Searching on the internet has got a considerable part of our everyday activity and studying the laws of this process with the goal of possible improvement of strategies and data structures is a real challenge. The approach of this project is not algorithmic; instead, we plan to understand how navigation strategies are influenced by cognitive and social biases, by the learning processes of individuals, their characteristics like age, gender, and cultural background. We will study the dynamics of the searches as recorded by clickstreams, which will be a tool to map out the complex information landscape. This landscape can be represented by a weighted network and it evolves during the learning process as exploration proceeds. A further interesting question will be to explore if the content, like Wikipedia categories of the target of search influence the strategy. We plan to study with similar tools the related problem of search for information, which, in contrast to navigation, has no a priory given target. An example can be the acquisition of news, where the search strategy can be a source of filter bubbles known as one of the origins of increasing political polarisation. We are planning to analyze big datasets from Wikipedia related games (Wiki Game and Wikispeedia), Wikipedia clickstream data, online social network data, and Mechanical Turk based experiments.