In the AF condition, feedback algorithms were predetermined based on cognitive-behavioral theory, focused specifically on behavioral changes from week to week, and suggested behavioral strategies to improve adherence and weight loss. A computer-tailored message was compiled instantaneously based on the weekly diary information. Weekly and average weight losses were compared with expected weight losses corresponding to the week in the program. Praise or feedback to build motivation for self-monitoring was included depending on the frequency of monitoring. Reported calories were compared with individualized goals. If calories were above the recommended level, suggestions for reducing calories were provided. Use of meal replacements was encouraged as one strategy to control caloric intake. A variety of portion-controlled meals were recommended when adherence was low, such as liquid shakes (eg, Slim-Fast), prepackaged frozen entrees (eg, Lean Cuisine [Stouffers, a division of Nestlé] or Weight Watchers' Smart Ones [H. J. Heinz Co]), a variety of meal bars, or other prepackaged foods that met caloric requirements of a low-calorie meal (eg, canned soup). The number of calories expended in physical activity was compared with the weekly goal. When exercise was less than prescribed, strategies for overcoming a variety of barriers were suggested. Finally, a summary compared reported behaviors with weight loss progress; provided ongoing support, praise, or motivation; and suggested next steps. In the HC condition, participants were randomized to 1 of 5 human e-counselors with behavioral weight loss experience and degrees in nutrition, exercise physiology, psychology, or health education. All human e-counselors were blinded to the algorithms used to program the AF counselor and were trained and supervised using procedures from other behavioral e-counseling weight loss trials.1,2 Although messages were individualized, e-counselors generally considered weekly weight loss compared with overall progress, progress toward behavioral goals, overcoming specific weight loss barriers, motivation, and answers to participants' questions. There was no predefined structure or content for HC e-mails, and counselors prioritized and selected the focus of the communication based on their clinical judgment.