It is a widespread phenomenon that individuals from the same population will differ in their speed and ability to learn and yet very few studies looked at why such variation occurs. One hypothesis is that individuals prioritise new environmental cues (current information) and previous experiences (old information) differently. For example, those that favour old information, may be slower to adapt to current information and struggle to repeat success on a task when the context is changed. To test this hypothesis, we presented the three-spined stickleback (Gasterosteus aculeatus) with a two-choice associative learning task. The correct chamber was randomised so that success was dependent on using an ecologically relevant landmark as an indicator of food location. Fish were given 45 trials to make 8 out of 10 correct decisions in a row (i.e. learning criterion) before they were presented with a second learning task, which was identical to the first except with a new landmark. To evaluate learning performance, we looked at number of trials required to reach criterion and time to find food. To test whether information use correlate with learning success, we analysed how food location and choice from previous trials affect the most recent choice and how this varies over time and across tasks between fast and slow learners. Our preliminary results suggest over time all fish made quicker decisions and find food faster. In task 2, the majority of fish reached criterion faster and showed less variation in performance suggesting information transfer from task 1 to task 2. The goal of this research is to better understand whether learning style affects performance; individual’s ability to acquire new information and adapt to new environments.