Despite concerns about generalizability, past mock trial research has concluded that effects of sample (i.e., students versus representative mock jurors) are negligible. The current study was conducted to explore this conclusion within the conceptual framework of cognitive–experiential self-theory (CEST). Through a mock civil commitment hearing of a sexually violent predator, responses of student (n = 138) and representative (n = 240) mock jurors were compared. Results revealed several important differences between samples: (a) the student sample scored higher on the rational processing measure (i.e., need for cognition); (b) students' verdicts were also significantly correlated to a measure of their cognitive processing style, an enduring personal characteristic related to the extent to which an individual engages in either effortful/effortless cognition; and (c) the representative sample was more punitive, was more persuaded by clinical expert testimony, and evidenced a greater gender effect in its decisions. Implications for jury decision-making research are discussed. Copyright © 2009 John Wiley & Sons, Ltd.