Collective aesthetic cognitivism

Team members: Bahador Bahrami, Doga Gulan, Sepehr Saedpouhr, Arghavan Aslani, Antonio Scarafone

Funding from:

Our visual exploration of any given artwork is individual. Our evaluation of it, however, is primarily collective: background expectations, learned from books, articles, classes and indeed, discussing our experience with others (friends, colleagues, etc.) shapes our understanding. What can we learn from art if we make the visual exploration collective? Cognitive scientists have, in the past, used eye tracking to examine the private, visual exploration of art but it is entirely unknown how the interactive and collective dynamics shape ways of looking to ways of understanding art. Recent technological developments mean that an entire eye-tracking setup can now fit into the frame of conventional glasses meaning it is now possible, for the first time, to track the gaze seamlessly under plausible ecological conditions of navigation and social interaction. Plus, new methods of data handling (cloud storage/computing, machine learning and artificial intelligence) enable us to establish real-time wireless loops between the viewer (i.e., eye tracking device), the cognitive scientist (i.e., the data collection/analysis unit), and the artist by integrating all of this into interactive displays of the same artwork given back to the viewers. This permits creating ‘gaze-contingent’ reactive displays that respond, in real-time, to how the visitors examine the exhibited artwork. In TRT-2021-10480 we do this in groups of interacting individuals (e.g., many visitors visiting an art gallery), each equipped with an eye-tracking device, collecting and analysing the data in parallel and integrating them into an interactive display. Using this data, we test two contrasting hypotheses about the role of social context in distribution of gaze. Our project explores the ways for discovering emergent phenomena that arise from the interaction between visitors and the artwork that would not have been possible if we only examined viewer behaviour individually.