When Paintings Change Time: Visual Structure and Circadian Context Shape Subjective Duration

Abstract Brief interval timing is typically studied with simple, uniform stimuli, leaving open how naturalistic visual input and circadian and seasonal factors shape perceived duration. We used a temporal-bisection task in which participants judged whether each image duration was closer to a learned short or long standard. Stimuli comprised 25 images derived from five Van Gogh paintings: five originals and 20 hue-shifted variants created by shifting each painting in four chromatic directions. We tested whether perceived duration varied across chromatic conditions and whether continuous image-statistics dimensions and circadian and seasonal factors predicted variation across paintings and observers. At the condition level, blue-shifted images were judged as lasting slightly longer than neutral and green-shifted images, as indexed by lower points of subjective equality, although these effects were modest and condition means did not differ from the objective midpoint. At the item level, perceived duration varied more strongly across paintings: the difference between the paintings with the shortest and longest durations was approximately 22 ms. This heterogeneity was predicted by three image-statistics components reflecting brightness and oriented texture, chromaticity–luminance dispersion, and cool structural complexity, together explaining 59% of painting-level variance. A participant-by-painting mixed-effects model confirmed these effects and showed that chronotype and day length independently predicted perceived duration. In this sample, the observer-level effects of chronotype and day length were larger than the image-statistics effects. These findings indicate that subjective duration under naturalistic viewing depends on multidimensional visual structure and individual circadian and seasonal context, rather than on chromatic condition alone.