The rapid advancement of Artificial Intelligence(AI)and Large Language Models(LLMs)has led to their increasing integration into various domains,from text generation and translation to question-answering.However,a crit...The rapid advancement of Artificial Intelligence(AI)and Large Language Models(LLMs)has led to their increasing integration into various domains,from text generation and translation to question-answering.However,a critical question remains:do these sophisticated models,much like humans,exhibit susceptibility to cognitive biases?Understanding the presence and nature of such biases in AI is paramount for assessing their reliability,enhancing their performance,and predicting their societal impact.This research specifically investigates the susceptibility of Google’s Gemini 1.5 Pro and DeepSeek,two prominent LLMs,to framing effects and confirmation bias.The study meticulously designed a series of experimental trials,systematically manipulating information proportions and presentation orders to evaluate these biases.In the framing effect experiment,a genetic testing decision-making scenario was constructed.The proportion of positive and negative information(e.g.,20%,50%,or 80%positive)and their presentation order were varied.The models’inclination towards undergoing genetic testing was recorded.For the confirmation bias experiment,two reports-one positive and one negative-about“RoboTaxi”autonomous vehicles were provided.The proportion of erroneous information within these reports(10%,30%,and 50%)and their presentation order were systematically altered,and the models’support for each report was assessed.The findings demonstrate that both Gemini 1.5 Pro and DeepSeek are susceptible to framing effects.In the genetic testing scenario,their decision-making was primarily influenced by the proportion of positive and negative information presented.When the proportion of positive information was higher,both models showed a greater inclination to recommend or proceed with genetic testing.Conversely,a higher proportion of negative information led to greater caution or a tendency not to recommend the testing.Importantly,the order in which this information was presented did not significantly influence their decisions in the framing effect scenarios.Regarding confirmation bias,the two models exhibited distinct behaviors.Gemini 1.5 Pro did not show an overall preference for either positive or negative reports.However,its judgments were significantly influenced by the order of information presentation,demonstrating a“recency effect,”meaning it tended to support the report presented later.The proportion of erroneous information within the reports had no significant impact on Gemini 1.5 Pro’s decisions.In contrast,DeepSeek exhibited an overall confirmation bias,showing a clear preference for positive reports.Similar to Gemini 1.5 Pro,DeepSeek’s decisions were also significantly affected by the order of information presentation,while the proportion of misinformation had no significant effect.These results reveal human-like cognitive vulnerabilities in advanced LLMs,highlighting critical challenges to their reliability and objectivity in decision-making processes.Gemini 1.5 Pro’s sensitivity to presentation order and DeepSeek’s general preference for positive information,coupled with its sensitivity to order,underscore the need for careful evaluation of potential cognitive biases during the development and application of AI.The study suggests that effective measures are necessary to mitigate these biases and prevent potential negative societal impacts.Future research should include a broader range of models for comparative analysis and explore more complex interactive scenarios to further understand and address these phenomena.The findings contribute significantly to understanding the limitations and capabilities of current AI systems,guiding their responsible development,and anticipating their potential societal implications.展开更多
People as third-party observers,without direct self-interest,may punish norm violators to maintain social norms.However,third-party judgment and the follow-up punishment might be susceptible to the way we frame(i.e.,v...People as third-party observers,without direct self-interest,may punish norm violators to maintain social norms.However,third-party judgment and the follow-up punishment might be susceptible to the way we frame(i.e.,verbally describe)a norm violation.We conducted a behavioral and a neuroimaging experiment to investigate the above phenomenon,which we call the“third-party framing effect”.In these experiments,participants observed an anonymous perpetrator deciding whether to keep her/his economic benefit while exposing a victim to a risk of physical pain(described as“harming others”in one condition and“not helping others”in the other condition),then they had a chance to punish that perpetrator at their own cost.Our results showed that the participants were more willing to execute third-party punishment under the harm frame compared to the help frame,manifesting a framing effect.Self-reported anger toward perpetrators mediated the relationship between empathy toward victims and the framing effect.Meanwhile,activation of the insula mediated the relationship between mid-cingulate cortex activation and the framing effect;the functional connectivity between these regions significantly predicted the size of the framing effect.These findings shed light on the psychological and neural mechanisms of the third-party framing effect.展开更多
The general relativistic frame dragging effect on the properties,such as the moments of inertia and the radiiof gyration of fast rotating neutron stars with a uniform strong magnetic field,is calculated accurate to th...The general relativistic frame dragging effect on the properties,such as the moments of inertia and the radiiof gyration of fast rotating neutron stars with a uniform strong magnetic field,is calculated accurate to the first orderin the uniform angular velocity.The results show that compared with the corresponding non-rotating static sphericalsymmetric neutron star with a weaker magnetic field,a fast rotating neutron star(millisecond pulsar)with a strongermagnetic field has a relative smaller moment of inertia and radius of gyration.展开更多
文摘The rapid advancement of Artificial Intelligence(AI)and Large Language Models(LLMs)has led to their increasing integration into various domains,from text generation and translation to question-answering.However,a critical question remains:do these sophisticated models,much like humans,exhibit susceptibility to cognitive biases?Understanding the presence and nature of such biases in AI is paramount for assessing their reliability,enhancing their performance,and predicting their societal impact.This research specifically investigates the susceptibility of Google’s Gemini 1.5 Pro and DeepSeek,two prominent LLMs,to framing effects and confirmation bias.The study meticulously designed a series of experimental trials,systematically manipulating information proportions and presentation orders to evaluate these biases.In the framing effect experiment,a genetic testing decision-making scenario was constructed.The proportion of positive and negative information(e.g.,20%,50%,or 80%positive)and their presentation order were varied.The models’inclination towards undergoing genetic testing was recorded.For the confirmation bias experiment,two reports-one positive and one negative-about“RoboTaxi”autonomous vehicles were provided.The proportion of erroneous information within these reports(10%,30%,and 50%)and their presentation order were systematically altered,and the models’support for each report was assessed.The findings demonstrate that both Gemini 1.5 Pro and DeepSeek are susceptible to framing effects.In the genetic testing scenario,their decision-making was primarily influenced by the proportion of positive and negative information presented.When the proportion of positive information was higher,both models showed a greater inclination to recommend or proceed with genetic testing.Conversely,a higher proportion of negative information led to greater caution or a tendency not to recommend the testing.Importantly,the order in which this information was presented did not significantly influence their decisions in the framing effect scenarios.Regarding confirmation bias,the two models exhibited distinct behaviors.Gemini 1.5 Pro did not show an overall preference for either positive or negative reports.However,its judgments were significantly influenced by the order of information presentation,demonstrating a“recency effect,”meaning it tended to support the report presented later.The proportion of erroneous information within the reports had no significant impact on Gemini 1.5 Pro’s decisions.In contrast,DeepSeek exhibited an overall confirmation bias,showing a clear preference for positive reports.Similar to Gemini 1.5 Pro,DeepSeek’s decisions were also significantly affected by the order of information presentation,while the proportion of misinformation had no significant effect.These results reveal human-like cognitive vulnerabilities in advanced LLMs,highlighting critical challenges to their reliability and objectivity in decision-making processes.Gemini 1.5 Pro’s sensitivity to presentation order and DeepSeek’s general preference for positive information,coupled with its sensitivity to order,underscore the need for careful evaluation of potential cognitive biases during the development and application of AI.The study suggests that effective measures are necessary to mitigate these biases and prevent potential negative societal impacts.Future research should include a broader range of models for comparative analysis and explore more complex interactive scenarios to further understand and address these phenomena.The findings contribute significantly to understanding the limitations and capabilities of current AI systems,guiding their responsible development,and anticipating their potential societal implications.
基金This work was supported by the National Natural Science Foundation of China(31871109,32071083,and 31900779)Shenzhen–Hong Kong Institute of Brain Science—Shenzhen Fundamental Research Institutions(2021SHIBS0003).
文摘People as third-party observers,without direct self-interest,may punish norm violators to maintain social norms.However,third-party judgment and the follow-up punishment might be susceptible to the way we frame(i.e.,verbally describe)a norm violation.We conducted a behavioral and a neuroimaging experiment to investigate the above phenomenon,which we call the“third-party framing effect”.In these experiments,participants observed an anonymous perpetrator deciding whether to keep her/his economic benefit while exposing a victim to a risk of physical pain(described as“harming others”in one condition and“not helping others”in the other condition),then they had a chance to punish that perpetrator at their own cost.Our results showed that the participants were more willing to execute third-party punishment under the harm frame compared to the help frame,manifesting a framing effect.Self-reported anger toward perpetrators mediated the relationship between empathy toward victims and the framing effect.Meanwhile,activation of the insula mediated the relationship between mid-cingulate cortex activation and the framing effect;the functional connectivity between these regions significantly predicted the size of the framing effect.These findings shed light on the psychological and neural mechanisms of the third-party framing effect.
基金National Natural Science Foundation of China under Grant Nos.10647116 and 10575140the China Postdoctoral Science Foundation under Grant No.2005037175
文摘The general relativistic frame dragging effect on the properties,such as the moments of inertia and the radiiof gyration of fast rotating neutron stars with a uniform strong magnetic field,is calculated accurate to the first orderin the uniform angular velocity.The results show that compared with the corresponding non-rotating static sphericalsymmetric neutron star with a weaker magnetic field,a fast rotating neutron star(millisecond pulsar)with a strongermagnetic field has a relative smaller moment of inertia and radius of gyration.
