Although it has been known for over 40 years that people tend to adopt similar communication styles with both infants and dogs, the visual component of infant-directed speech (motherese) has been overlooked in research. The aim of the current study was to investigate variations in visual prosodic features across different addressees (i.e. dogs, infants, and adults) and the intended goal of communication. Parents who were also dog owners interacted with their infant (6–18 m.o.), their family dog and their spouses as adult-directed control in three different, 1–2-minute-long scenarios: language tutoring, attention-getting, and fixed sentences. We comparatively analysed speakers’ (N=46; 23 females, Mage= 34.41 years, SD=4.41) facial expressions (i.e., happy, surprised, neutral) and emotional states (valence, arousal) across scenarios and conditions using Noldus FaceReader software. Results revealed a significant interaction effect of condition and situation. As expected, visual prosody was the least intense in the dog-directed condition, with language tutoring showing the lowest intensity. Fixed sentences, with lexical content intended for infants (i.e. nursery rhyme), exhibited consistently the highest intensity. Interestingly, adult-directed prosody toward spouses resembled infant-directed prosody more than expected. In summary, our results highlight disparities in the interpretation of facial expressions in both con- and interspecies communication. Despite the study’s limitations, it emphasizes the importance of ongoing visual prosody characterisation, inviting further clarification and reconsideration of the dynamics between the visual and acoustic aspects of motherese and dog-directed talk as well as the role of bonding between interactants. These insights underscore the benefits of comparative cognition involving canines.