In this experiment we studied the effect of goal setting on the strategies used to perform a block design task called SAMUEL. SAMUEL can measure many indicators, which are then combined to determine the strategies used by participants when solving SAMUEL problems. Two experimental groups were created: one group was given an explicit, difficult goal and the other was not given a goal. The two groups were comparable in their average visual-spatial ability. The results indicated no goal effect on the strategies, defined in terms of the combined indicators. However, the goal did have an effect on some of the indicators taken alone (total problem-solving time, total viewing time, and model-viewing frequency) but this was true only for subjects with a low cognitive ability. These findings demonstrate that setting a goal can have an effect on some strategy indexes used to assess performance on a visual-intelligence design task. This research has implications for defining intelligence-test instructions and educational requirements in school. (Contains 1 table and 8 figures.)