Fish usually alter their swimming behavior to adapt to changes in food resources in their natural habitats. In this study, we investigated the effects of nutritional status on the preferred swimming behavior of the juvenile crucian carp (). Three experimental treatment groups with different nutritional status were established: the control group (fasted for 48 h), the feeding group (fed to satiation), and the starvation group (starved for 14 days). The experimental fish were videoed individually with a hand-made device for determining the preferred water velocity: a 1.5 m long conical raceway in which the water speed gradually increased from 11.86 cm/s to 65.45 cm/s at (25±1)℃. The water velocity of the 1st, 2nd, 3rd, 4th and 5th waster velocity intervals were 11.86-15.18 cm/s, 15.18-20.12 cm/s, 20.12-27.91 cm/s, 27.91-41.30 cm/s, 41.30-65.45 cm/s, respectively. The videos were analyzed with the Ethovision XT9 software, and several parameters were calculated in each water velocity interval: frequency(), and the percentage of time stayed( increased as the water velocity increased for both the control and feeding groups. showed no significant variation across different water velocity intervals in the fasting group, whereas the feeding group displayed the longest t, was significantly higher in the third and fourth water velocity intervals than in the other intervals in both the control and feeding groups, and the preferred swimming speeds of both the control and feeding groups were 20.12-41.30 cm/s. Two distinctly different preferred swimming behaviors were apparent in the starved group, which we arbitrarily designated the type I and type II preferred swimming behaviors. The type-I fish showed the largest values in the first water velocity interval, which were significantly higher than those in all other intervals. However, the of these fish showed extremely low values across all water velocity intervals, and was much lower than those of the other groups. Therefore, the preferred swimming speed of the type I fish from the starvation group was 11.86-15.18 cm/s. However, of the type II fish were similar to those of the control group, i.e., , of the type II fish was significantly higher in the third and fourth water velocity intervals than in the other intervals, and was significantly higher in the fourth water velocity interval than in the first, second, or fifth, but was not higher than that in the third water velocity interval. Consequently, the preferred swimming speed of the type II fish from the starvation group was 20.12-41.30 cm/s. This study shows that satiation feeding does not affect the preferred swimming speed of the crucian carp. However, during digestion, the crucian carp showed fewer traverses across different water speed intervals, but a longer period of stay each time they arrived at a given interval. Starvation may enhance the interindividual variations in energy reserves and hence the maintenance of swimming capacity, resulting in two distinctly different types of preferred swimming behavior in the fasting crucian carp. The type I fish showed a profoundly reduced preferred swimming speed, whereas the preferred swimming speed of the type II fish did not differ from that of the normally fed fish.