The role of attention in semantic processing of visual scenes: a research-based overview

The role of attention in visual processing is a complex and commonly studied issue. One of the ways to conduct investigations concerning its mechanisms is by using real-world scenes with a clearly defined hierarchical structure. Furthermore, comparisons of differences in processing of semantically congruent (i.e. containing object expected in a given context provided by scene background) and semantically incongruent scenes (i.e. containing object unexpected in a given context) provides a basis for in-depth probing of the attention role. This thesis presents two studies carried out with the use of real-world scenes. Study 1 investigated whether semantically incongruent scenes capture attention, while Study 2 analyzed temporal hierarchy of scenes semantics processing when attention was focused. Results contributed to clarify the role of attention in the visual processing of semantically incongruent stimuli. Although attention is not automatically attracted by semantic incongruencies, underlying recognition mechanisms may be modulated, thus supporting the idea of multiple interactive structure for visual processing.