Abstract
A painted portrait differs from a photo in that the artist intentionally selects only certain regions for fine detail while leaving other areas less well defined. Although artists and art critics have claimed that these choices guide the viewer's gaze, this claim has not been thoroughly tested. Past studies involving the viewing of original artwork are confounded by unsystematic variation of the regions of fine and coarse detail (e.g., whether they are foreground or background). Here we monitored the gaze of participants viewing original photos and paintings of the same model posing as one of Rembrandt's subjects (e.g., Self Portrait with Beret, 1659). The paintings were rendered with a non-photorealistic technique designed to mimic Rembrandt's painting style (Di Paolo, 2007). Each painting contained four regions of interest in which the level of detail was systematically varied: left versus right eye region in finer detail and left versus right collar region in finer detail. Both original and mirror image views were tested to control for side biases. Participants viewed each portrait along with many other portraits that encompassed a wide range of artistic styles, creating a context in which participants could compare and rate the portraits for “artistic merit.” Analyses revealed that overall fewer fixations were made when viewing paintings than photos, and viewers' gaze was attracted to and held longer by an eye region in the portrait rendered in finer detail. Even regions of paintings that were rarely fixated (i.e., collar regions below the face) nevertheless guided gaze by enhancing the salience of a finely detailed eye on the same side of the portrait as a coarsely detailed collar region. This implies that Rembrandt and other portraitists incorporate an implicit understanding of how gaze is directed by relative detail.