This paper describes the development, implementation, and effects of new approaches to cognitive assessment within an undergraduate engineering course at the U.S. Air Force Academy. The course, ENGR 110, "Introduction to Engineering," was designed to be a problem-based learning environment in which cadets worked in teams to solve problems integral to a "Mission to Mars," for example, getting to Mars, constructing a research site on Mars, and developing a renewable power source there. In addition to traditional knowledge and skill objectives, the course was focused on "higher order" outcomes such as "framing and resolving ill-defined problems,""communicating via multiple media,""exhibiting intellectual curiosity," and "developing a rich conceptualization of engineering." Forty-two freshman cadets participated in the pilot testing of the cognitive assessment methods used to assess student achievement with respect to these outcomes. Assessments included: (1) reflective judgment exercises; (2) self-assessment questionnaires; (3) concept maps; (4) focus groups; (5) e-mail journals; (6) observations; and (7) individual and group interviews with faculty. Results indicated statistically and educationally significant differences in "problem solving" between ENGR 110 students and two control classes of sophomore engineering students. Other results supported research and development. (Contains 2 tables, 2 figures, and 14 references.) (Author/SLD)