AUTHOR=Gardezi Maham , Fung King Hei , Baig Usman Mirza , Ismail Mariam , Kadosh Oren , Bonneh Yoram S. , Sheth Bhavin R. 

TITLE=What Makes an Image Interesting and How Can We Explain It

JOURNAL=Frontiers in Psychology

VOLUME=Volume 12 - 2021

YEAR=2021

URL=https://www.frontiersin.org/journals/psychology/articles/10.3389/fpsyg.2021.668651

DOI=10.3389/fpsyg.2021.668651

ISSN=1664-1078

ABSTRACT=Here, we explore the question: What makes a photograph interesting? Answering this question deepens our understanding of human visual cognition and knowledge gained can prove useful in understanding how to reliably disseminate information. Observers viewed images belonging to different categories, which cover a wide, representative spectrum of real-world scenes, in a self-paced manner; observers judged the image's interestingness at trial's end. Our studies revealed the following: landscapes were the most interesting of all categories tested, followed by scenes with people and cityscapes, followed still by aerial scenes, with indoor scenes of homes and offices being least interesting. Interestingness is an intrinsic property of the image, not one of a specific experimental condition: relative judgment tasks with image pairs, fixed viewing duration, or changes in viewing history all failed to affect relative interestingness of image category. Low-level accounts based on computational image complexity, color, or viewing time failed to explain image interestingness: more interesting images were not viewed for longer. On the other hand, a single higher-order variable, namely image uprightness, significantly improved models of average interest. Observers' eye movements partially predicted overall average interest: a regression model with number of fixations, mean fixation duration, and a custom measure of novel fixations explained 37% of variance. Our research revealed a clearly hierarchy of image category for interestingness that low-level accounts based on image complexity and color combined cannot explain but eye movements, in part, can.