In most visual search experiments, target and distractor stimuli are presented in a static, non-overlapping, high-contrast, two-dimensional configuration from an aerial perspective. The real world is usually far from this ideal, which often makes it a little more difficult for us to find the things we are looking for. This pre-data poster describes an experiment in which we will transfer the location probability learning (LPL) paradigm to a more naturalistic context. Participants will complete a classic visual search task under two conditions: (1) A standard visual search task with two-dimensional stimuli on a computer monitor, and (2) an interactive visual search task with three-dimensional Lego bricks in an assortment box. In both conditions, target stimuli will appear with a probability of 50 percent in one quadrant and with 16.7 percent in one of the other three quadrants without the participants' knowledge. We will compare reaction times between high and low probability target locations to find out whether participants implicitly learn this probability bias and adjust their search strategy accordingly. We are also interested in whether there are transfer effects from the standard visual search task to the interactive visual search tasks, and vice versa. We welcome any feedback and look forward to discussing the planned project with you.