Mice produce ultrasonic vocalizations featuring a variety of syllables. Vocalizations are observed during social interactions. In particular, males produce numerous syllables during courtship. Previous studies have shown that vocalizations change according to sexual behavior, suggesting that males vary their vocalizations depending on the phase of the courtship sequence. To examine this process, we recorded large sets of mouse vocalizations during male–female interactions and acoustically categorized these sounds into 12 vocal types. We found that males emitted predominantly short syllables during the first minute of interaction, more long syllables in the later phases, and mainly harmonic sounds during mounting. These context- and time-dependent changes in vocalization indicate that vocal communication during courtship in mice consists of at least three stages and imply that each vocalization type has a specific role in a phase of the courtship sequence. Our findings suggest that recording for a sufficiently long time and taking the phase of courtship into consideration could provide more insights into the role of vocalization in mouse courtship behavior in future study.

One of the cognitive capacities underlying language is core-Merge, which allows senders to combine two words into a sequence and receivers to recognize it as a single unit. Recent field studies suggest intriguing parallels in non-human animals, e.g., Japanese tits (Parus minor) combine two meaning-bearing calls into a sequence when prompting antipredator displays in other individuals. However, whether such examples represent core-Merge remains unclear; receivers may perceive a two-call sequence as two individual calls that are arbitrarily produced in close time proximity, not as a single unit. If an animal species has evolved core-Merge, its receivers should treat a two-call sequence produced by a single individual differently from the same two calls produced by two individuals with the same timing. Here, we show that Japanese tit receivers exhibit antipredator displays when perceiving two-call sequences broadcast from a single source, but not from two sources, providing evidence for core-Merge in animals.

Mice produce various sounds within the ultrasonic range in social contexts. Although these sounds are often used as an index of sociability in biomedical research, their biological significance remains poorly understood. We previously showed that mice repeatedly produced calls in a sequence (i.e. calling bout), which can vary in their structure, such as Simple, Complex or Harmonics. In this study, we investigated the use of the three types of calling bouts in different sociosexual interactions, including both same- and opposite-sex contexts. In same-sex contexts, males typically produced a Simple calling bout, whereas females mostly produced a Complex one. By contrast, in the opposite-sex context, they produced all the three types of calling bouts, but the use of each calling type varied according to the progress and mode of sociosexual interaction (e.g. Harmonic calling bout was specifically produced during reproductive behaviour). These results indicate that mice change the structure of calling bout according to sociosexual contexts, suggesting the presence of multiple functional signals in their ultrasonic communication.