Search

Deep convolutional neural networks (DCNNs) have been found to demonstrate hierarchical mapping to human brain regions on tasks such as object recognition. However, it remains unclear if such hierarchical mapping also applies to action recognition, which involves dynamic visual information processing. Here, we compared action representations of two-stream DCNNs to the human visual system. Five visual areas that are associated with object and action processing were selected. Nine human action categories were adopted from three semantic classes to examine the action representations of both DCNNs and human visual areas. In two fMRI experiments, actions were presented in the forms of computer-rendered videos and point-light biological motion videos. Results showed that although two-stream DCNNs demonstrated hierarchical representations of actions as layers grow deeper, DCNNs lack a hierarchical mapping to human visual areas. Consistently across different video displays and DCNN pathways, only the top DCNN layers demonstrated high similarity to representations in the human visual system. The results suggest that the dynamic representations of human actions may be different in DCNNs compared to the human visual system, even after big-data training.

BackgroundThe heterogeneous nature of mood and anxiety disorders highlights a need for dimensionally based descriptions of psychopathology that inform better classification and treatment approaches. Following the Research Domain Criteria approach, this investigation sought to derive constructs assessing positive and negative valence domains across multiple units of analysis.MethodsAdults with clinically impairing mood and anxiety symptoms (N = 225) completed comprehensive assessments across several units of analysis. Self-report assessments included nine questionnaires that assess mood and anxiety symptoms and traits reflecting the negative and positive valence systems. Behavioral assessments included emotional reactivity and distress tolerance tasks, during which skin conductance and heart rate were measured. Neuroimaging assessments included fear conditioning and a reward processing task. The latent variable structure underlying these measures was explored using sparse Bayesian group factor analysis.ResultsGroup factor analysis identified 11 latent variables explaining 31.2% of the variance across tasks, none of which loaded across units of analysis or tasks. Instead, variance was best explained by individual latent variables for each unit of analysis within each task. Post hoc analyses 1) showed associations with small effect sizes between latent variables that were derived separately from functional magnetic resonance imaging and self-report data and 2) showed that some latent variables are not directly related to individual valence system constructs.ConclusionsThe lack of latent structure across units of analysis highlights challenges of the Research Domain Criteria approach and suggests that while dimensional analyses work well to reveal within-task features, more targeted approaches are needed to reveal latent cross-modal relationships that could illuminate psychopathology.

We see the world as continuous with smooth movements ofobjects and people, even though visual inputs can consist ofstationary frames. The perceptual construction of smoothmovements depends not only on low-level spatiotemporalfeatures but also high-level knowledge. Here, we examined therole of causality in guiding perceptual interpolation of motion inthe observation of human actions. We recorded videos of naturalhuman-object interactions. Frame rate was manipulated to yieldshort and long stimulus-onset-asynchrony (SOA) displays for ashort clip in which a catcher prepared to receive a ball. Thefacing direction of the catcher was either maintained intact togenerate a meaningful interaction consistent with causality, orwas transformed by a mirror reflection to create a non-causalsituation lacking a meaningful interaction. Across threeexperiments, participants were asked to judge whether thecatcher’s action showed smooth movements or sudden changes.Participants were more likely to judge the catcher’s actions to becontinuous in the causal condition than in the non-causalcondition, even with long SOA displays. This causalinterpolation effect was robust to manipulations of bodyorientation (i.e. upright versus inverted). These findings indicatethat causality in human actions guides interpolation of bodymovements, thereby completing the history of an observedaction despite gaps in the sensory information. Hence, causalknowledge not only makes us see the future, but also fills ininformation about recent history.

Humans demonstrate remarkable abilities to perceive physi-cal and social events based on very limited information (e.g.,movements of a few simple geometric shapes). However, thecomputational mechanisms underlying intuitive physics andsocial perception remain unclear. In an effort to identify thekey computational components, we propose a unified psycho-logical space that reveals the partition between the perceptionof physical events involving inanimate objects and the percep-tion of social events involving human interactions with otheragents. This unified space consists of two prominent dimen-sions: an intuitive sense of whether physical laws are obeyedor violated; and an impression of whether an agent possessesintentions, as inferred from movements. We adopt a physicsengine and a deep reinforcement learning model to synthe-size a rich set of motion patterns. In two experiments, humanjudgments were used to demonstrate that the constructed psy-chological space successfully partitions human perception ofphysical versus social events.

Deep convolutional neural networks (DCNNs) have attained human-level performance for object categorization and exhibited representation alignment between network layers and brain regions. Does such representation alignment naturally extend to other visual tasks beyond recognizing objects in static images? In this study, we expanded the exploration to the recognition of human actions from videos and assessed the representation capabilities and alignment of two-stream DCNNs in comparison with brain regions situated along ventral and dorsal pathways. Using decoding analysis and representational similarity analysis, we show that DCNN models do not show hierarchical representation alignment to human brain across visual regions when processing action videos. Instead, later layers of DCNN models demonstrate greater representation similarities to the human visual cortex. These findings were revealed for two display formats: photorealistic avatars with full-body information and simplified stimuli in the point-light display. The discrepancies in representation alignment suggest fundamental differences in how DCNNs and the human brain represent dynamic visual information related to actions. [ABSTRACT FROM AUTHOR]

Significant progress has recently been made in the biosynthesis of germacrene A using microbial cell factories. Germacrene A is a crucial precursor for the synthesis of anti‐cancer active compounds. However, its hydrophobic characteristics lead to its aggregation in cell membranes and cause severe cytotoxicity. In the present study, we found that rhamnolipids (RLs), as toxicity antidotes, could promote the production of germacrene A. An optimal RLs concentration of 1.25 g L−1 resulted in an increase of over 30% in the germacrene A titer at both shake flask and bioreactor scales. Mechanistic analysis showed that the addition of RLs could dramatically reduce aqueous‐phase surface tension and cell surface hydrophobicity (CSH), and increase the cell membrane permeability. This, in turn, promoted an efficient transfer of germacrene A from cell membrane to extraction phases. The addition of RLs also increased the adenosine triphosphate (ATP) concentration and the nicotinamide adenine dinucleotide (NAD+/NADH) ratio, while reducing reactive oxygen species (ROS) levels. Correspondingly, gene transcripts for key enzymes associated with germacrene A biosynthesis, the respiratory chain, and ROS scavenging were upregulated significantly. This study provides an effective RLs‐regulated fermentation method for the biosynthesis of hydrophobic natural products. [ABSTRACT FROM AUTHOR]

We examined mechanisms underlying infants' ability to categorize human biological motion stimuli from sex-typed walk motions, focusing on how visual attention to dynamic information in point-light displays (PLDs) contributes to infants' social category formation. We tested for categorization of PLDs produced by women and men by habituating infants to a series of female or male walk motions and then recording posthabituation preferences for new PLDs from the familiar or novel category (Experiment 1). We also tested for intrinsic preferences for female or male walk motions (Experiment 2). We found that infant boys were better able to categorize PLDs than were girls and that male PLDs were preferred overall. Neither of these effects was found to change with development across the observed age range (∼4-18 months). We conclude that infants' categorization of walk motions in PLDs is constrained by intrinsic preferences for higher motion speeds and higher spans of motion and, relatedly, by differences in walk motions produced by men and women.

Human understanding of relations between objects depends onthe ability to code meaningful role bindings. Computationalmodels of relational reasoning have proposed that neuraloscillations provide a basic mechanism enabling workingmemory to code the bindings of objects into relational roles.We adapted a behavioral oscillation paradigm to investigatemoment-to-moment changes in representations of semanticroles. On each trial, a picture was presented showing an action(chasing) relating two animals, one animal playing an agentrole (chaser) and the other playing a patient role (chased). Afterthe picture disappeared, the inter-stimulus interval (ISI) wasvaried in densely-sampled increments followed by a verbalprobe indicating an animal in a role. Reaction time (RT) todecide the validity of the verbal probe was recorded. We foundthat RTs varied systematically with ISI in an oscillatoryfashion. A task that required memory for a relational roleevoked stronger theta- and alpha-band oscillations than did amemory task not involving relational roles. The behavioraloscillation patterns in the role-identification task revealed aphase shift between the two semantic roles in the alpha band.

People are adept at perceiving interactions from movementsof simple shapes but the underlying mechanism remains un-known. Previous studies have often used object movementsdefined by experimenters. The present study used aerial videosrecorded by drones in a real-life environment to generate de-contextualized motion stimuli. Motion trajectories of dis-played elements were the only visual input. We measuredhuman judgments of interactiveness between two moving el-ements, and the dynamic change of such judgments over time.A hierarchical model was developed to account for human per-formance in this task, which represents interactivity using la-tent variables, and learns the distribution of critical movementfeatures that signal potential interactivity. The model providesa good fit to human judgments and can also be generalized tothe original Heider-Simmel animations (1944). The model canalso synthesize decontextualized animations with controlleddegree of interactiveness, providing a viable tool for studyinganimacy and social perception.

BACKGROUND: Insect‐resistance genetically modified (GM) plants derived from Bacillus thuringiensis (Bt) have been cultivated to control pests, but continuous cultivation of Bt‐transgenic plants at large‐scale regions leads to the resistance evolution of target insects to transgenic plants. RNA interference (RNAi) technology is considered an effective strategy in delaying the resistance evolution of target insects. RESULTS: We here developed a single transgenic oilseed rape (Brassica napus) line with hairpin RNA of the chitin‐synthase 1 gene (CHS1) of Plutella xylostella (hpPxCHS1) and a pyramid transgenic B. napus line harboring hpPxCHS1 and Bt gene (Cry1Ac). Escherichia coli HT115 delivered hpPxCHS1 showed negative effects on the growth of P. xylostella. The single transgenic and pyramid transgenic B. napus significantly reduced the larval weight and length of P. xylostella and increased its lethality rate, with down‐regulation expression of the PxCHS1 gene in insects. CONCLUSION: Compared to Bt‐transgenic B. napus, pyramid‐transgenic B. napus shorted the mortality time of P. xylostella, indicating that RNAi technology synergistic with Bt protein improves the effectiveness of controlling target insects. Our results proved that RNAi can delay the resistance evolution of target insects to Bt‐transgenic plants. © 2024 Society of Chemical Industry. [ABSTRACT FROM AUTHOR]

Human actions are more than mere body movements. Incontrast to other dynamic events in the natural world, humanactions involve mental processes that enable willful bodilymovements. We reported two experiments to demonstrate thathuman observers spontaneously assign the role of cause torelative limb movements, and the role of effect to body motion(i.e., the position changes of the body center of mass) whenobserving actions of others. Experiment 1 showed that thiscausal action constraint impacts people’s impression on thenaturalness of observed actions. Experiment 2a/b revealed thatthe causal constraint guides the integration of different motioncues within a relational schema. We developed an idealobserver model to rule out the possibility that these effectsresulted from the learning of statistical regularity in actionstimuli. These findings demonstrate that causal relationsconcerning bodily movements play an important role inperceiving and understanding actions.

BACKGROUND: Yeast biomass, encompassing fatty acids, terpenoids, vitamins, antioxidants, enzymes, and other bioactive compounds have been extensively utilized in food‐related fields. The safety and potential bioactivities of Scheffersomyces segobiensis DSM 27193, an oleaginous yeast strain, are unclear. RESULTS: Scheffersomyces segobiensis DSM 27193 accumulated large palmitoleic acid (POA) levels (43.4 g kg−1 biomass) according to the results of whole‐cell components. We annotated the Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis, and predicted the categories and host of the pathogen–host interactions (PHI) genes in S. segobiensis DSM 27193. However, S. segobiensis DSM 27193 did not exert toxic effects in mice. Administration of S. segobiensis DSM 27193 led to substantial weight reduction by diminishing food intake in an obesity mouse model. Additionally, it reversed hepatic steatosis and adipose tissue hypertrophy, and improved abnormalities in serum biochemical profiles such as triglyceride, total cholesterol, low‐density lipoprotein cholesterol, lipopolysaccharide, tumor necrosis factor‐α, interleukin‐1β, and interleukin‐6. CONCLUSION: This study is the first to illustrate the safety and effects of S. segobiensis DSM 27193 against obesity and offers a scientific rationale for its application in functional food supplements. © 2023 Society of Chemical Industry. [ABSTRACT FROM AUTHOR]