Contingent valuation is a survey-based method that randomly selects families or individuals as samples. It reveals consumer preferences for public goods and services such as ecological environment resources in a hypothetical market, and infers respondents' willingness to pay (WTP) to improve, for example, environmental quality. The method can also be used to calculate the respondents' WTP (or willingness to accept, WTA) and extend the samples to the whole study region. The average WTP (or WTA) can then be used to obtain the economic benefits or losses brought about by a planning project. A great variety of questionnaire formats have been developed and applied in practice, among which the Dichotomous Choice CVM is considered one of the most advanced methods nowadays. With the continuous development of CVM, it is important to analyze what factors influence WTP in the related empirical studies. Different scholars have reached an agreement that the respondents' willingness to pay for environmental improvement is closely related to the distance between the evaluation objects and the environmental resources. To a certain extent, the distance between the respondents and the evaluation objects can adjust demand for environmental goods as an alternative of price mechanism. On the one hand, the distance factor can influence the respondents' awareness level on environmental goods, which is to produce environmental preferences through affecting the information validity and the accessibility; on the other hand, it can also affect the possibility of usability and substitutability of environmental goods. Logically, the distance should have negative effects on affecting respondents' WTP in a given region in the assumption. The farther the distance of respondents' live from the assessment object, the less possible improvement and protection WTP for environmental goods will have. This is the "Distance Decay Effect". Based on the study on WTP distance-decay, this paper, combining the Fishbein theory with the contingent valuation method, assumes that individuals' cognition and attitude towards the goods are imbalanced and the willingness to pay (WTP) of respondents in different space exists difference. The random utility model is applied to build the Double-bounded Dichotomous Choice CVM Data Analysis Model,and establishes the influential factor of WTP. Taking Sanjiang plain wetland as the application object, the samples are divided into core zone, radiation zone and peripheral zone by using double-bounded dichotomous contingent valuation technique and discussing the payment ability, payment willing and their influential factors so as to establish the spatial differentiation model of the willingness to pay to the non-use value WTP based on the cognition and attitude. Respectively, the average WTP of residents is: 142.23, 105.01 and 77.62 RMB per year, exhibiting a step-decreasing trend, which means the distance, cognition and the WTP are closely related. Research results show that, involving the individuals' cognition in WTP model, the effectiveness of CVM can be improved. The findings will provide useful references for government to make related ecological policies and the conclusion of the study will lay a foundation and provide reference for the government policy and decision making.