This study investigates distributed neural representations of conditioned threat in the human brain using functional MRI (fMRI) data from 1465 participants. The research combines multivariate pattern analysis (MVPA) with fMRI data across various threat conditioning and negative affect paradigms to decode threat and safety cues. The results show that trained decoders can sensitively and specifically distinguish between threat and safety cues across multiple datasets. The study also demonstrates that many neural nodes dynamically shift representations between threat and safety. The 'threat circuit' includes subregions of the amygdala, hippocampus, insular cortex, and medial prefrontal cortex. However, the study shows that more distributed neural systems are engaged in threat processing, including sensory and cognitive nodes. The researchers constructed an 'extended threat detection and responding circuit' that integrates the 'threat circuit' with sensory, working memory, and cognitive nodes. This extended circuit was used to train new classifiers that can distinguish threat from safety cues across different paradigms. The study found that the extended circuit and the newly trained classifiers can robustly decode threat and safety cues using activation patterns within and beyond the 'threat circuit'. The results suggest that the neural decoders are sensitive and specific in distinguishing threat from safety cues across multiple datasets. The study also highlights the importance of considering multiple neural systems for a comprehensive understanding of threat processing. The findings support the idea that threat processing involves multiple processes encompassing multiple brain circuits. The study also shows that the neural representations for conditioned threat are distinct from those evoked by intrinsically salient stimuli. The results provide compelling empirical evidence for conceptual frameworks that threat processing involves multiple processes encompassing multiple brain circuits. The study also highlights the value of other nodes that are important in conditioned threat processing. The results suggest that multiple sensory and cognitive systems are engaged during the encounter of threat and safety cues, which is likely needed for the processing of these complex representations.This study investigates distributed neural representations of conditioned threat in the human brain using functional MRI (fMRI) data from 1465 participants. The research combines multivariate pattern analysis (MVPA) with fMRI data across various threat conditioning and negative affect paradigms to decode threat and safety cues. The results show that trained decoders can sensitively and specifically distinguish between threat and safety cues across multiple datasets. The study also demonstrates that many neural nodes dynamically shift representations between threat and safety. The 'threat circuit' includes subregions of the amygdala, hippocampus, insular cortex, and medial prefrontal cortex. However, the study shows that more distributed neural systems are engaged in threat processing, including sensory and cognitive nodes. The researchers constructed an 'extended threat detection and responding circuit' that integrates the 'threat circuit' with sensory, working memory, and cognitive nodes. This extended circuit was used to train new classifiers that can distinguish threat from safety cues across different paradigms. The study found that the extended circuit and the newly trained classifiers can robustly decode threat and safety cues using activation patterns within and beyond the 'threat circuit'. The results suggest that the neural decoders are sensitive and specific in distinguishing threat from safety cues across multiple datasets. The study also highlights the importance of considering multiple neural systems for a comprehensive understanding of threat processing. The findings support the idea that threat processing involves multiple processes encompassing multiple brain circuits. The study also shows that the neural representations for conditioned threat are distinct from those evoked by intrinsically salient stimuli. The results provide compelling empirical evidence for conceptual frameworks that threat processing involves multiple processes encompassing multiple brain circuits. The study also highlights the value of other nodes that are important in conditioned threat processing. The results suggest that multiple sensory and cognitive systems are engaged during the encounter of threat and safety cues, which is likely needed for the processing of these complex representations.