Active transportation –cycling and biking– not only are sustainable travel modes with zero environmental impact, but also have associated health benefits. However, in comparison with motorized transportation, the motives underlying demand for active transportation –especially beyond recreational purposes– is poorly understood, especially because the standard tradeoff between travel time and cost does not apply to active modes (as it is virtually free and usually takes longer). We propose to further investigate the factors that explain demand for active transportation, including non-instrumental attributes, nonstandard observed attributes (e.g. calories burned), and extended decision rules. To integrate non-instrumental attributes (attitudes and perceptions) we will use extensions of the hybrid choice models (HCM) and a structural model. In particular, this line of research is to extend our previous work, where we used a hybrid choice model with non-instrumental variables that not only enter utility but also inform assignment to latent classes. Using a discrete choice experiment we analyzed the effects of weather (temperature, rain, and snow), cycling time, slope, cycling facilities (bike lanes), and traffic on cycling decisions by members of Cornell University (in an area with cold and snowy winters and hilly topography). We showed that cyclists can be separated into two segments based on a latent factor that summarizes cycling skills and experience. Specifically, cyclists with more skills and experience are less affected by adverse weather conditions. We envision to extend the hybrid choice model to a specification with a semi-parametric representation of how preferences for cycling vary in the population. Semiparametric models are attractive because they don’t impose any specific shape to preferences, instead preference distributions are revealed by actual behavioral responses.