The visual world is full of redundant information. Natural images contain largely the same information across multiple spatial frequency bands and adjacent pixels are strongly correlated. For a system that is concerned with the economical, both in terms of time and resources, processing of vast amounts of data, reducing this data to extract the relevant information is crucial. The work presented in this thesis was to investigate whether and how the human visual system makes use of the statistical regularities of its environment to facilitate the efficient processing of visual stimuli. (PSYE - Sciences psychologiques et de l'éducation) -- UCL, 2020