Chinese cognition and conceptual system is built on metaphors. Metaphor comprehension is an important task in natural language processing. Based on the interaction theory, the cooperative network model (CNM) is proposed, which better conforms to the character of metaphor, to perform metaphor comprehension. The CNM reveals the semantic relations between target and source domain, i.e., seeking common ground while reserving differences. Besides, the CNM addresses bidirectional semantic relations between concepts. By computing cooperative strength of relations, the CNM obtains the meaning of metaphorical context. The experimental results demonstrate the effectiveness of the CNM. The outputs of the CNM preliminarily reflect the dynamic and prominent mechanisms of metaphor comprehension.