Themed scenic spots, as a new extension of theme parks, have been developed fast in China these years. Having attracted numerous investment, themed scenic spots saw a large increase in number and scale but suffer from low profit. To improve the situation, it is very necessary to evaluate tourist satisfaction more scientifically and reasonably. Traditionally, multiple regression models are used to identify key destination attributes which the destination authorities should invest in to strengthen tourist overall satisfaction. The underlying assumption of multiple regression models is that there is linear and symmetric relationship between attribute- level performance and overall satisfaction. However, this assumption is problematic. Recently, the asymmetric relationships between attribute- level and overall customer satisfaction and the classification of product or service attributes in terms of these relationships have been revealed in many research areas, and some studies have focused on tourism-related phenomena. But the number of studies relating to the attribute- level performance of themed scenic spots and their influences on overall tourist satisfaction is limited. The theoretical assumption of asymmetric relationships between attribute-level performance and satisfaction shows the direction of refinement and extension of the relatively stagnant tourist satisfaction research. As the use of linear assumption in satisfaction evaluation is questioned, this paper, taking typical historical and cultural themed scenic spots (including Tang Paradise in Xi'an, Song Dynasty Town in Hangzhou, Lingshan in Wuxi and Mystical Yancheng Chun-Qiu Dream in Changzhou) as examples, examines the asymmetric effects of attribute- level performance on tourist overall satisfaction using dummy variable regression. The empirical results mainly include four aspects: (1) The attributes of historical and cultural themed scenic spots are divided into 7 dimensions, including landscape environment, entertainment activities, experience in scenic spots, service facilities, service quality, theme display and consumption prices. (2) For historical and cultural themed scenic spots, the asymmetric effect of attribute- level performance on tourist overall satisfaction is confirmed, and the asymmetrical effect is not equivalent for each attribute. Only the effect of two attributes, landscape environment and experience in scenic spots, on tourist satisfaction is linear and symmetric, and these two attributes are classified as performance factors of tourist overall satisfaction. Three attributes, entertainment activities, service facilities and theme display, have showed negative asymmetric effect and they belong to basic factors. One attribute, service quality, has exhibited positive asymmetric effect on tourist satisfaction and it is excitement factors. Besides, consumption prices also have positive asymmetric effect, but this attribute should be classified as basic factors actually. (3) The paper proposes a three- factor structure model of tourist satisfaction for historical and cultural themed scenic spots. This model not only includes the relationships between the attribute-level performance of themed scenic spots and tourist satisfaction, but also reveals the corresponding levels of tourist demand of each attribute and the market significance of each attribute. Entertainment activities, service facilities, theme display and consumption prices, which are entirely expected by tourists and are tourists' minimum requirements for such scenic spot marketing, establish a market entry "threshold". Landscape environment and experience in scenic spots are directly connected to tourists' explicit needs and desires, and a themed scenic spot should be competitive with regard to these two attributes. Service quality is unexpected and may surprise the tourists, so this attribute is the origin of competitive advantage for themed scenic spots in the future. Finally, the landscape environment performance must be maintained and improved as much as possible to ensure the core competitiveness of scenic spots. All in all, the focal points of enhancing tourist overall satisfaction is that attribute- level performances should be optimized, not simply maximized. [ABSTRACT FROM AUTHOR]