Research indicates that the ability to perform a task can be affected by the composition of the meal preceding the task. This study investigated the effect of shift workers' consumption of a medium-fat, medium-carbohydrate meal on alertness scores. Six subjects (four men, two women) aged 19 to 44 recorded food intake, sleep, and quality of sleep for two weeks, and measured their body temperature and performed cognitive tests during two night shifts at baseline and in test periods. The Stanford Sleepiness Scale (SSS) was used to quantify sleepiness, and a Paced Auditory Serial Addition Test (PASAT) was used to measure cognitive performance. In comparison with the score at baseline, when subjects had a low-fat, high-carbohydrate dietary intake (1,335 kcal/5,588 kJ, 56% carbohydrate, 28% fat), the 1.6-second PASAT score improved significantly (p=0.042) during night shifts when subjects consumed a test meal (987 kcal/4,131 kJ, 46% carbohydrate, 42% fat). No statistically significant difference in SSS was found between baseline and test periods. The reduced body temperature between 2400 hours and 0530 hours was similar for both baseline and test periods. Meal composition and size during night shifts may affect cognitive performance.