This article attempts to further our understanding of alexithymia by testing two conceptual questions about the construct: (a) Is alexithymia characterized by reduced autonomic activity? and (b) Can it be clearly distinguished from defensiveness? Eighty healthy university students completed a battery of personality scales including the Toronto Alexithymia Scale, measures of self-deception and impression management, depression, and anger-in. They also participated in three lab stress tasks: isometric handgrip; mental arithmetic; and a negative affect provocation task. Blood pressure and heart rate were monitored throughout the lab procedure. Analyses were conducted with tercile groups of low, medium, and high alexithymia scorers. The "high alexithymia" tercile showed smaller heart rate responses to the stress tasks and more anger-in behavior. Blood pressure responses did not differentiate the low/ medium/high alexithymia subgroups. Alexithymia scores were unrelated to defensiveness, that is, there was no relationship between alexithymia and impression management or self-deception, and alexithymia was unrelated to depression. We conclude that students defined as "high alexithymia" on the Toronto Alexithymia Scale are not self-deceptive nor do they try to leave a particular impression; they tend to be somewhat hypoaroused autonomically, and they report as many psychological distress symptoms as do subjects with lower TAS scores.