Computational analysis of paintings and preference based on neural visual processing models

2016 Fiscal Year Final Research Report

• Project/Area Number: 15K12042
• Research Category: Grant-in-Aid for Challenging Exploratory Research
• Allocation Type: Multi-year Fund
• Research Field: Cognitive science
• Research Institution: The University of Tokyo
• Principal Investigator: MOTOYOSHI Isamu 東京大学, 大学院総合文化研究科, 准教授 (60447034)
• Research Collaborator: MORI Shiori
• Project Period (FY): 2015-04-01 – 2017-03-31
• Keywords: 視覚情報処理 / 芸術 / 嗜好

Outline of Final Research Achievements

The present study was aimed to investigate how the style of paintings is shaped by the structure of image features contained in paintings, and how the human visual system processes these image features to determine subjective judgments regarding the beauty and ugliness of various images including paintings. Since the other research groups has recently published a similar line of studies about the former style analysis, we focused on examining the latter preference processing via a series of psychophysical and electrophysiological (EEG) experiments, which revealed robust relationships between human preferential judgments and a small set of low-level statistics of texture images. The results suggest the existence of rapid implicit neural processing that utilize image features to directly summon preferential responses.

Free Research Field

実験心理学