This study is the third in a series of studies exploring environmental components that predict the likelihood of psychological restoration. The focus in this research is on small urban parks and open spaces. These spaces presumably will become increasingly important as settings for restoration as demands for densification of cities increases. Small parks provide opportunities for people in cities to sit down and rest for a little while near work or home. They function as spaces where they can get away from daily demands both mentally and physically as well as become fascinated by the greenery and other features in the park. As concluded in our earlier studies, the possibility afforded for restoration is not only a matter of the size of an available park or open space, but also a matter of its design and the components used to create it. By referring to specific components in the environment, rather than broad categories of natural versus built, as in much research on restorative environments, this study helps to fill a gap in the empirical literature concerning the application of restorative environments theory. A focus on the specific components in parks and open spaces is highly relevant from a design and planning perspective; it gives professionals more concrete information about design that can be applied in practice. In our earlier studies, we found that the environmental components most predictive of the likelihood of restoration were the percentage of ground surface covered by grass and the amount of trees and bushes visible from the given viewing point. That earlier work focused on visual aspects of small parks and open spaces, using photographs as the media for presentation to research subjects. In the present study, we instead present people with words that describe different sets of possible components in a small urban park or open space. By presenting them with words instead of photos, the approach relies on their ability to imagine the different alternatives by referring to their own experiences. In using descriptive words as the media of presentation, we apply a common standard procedure in choice-based conjoint analysis, a methodological approach that enables examination of preferences for various combinations of features. The method involves presenting subjects with pairs of park alternatives, which differ in the levels of different components. Conjoint analysis has mainly been used in marketing research in the development of new products. In our case, the park is the “product,” elaborated in terms of different park components. Statistically, conjoint analysis enables examination of more variables and combinations of variables than possible in simple regression analysis. The method also provides a setup that reminds of real world choices; it creates a realistic choice situation that people can relate to. The components assessed in this study are grass, bushes, trees, flower beds, water, and the number of other people in the park. Each component is presented at one of three levels in a given park alternative (e.g., no trees, a few trees, many trees). Given a pair of alternatives, the respondent’s task is to choose the one that is best for them. The matter of “best” follows from a scenario that frames the choice task; the subject is to imagine being in need of rest after a period of intense mental work and looking for a place to rest for a while. The study is web-based, with access provided to residents of Oslo, Norway, who experience the urban environment on a daily basis and are familiar with the type of small urban parks and open spaces in focus. Our presentation will cover the initial results from this study, which is now underway. The present application of conjoint methodology breaks new ground for quantitative research in landscape architecture and environmental psychology.