptability from signals.

Language: Английский