Autism spectrum disorder (ASD) involves impaired social behavior and atypical face processing. Hundreds of risk genes for ASD have been identified. However, it is still unknown if mutations in a particular gene could result in atypical face processing like that of ASD. Mutations in the SH3 and multiple ankyrin repeat domains 3 (Shank3) gene, which are linked to ASD, have been shown to cause social problems in animals though facial recognition impairments have not previously been reported.
Domestic dogs (Canis familiaris) developed remarkable face-processing skills. They could serve as useful animal models for researching the genetic correlations of abnormal face processing like ASD. A recent animal study published in Science Advances examined whether dogs with Shank3 mutations produced by the clustered regularly interspaced short palindromic repeats–CRISPR-associated protein 9 (CRISPR-Cas9) editing method would exhibit facial processing deficiencies similar to those seen in ASD.
A total of 23 wild-type (WT) control beagles and 15 Shank3 mutant beagles were included in this study. There was no significant difference in mean age between the two groups. The mean age of the WT group was found to be 19.3 ± 1.33 months, and the mutant group was 24.5 ± 2.75 months (Mann–Whitney U test [U] = 125.5, P = 0.164).
Photographs of 16 beagle faces, 8 poodle faces, 16 autos, 16 human faces, 16 residences, and 16 chimpanzee faces were used as the visual stimuli in this study. All stimuli were shown in black and white to account for any possible effects of color disparities between photographs of houses and faces. Various analysis techniques were used in this study, including eye-tracking data analysis, behavioral analysis, and electrocorticogram (ECoG) data analysis.
The results demonstrated that WT controls examined facial stimuli significantly more frequently compared to a 50% chance level with a mean of 65.2 ± 2.55%, t(15) = 5.988, P < 0.001, whereas Shank3 mutants examined less frequently compared to chance with a mean of 38.8 ± 2.64%, t(13) = 4.235, P = 0.001. These findings indicate that WT controls and mutants exhibited opposite behavioral patterns. It was also reported that neither mutants nor controls approached houses at a rate significantly different from the 50% chance (mutants: 50.6 ± 2.59%, P = 0.938; WT: 48.4 ± 3.28%, P = 0.766). The overall amount of time spent looking at the screen was similar for controls and mutants (2.0 ± 0.08 s versus 1.9 ± 0.09 s) with U = 6 and P = 0.111.
In the non-parametric Wilcoxon test, WT controls had higher gaze fixations on faces compared to houses with Z = −2.023 and P = 0.031, whereas the reverse pattern was observed in mutants who showed greater fixations on houses compared to faces with Z = -2.023 and P = 0.031. Moreover, mutant dogs exhibited shorter duration of gaze movements (44.9 ± 8.16% versus 21.5 ± 3.72%, U = 3, P = 0.028) and fewer fixations in the eye region of the face stimuli (43.1 ± 7.80% versus 19.9 ± 3.15%, U = 3, P = 0.028) compared to WT dogs.
Mutants’ face-specific neuronal process was slower and less developed in the N1 time window compared to WT beagles. Controls exhibited significantly more gaze fixations on beagle faces compared to husky faces with Z = −2.023 and P = 0.031, while this difference was not observed in mutants with Z = −1.214 and P = 0.156. In ECoG signals analysis, mutant dogs exhibited abnormalities in the brain processes included in breed or species-specific facial classification.
In conclusion, dogs with Shank3 mutations exhibit atypical facial processing and offer a helpful animal model for researching the mechanisms and therapies of ASD.
Reference: Yuan S, Pang C, Wu L, et al. Autism-like atypical face processing in Shank3 mutant dogs. Sci Adv. 2025;11(14):eadu3793. doi:10.1126/sciadv.adu3793
