While mental work has been shown to favor overconsumption, the present study compared the effect of a cognitive task alone, followed by acute exercise, or performed on a cycling desk, on short-term food intake and appetite in adults. A total of 19 normal-weight adults randomly completed: resting session (CON), 30-minute cognitive task (CT), 30-minute cognitive task followed by a 15-minute high-intensity interval exercise bout (CT-EX), and 30-minute cognitive task performed on a cycling desk (CT-CD). Energy expenditure was estimated (heart rate-workload relationship), and energy intake (EI; ad libitum) and appetite (visual analog scales) were assessed. Energy expenditure was higher in CT-EX (P < .001) compared with the other conditions and in CT-CD compared with CON and CT (P < .01). EI was higher in CON (P < .05) and CT-CD compared with CT (P < .01). Relative EI was higher in CON compared with CT (P < .05) and lower in CT-EX compared with CT, CT-CD, and CON (all Ps < .001). Area under the curve desire to eat was higher in CON compared with CT (P < .05) and CT-EX (P < .01). Area under the curve prospective food consumption was higher in CON compared with CT-EX (P < .01). Overall composite appetite score was not different between conditions. While cycling desks are recommended to break up sedentary time, the induced increase in energy expenditure might not be enough to significantly reduce overall short-term relative EI after mental work.