Posttraumatic stress disorder (PTSD) is the most common psychological problem after experiencing traumatic events. Based on the latent variable model, a large number of traditional psychopathological studies on PTSD were conducted to investigate the potential symptom dimension structure of PTSD by means of factor analysis, and to explore the influencing factors of these different symptom dimensions and their correlation with other mental health factors. With the introduction of psychopathology network theory into the field of mental health research, there have been a lot of researches on PTSD symptom network in recent years. These studies focus on the network structure formed by the dynamic correlation between different symptoms of PTSD. The two approaches complement each other in content, but they also have many methodological limitations. In theory, latent variable model and network model are based on different hypotheses about mental disorder, but they are not mutually exclusive. There is a lot of evidence that for complex mental disorders such as PTSD, the underlying symptom dimensions and the network structure of interaction between symptom dimensions may exist at the same time. The investigation of the latent symptom dimension network of PTSD can combine the two perspectives, make up for the deficiency of their methodology, and provide the basis for a more comprehensive understanding of the psychological and pathological characteristics of PTSD. < p > < p > on the basis of previous studies on the structure of latent symptom dimensions of PTSD in adult earthquake survivors, Wang Li research group of Key Laboratory of mental health, Chinese Academy of Sciences, explored the network structure of these symptom dimensions, estimated the latent variable network model among 1196 adult earthquake survivors, and repeated samples in a child and adolescent with mixed trauma types The estimated symptom dimension network of PTSD was verified. The results show that the latent variable network model has better model fitting in both samples than the traditional latent variable model. In the latent symptom dimension network of PTSD, the strongest symptom association exists between intrusion and avoidance dimensions. There was also a strong symptom association between the dimensions of externalization behavior and negative emotion, anxiety arousal and mental distress arousal. According to the centrality of each symptom dimension, externalization behavior has the highest centrality in the latent symptom dimension network of PTSD. < p > < p > this study revealed the network structure of latent symptom dimension of PTSD. The results support the existence of latent symptom dimension structure and causal network structure among symptom dimensions, suggesting that more attention should be paid to dynamic interaction at symptom dimension level rather than single symptom level in future research. The results also suggest that more attention should be paid to the association between specific symptom dimensions, such as the association between intrusion and avoidance symptoms, in the intervention of PTSD. The relevant results were published in the European Journal of psychology. The research is supported by the National Natural Science Foundation of China, the major project of the key research base of Humanities and social sciences of the Ministry of education, the key foreign cooperation program of the Chinese Academy of Sciences and the Beijing Science and Technology Commission.