Empowered by the rapid advancement of digital technologies,including Big Data,artificial intelligence(AI),and virtual reality,human society has transformed from the era of information to the era of digital intelligenc...Empowered by the rapid advancement of digital technologies,including Big Data,artificial intelligence(AI),and virtual reality,human society has transformed from the era of information to the era of digital intelligence.Unlike previous social formations,the digital-intelligent society has disrupted many long-held consensus norms and introduced numerous difficult challenges.To cultivate adaptive talents with general literacy of digital intelligence and specific professional competences,psychology,as one of the foundations of social sciences,must launch a revolution in future-oriented education.In higher education,the two principal components,defined by their nature and objective,are knowledge-oriented and research-oriented teaching.The former is designed to provide an introduction to the fundamental principles and basic knowledge of psychology for freshmen and sophomores,while the latter is intended to equip junior and senior undergraduates with the skills necessary for conducting scientific research.First,it is both possible and necessary to integrate AI throughout the processes of knowledge-oriented teaching.In this article,we propose a“loop model”to demonstrate the applications of AI in the knowledge-oriented phase.Furthermore,to provide a reference criterion for nurturing innovative and research-oriented students,we present a theoretical framework of“chimeric research”to provide a comprehensive overview of psychology research in the era of AI.In conclusion,psychology education needs to be aligned with the demands of the modern society and embrace digital intelligence in both knowledge-and research-oriented teaching phases.展开更多
Understanding how drivers perceive and respond to external stimuli in driving tasks is important for the development of advanced driving technologies and human-computer interaction.In this paper,we conducted a tempora...Understanding how drivers perceive and respond to external stimuli in driving tasks is important for the development of advanced driving technologies and human-computer interaction.In this paper,we conducted a temporal response analysis between driving data and cortical activation data measured by functional near-infrared spectroscopy(fNIRS),based on a naturalistic driving experiment.Temporal response function analysis indicates that stimuli,which elicit significant responses of drivers include distance,acceleration,time headway,and the velocity of the preceding vehicle.For these stimuli,the time lags and response patterns were further discussed.The influencing factors on drivers’perception were also studied based on various driver characteristics.These conclusions can provide guidance for the construction of car-following models,the safety assessment of drivers and the improvement of advanced driving technologies.展开更多
The visual system continuously adapts to the statistical properties of the environment. Existing evidence shows a close resemblance between deep convolutional neural networks(CNNs) and primate visual stream in neural ...The visual system continuously adapts to the statistical properties of the environment. Existing evidence shows a close resemblance between deep convolutional neural networks(CNNs) and primate visual stream in neural selectivity to naturalistic textures above the primary visual processing stage. This study delves into the mechanisms of perceptual learning in CNNs,focusing on how they assimilate the high-order statistics of natural textures. Our results show that a CNN model achieves a similar performance improvement as humans, as manifested in the learning pattern across different types of high-order image statistics. While L2 was the first stage exhibiting texture selectivity, we found that stages beyond L2 were critically involved in learning. The significant contribution of L4 to learning was manifested both in the modulations of texture-selective responses and in the consequences of training with frozen connection weights. Our findings highlight learning-dependent plasticity in the mid-to-high-level areas of the visual hierarchy. This research introduces an AI-inspired approach for studying learning-induced cortical plasticity, utilizing DCNNs as an experimental framework to formulate testable predictions for empirical brain studies.展开更多
Dear Editor,The adult human visual system is capable of reshaping its oculomotor control and sensory coding to adapt to impoverished visual inputs.When one's central vision is deprived,a spared part of the periphe...Dear Editor,The adult human visual system is capable of reshaping its oculomotor control and sensory coding to adapt to impoverished visual inputs.When one's central vision is deprived,a spared part of the peripheral retina acts as a pseudo fovea termed as preferred retinal locus(PRL).In people with normal vision,a PRL can be induced via oculomotor training with simulated central vision loss[1,2].展开更多
While advanced Large Language Models(LLMs)can simulate human-like prosocial behaviors,the degree to which they align with human prosocial values and the underlying afective mechanisms remain unclear.This study address...While advanced Large Language Models(LLMs)can simulate human-like prosocial behaviors,the degree to which they align with human prosocial values and the underlying afective mechanisms remain unclear.This study addressed these gaps using the third-party punishment(TPP)paradigm,comparing LLM agents(GPT and DeepSeek series)with human participants(n=100).The LLM agents(n=500,100 agents per model)were one-to-one constructed based on the demographic and psychological features of human participants.Prompt engineering was employed to initiate TPP games and record punitive decisions and afective responses in LLM agents.Results revealed that:(1)GPT-4o,DeepSeek-V3,and DeepSeek-R1 models demonstrated stronger fairness value alignment,choosing punitive options more frequently than humans in TPP games;(2)all LLMs replicated the human pathway from unfairness through negative afective response to punitive decisions,with stronger mediation efects of negative emotions observed in DeepSeek models than GPT models;(3)only DeepSeek-R1 exhibited the human-like positive feedback loop from previous punitive decisions to positive afective feedback and subsequent punitive choices;(4)most LLMs(excluding GPT-3.5)showed signifcant representational similarity to human afect-decision patterns;(5)notably,all LLMs displayed rigid afective dynamics,characterized by lower afective variability and higher afective inertia than the fexible,contextsensitive fuctuations observed in humans.These fndings highlight notable advances in prosocial value alignment but underscore the necessity to enhance their afective dynamics to foster robust,adaptive prosocial LLMs.Such advancements could not only accelerate LLMs'alignment with human values but also provide empirical support for the broader applicability of prosocial theories to LLM agents.展开更多
The brain processes information in the visual hierarchy through parallel streams.In this review,we first introduce the visual hierarchy at the macroscopic level along the dorsal and ventral streams.We then delve into ...The brain processes information in the visual hierarchy through parallel streams.In this review,we first introduce the visual hierarchy at the macroscopic level along the dorsal and ventral streams.We then delve into the mesoscopic modular architecture of the extrastriate area to illustrate how parallel and hierarchical processing are implemented in the early-to-mid visual system.These organizational principles have influenced the design of artificial neural networks.Advances in neuroimaging and neural networks now enable the exploration of modular organization in the visual cortex,as well as their digital counterparts.We propose that integrating biological principles with computational approaches can deepen our understanding of the visual system and promote the development of biologically plausible artificial intelligence.展开更多
The subgenual anterior cingulate cortex(sgACC)plays a central role in the pathophysiology of major depressive disorder(MDD).Its functional interactive profile with the left dorsal lateral prefrontal cortex(DLPFC)is as...The subgenual anterior cingulate cortex(sgACC)plays a central role in the pathophysiology of major depressive disorder(MDD).Its functional interactive profile with the left dorsal lateral prefrontal cortex(DLPFC)is associated with transcranial magnetic stimulation(TMS)treatment outcomes.Previous research on sgACC functional connectivity(FC)in MDD has yielded inconsistent results,partly due to small sample sizes and limited statistical power.Furthermore,calculating sgACC-FC to target TMS individually is challenging.We used a large multi-site cross-sectional sample(1660 patients with MDD vs.1341 healthy controls)from Phase Ⅱ of the Depression Imaging REsearch ConsorTium(DIRECT)to systematically delineate case-control difference maps of sgACC-FC.We explored the potential impact of group-level abnormality profiles on TMS target localization and clinical efficacy.Next,we developed an MDD big data-guided,individualized TMS targeting algorithm to integrate group-level statistical maps with individual-level brain activity to individually localize TMS targets.We found enhanced sgACCDLPFC FC in patients with MDD compared with healthy controls(HC).These group differences altered the position of the sgACC anti-correlation peak in the left DLPFC.We showed that the magnitude of case-control differences in the sgACC-FC was related to clinical improvement in two independent clinical samples.This targeting algorithm may generate targets demonstrating stronger associations with clinical efficiency than group-level targets.We reliably delineated MDD-related abnormalities of sgACC-FC profiles in a large,independently ascertained sample and demonstrated the potential impact of such casecontrol differences on FC-guided localization of TMS targets.展开更多
文摘Empowered by the rapid advancement of digital technologies,including Big Data,artificial intelligence(AI),and virtual reality,human society has transformed from the era of information to the era of digital intelligence.Unlike previous social formations,the digital-intelligent society has disrupted many long-held consensus norms and introduced numerous difficult challenges.To cultivate adaptive talents with general literacy of digital intelligence and specific professional competences,psychology,as one of the foundations of social sciences,must launch a revolution in future-oriented education.In higher education,the two principal components,defined by their nature and objective,are knowledge-oriented and research-oriented teaching.The former is designed to provide an introduction to the fundamental principles and basic knowledge of psychology for freshmen and sophomores,while the latter is intended to equip junior and senior undergraduates with the skills necessary for conducting scientific research.First,it is both possible and necessary to integrate AI throughout the processes of knowledge-oriented teaching.In this article,we propose a“loop model”to demonstrate the applications of AI in the knowledge-oriented phase.Furthermore,to provide a reference criterion for nurturing innovative and research-oriented students,we present a theoretical framework of“chimeric research”to provide a comprehensive overview of psychology research in the era of AI.In conclusion,psychology education needs to be aligned with the demands of the modern society and embrace digital intelligence in both knowledge-and research-oriented teaching phases.
基金supported by the National Key R&D Program of China(No.2021YFC3001500).
文摘Understanding how drivers perceive and respond to external stimuli in driving tasks is important for the development of advanced driving technologies and human-computer interaction.In this paper,we conducted a temporal response analysis between driving data and cortical activation data measured by functional near-infrared spectroscopy(fNIRS),based on a naturalistic driving experiment.Temporal response function analysis indicates that stimuli,which elicit significant responses of drivers include distance,acceleration,time headway,and the velocity of the preceding vehicle.For these stimuli,the time lags and response patterns were further discussed.The influencing factors on drivers’perception were also studied based on various driver characteristics.These conclusions can provide guidance for the construction of car-following models,the safety assessment of drivers and the improvement of advanced driving technologies.
基金supported by the National Natural Science Foundation of China (Grant Nos. 31971031, 31930053, and 32171039)the STI2030Major Projects (Grant Nos. 2021ZD0203600, 2022ZD0204802, and 2022ZD0204804)。
文摘The visual system continuously adapts to the statistical properties of the environment. Existing evidence shows a close resemblance between deep convolutional neural networks(CNNs) and primate visual stream in neural selectivity to naturalistic textures above the primary visual processing stage. This study delves into the mechanisms of perceptual learning in CNNs,focusing on how they assimilate the high-order statistics of natural textures. Our results show that a CNN model achieves a similar performance improvement as humans, as manifested in the learning pattern across different types of high-order image statistics. While L2 was the first stage exhibiting texture selectivity, we found that stages beyond L2 were critically involved in learning. The significant contribution of L4 to learning was manifested both in the modulations of texture-selective responses and in the consequences of training with frozen connection weights. Our findings highlight learning-dependent plasticity in the mid-to-high-level areas of the visual hierarchy. This research introduces an AI-inspired approach for studying learning-induced cortical plasticity, utilizing DCNNs as an experimental framework to formulate testable predictions for empirical brain studies.
基金supported by the National Natural Science Foundation of China(31971031and 31930053)STI2030-Major Projects(2021ZD0203600)supported by NIH R01-EY017707.
文摘Dear Editor,The adult human visual system is capable of reshaping its oculomotor control and sensory coding to adapt to impoverished visual inputs.When one's central vision is deprived,a spared part of the peripheral retina acts as a pseudo fovea termed as preferred retinal locus(PRL).In people with normal vision,a PRL can be induced via oculomotor training with simulated central vision loss[1,2].
基金supported by the National Natural Science Foundation of China(Grant Nos.32271110,62441614)the Tsinghua University Initiative Scientific Research Program(Grant No.20235080047)。
文摘While advanced Large Language Models(LLMs)can simulate human-like prosocial behaviors,the degree to which they align with human prosocial values and the underlying afective mechanisms remain unclear.This study addressed these gaps using the third-party punishment(TPP)paradigm,comparing LLM agents(GPT and DeepSeek series)with human participants(n=100).The LLM agents(n=500,100 agents per model)were one-to-one constructed based on the demographic and psychological features of human participants.Prompt engineering was employed to initiate TPP games and record punitive decisions and afective responses in LLM agents.Results revealed that:(1)GPT-4o,DeepSeek-V3,and DeepSeek-R1 models demonstrated stronger fairness value alignment,choosing punitive options more frequently than humans in TPP games;(2)all LLMs replicated the human pathway from unfairness through negative afective response to punitive decisions,with stronger mediation efects of negative emotions observed in DeepSeek models than GPT models;(3)only DeepSeek-R1 exhibited the human-like positive feedback loop from previous punitive decisions to positive afective feedback and subsequent punitive choices;(4)most LLMs(excluding GPT-3.5)showed signifcant representational similarity to human afect-decision patterns;(5)notably,all LLMs displayed rigid afective dynamics,characterized by lower afective variability and higher afective inertia than the fexible,contextsensitive fuctuations observed in humans.These fndings highlight notable advances in prosocial value alignment but underscore the necessity to enhance their afective dynamics to foster robust,adaptive prosocial LLMs.Such advancements could not only accelerate LLMs'alignment with human values but also provide empirical support for the broader applicability of prosocial theories to LLM agents.
基金supported by the STI2030-Major Projects(Grant No.2021ZD0203600)the National Natural Science Foundation of China(Grant Nos.31971031,31930053)the Research Center for Brain Cognition and Human Development,Guangdong,China(Grant No.2024B0303390003)。
文摘The brain processes information in the visual hierarchy through parallel streams.In this review,we first introduce the visual hierarchy at the macroscopic level along the dorsal and ventral streams.We then delve into the mesoscopic modular architecture of the extrastriate area to illustrate how parallel and hierarchical processing are implemented in the early-to-mid visual system.These organizational principles have influenced the design of artificial neural networks.Advances in neuroimaging and neural networks now enable the exploration of modular organization in the visual cortex,as well as their digital counterparts.We propose that integrating biological principles with computational approaches can deepen our understanding of the visual system and promote the development of biologically plausible artificial intelligence.
基金the Beijing Nova Program of Science and Technology(20230484465)the Beijing Natural Science Foundation(J230040)+12 种基金the National Natural Science Foundation of China(82122035,81671774,81630031,and 32300933)the Sci-Tech Innovation 2030–Major Project of Brain Science and Braininspired Intelligence Technology(2021ZD0200600)the National Key R&D Program of China(2017YFC1309902)the Key Research Program of the Chinese Academy of Sciences(ZDBS-SSW-JSC006)the Scientific Foundation of Institute of Psychology,Chinese Academy of Sciences(E2CX4425YZ,E3CX1315,and Y9CX422005)the China Postdoctoral Science Foundation(2019M660847)the China National Postdoctoral Program for Innovative Talents(BX20200360)the Special Research Assistant Program of the Chinese Academy of Sciences(E2CX0624)the Key R&D Program of Sichuan Province(2023YFS0076)the Canadian Institutes of Health Research(CIHR),the National Institutes of Health–US(NIH)the Brain Canada Foundationthe Temerty Family through the Centre for Addiction and Mental Health(CAMH)Foundation and the Campbell Family Research Institutethe China Scholarship Council(202104910248)during a visit of Xiao Chen to the Centre for Addiction and Mental Health is acknowledged.
文摘The subgenual anterior cingulate cortex(sgACC)plays a central role in the pathophysiology of major depressive disorder(MDD).Its functional interactive profile with the left dorsal lateral prefrontal cortex(DLPFC)is associated with transcranial magnetic stimulation(TMS)treatment outcomes.Previous research on sgACC functional connectivity(FC)in MDD has yielded inconsistent results,partly due to small sample sizes and limited statistical power.Furthermore,calculating sgACC-FC to target TMS individually is challenging.We used a large multi-site cross-sectional sample(1660 patients with MDD vs.1341 healthy controls)from Phase Ⅱ of the Depression Imaging REsearch ConsorTium(DIRECT)to systematically delineate case-control difference maps of sgACC-FC.We explored the potential impact of group-level abnormality profiles on TMS target localization and clinical efficacy.Next,we developed an MDD big data-guided,individualized TMS targeting algorithm to integrate group-level statistical maps with individual-level brain activity to individually localize TMS targets.We found enhanced sgACCDLPFC FC in patients with MDD compared with healthy controls(HC).These group differences altered the position of the sgACC anti-correlation peak in the left DLPFC.We showed that the magnitude of case-control differences in the sgACC-FC was related to clinical improvement in two independent clinical samples.This targeting algorithm may generate targets demonstrating stronger associations with clinical efficiency than group-level targets.We reliably delineated MDD-related abnormalities of sgACC-FC profiles in a large,independently ascertained sample and demonstrated the potential impact of such casecontrol differences on FC-guided localization of TMS targets.
