Theoretical investigations into the impact that predators exert on prey species suggest that two parameters, growth rate and mortality rate, should be the most influential in determining when animals should risk exposure to a predator in order to achieve higher feeding rates. While these two parameters have usually been assumed to be environmentally determined, we used thyroid hormone (3,3',5-triiodo-L-thyronine (T-3)) to manipulate growth rates and examine the behavioural consequences associated with these manipulations. In two experiments, we examined how the growth rate of fathead minnows (Pimephales promelas) is affected by treatment with T-3, and used the results from this experiment to make a priori predictions about their relative willingness to risk exposure to a predator in order to receive increased feeding rates. The first experiment demonstrated that T-3 significantly reduced the growth rates of fathead minnows compared with an unmanipulated control. When groups were compared in their relative willingness to risk exposure to a predator, manipulated growth rates in the first experiment were an accurate predictor of behaviour; groups with relatively high growth rates were more willing to risk exposure to a predator. These results are consistent with the theoretical expectation that growth rates should be an important factor determining decisions that involve trade-offs.