Modality styles are defined as behavioral predispositions characterized by a certain mode or set of modes that an individual habitually uses. They are reflective of higher-level orientations, or lifestyles, that are hypothesized to influence all dimensions of an individual’s travel and activity behavior. The objectives of this paper are to understand and quantify different modality styles, and to show how the modality styles construct can be operationalized within the context of traditional travel demand models. We employ the six-week MOBIDRIVE travel diary and estimate behavioral mixture models in which the modality style provides a behavioral rationale to the way in which unobserved heterogeneity is specified in the travel model. Our analysis consists of two stages: First, we explore the presence and types of modality styles suggested by the data through the means of a descriptive analysis. Next, we develop a model that captures the influence of modality styles on two dimensions of an individual’s travel behavior: mode choice for work tours and mode choice for non-work tours. The modality styles are specified as latent classes; heterogeneity across modality styles include both the modes considered (choice set) and the values of taste parameters. The modality style of an individual then influences all of his/her mode choice decisions for work-tours and non-work tours. In addition, error components capture unobserved correlation between alternatives and across mode choice decisions made by the same individual. Our results from both stages indicate the presence of “quasi-unimodal auto” individuals who display a strong bias for using the automobile and “multimodal” individuals who exhibit variation in their modal preferences. Multimodal behavior is further distinguished by those whose auto use is minimal at best (termed “multimodal green”) and those who display significant auto use (“multimodal all”). These three modality styles comprise roughly equal segments within the sample population. The behavioral mixtures model provides a better fit to the data than other modeling approaches, such as random parameters. Moreover, by providing a behavioral underpinning to unobserved heterogeneity, it provides a richer and more robust framework for interpretation and policy analysis.