The role of monocular cues in dogs perception of three-dimensionality

Understanding depth perception in dogs is crucial for interpreting their behavior, optimizing their environments, and advancing comparative visual cognition. A previous study by our group showed that dogs can use a combination of linear perspective and shading to infer tridimensionality from pictorial cues in two-dimensional images. Here we examined the individual contribution of the same two monocular, pictorial cues, in three experiments, involving a total of 84 dogs (46 females, 38 males; mean age ± SD = 4.2 ± 2.1 years).Using a violation of expectation paradigm, we measured dogs looking time – as a proxy for surprise – after observing a ball rolling onto an apparatus and then disappearing into a real hole or rolling over a fake hole, where either perspective lines or shading mimicked those of the real hole. Linear perspective alone did not elicit surprise, suggesting it is insufficient to elicit depth perception (Wald Chi-square = 2.012, p = 0.152), nor did low contrast shading (Wald Chi-square = 0.977, p = 0.442). However, high contrast shading successfully induced depth perception (Wald Chi-square = 7.405, p = 0.006), with longer looking times in the illusory condition (estimated mean ± std. error: 8.12 ± 1.22 s) compared to the real condition (6.11 ± 0.97 s).Results suggest that dogs can perceive depth from 2D images, but the effectiveness depends on the specific pictorial cues used. Shading appears to be a more salient depth cue for dogs compared to linear perspective when presented individually. These findings have implications for understanding visual processing in dogs and optimizing visual stimulus presentation in canine cognitive tests. Further research is needed to elucidate the mechanisms and ecological relevance of different pictorial depth cues in canine vision, as well as to explore the integration of multiple cues in depth perception.