In recent years, the concepts of creativity and art production intended as a prerogative of human beings have been challenged by the rise of Artificial Intelligence (AI). In light of the advancement of AI applications in the art field, we investigated whether the beholder’s response to an artwork is affected by the a priori knowledge of AI authorship. 

In order to answer this question, in 2019 we conducted an experiment in the ecological context of the art fair ArtVerona by exhibiting two abstract paintings, both created by

a human artist, yet declaring to visitors, before showing them the artworks, that one was made by a human being and one by an AI. While looking at the paintings, participants' implicit activation (heartbeat and skin conductance) was measured, while after having looked at each painting, they explicitly indicated how much they liked the artwork.

The results of the experiment showed that the pleasantness ratings were lower for the artwork declared as AI-made,

and also that the participants' physiological responses

were influenced by the authoriality manipulation.

The results of this study show that there is a negative bias against artworks associated with AI. Furthermore,

they corroborate the importance of the role of top-down processing in the relationship with an artwork and implement those of other researchers and scholars who had already suggested that a priori contextual information are key factors able to influence the aesthetic experience.