This study aims to advance our conceptual understanding of selection procedures by exploring the effect of response fidelity (i.e., written constructed response vs. behavioral constructed response) on test performance, validity, and applicant perceptions. Stimulus fidelity (multimedia stimulus) was kept constant. In a field experiment, 208 applicants for entry-level police officer jobs completed a multimedia situational judgment test with written constructed responses and behavioral responses. We hypothesized the behavioral response mode (a) to be a better predictor of police trainee performance one year later, (b) to be less cognitively saturated, (c) to exhibit higher personality (extraversion) saturation, and (d) to be perceived more positively in terms of media richness. Results suggested support for these hypotheses, although most effect sizes were not large. Implications for a building block approach to personnel selection procedures are discussed.