The AAT results showed a stronger approach tendency for forests compared to houses, but no significant difference between forests and parks. Could the authors discuss whether the spatial depth of forest pictures (compared to the enclosed walls of house pictures) might have influenced the approach-avoidance tendencies?
