Limits...
A role for ultrasonic vocalisation in social communication and divergence of natural populations of the house mouse (Mus musculus domesticus).

von Merten S, Hoier S, Pfeifle C, Tautz D - PLoS ONE (2014)

Bottom Line: We have analysed song frequency and duration, as well as spectral features of songs and syllables.Using a statistical syntax analysis we find complex temporal sequencing patterns that could suggest that the syntax conveys meaningful information to the receivers.We conclude that wild mice use USV for complex social interactions and that USV patterns can diverge fast between populations.

View Article: PubMed Central - PubMed

Affiliation: Department Evolutionary Genetics, Max Planck Institute for Evolutionary Biology, Plön, Germany.

ABSTRACT
It has long been known that rodents emit signals in the ultrasonic range, but their role in social communication and mating is still under active exploration. While inbred strains of house mice have emerged as a favourite model to study ultrasonic vocalisation (USV) patterns, studies in wild animals and natural situations are still rare. We focus here on two wild derived mouse populations. We recorded them in dyadic encounters for extended periods of time to assess possible roles of USVs and their divergence between allopatric populations. We have analysed song frequency and duration, as well as spectral features of songs and syllables. We show that the populations have indeed diverged in several of these aspects and that USV patterns emitted in a mating context differ from those emitted in same sex encounters. We find that females vocalize not less, in encounters with another female even more than males. This implies that the current focus of USVs being emitted mainly by males within the mating context needs to be reconsidered. Using a statistical syntax analysis we find complex temporal sequencing patterns that could suggest that the syntax conveys meaningful information to the receivers. We conclude that wild mice use USV for complex social interactions and that USV patterns can diverge fast between populations.

Show MeSH
Spectrograms of the 13 syllable types.Spectrograms were generated with 256(FFT) using the software Selena (Department of Animal Physiology, University of Tübingen; Germany). For abbreviations see table 2.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4016290&req=5

pone-0097244-g002: Spectrograms of the 13 syllable types.Spectrograms were generated with 256(FFT) using the software Selena (Department of Animal Physiology, University of Tübingen; Germany). For abbreviations see table 2.

Mentions: To find a reasonable number of syllable types, we tested a 5-syllable-type model against a 13-syllable type model (Figure 2). We defined 5 or 13 syllable types, respectively, using the extracted parameters and comparable criteria that have been used so far (see e.g. [12], [41]). We assigned each syllable to one of those types for both models. Our types and criteria for the 5-type model were as follows: Jumps, syllables that have one or more frequency jumps (at least 20 kHz change in less than 4 instantaneous FFT bins); Turns, syllables that have one or more frequency turns (one turn consisting of two frequency changes, each at least 0.8 kHz in less than 3 ms); Up, syllables with an upward frequency modulation (at least 0.05 kHz per 1 ms); Down, syllables with an downward frequency modulation (at least 0.05 kHz per 1 ms); Simple, all other syllable types. As visual inspection of syllables suggests that especially turns and jumps are much more variable, we additionally generated the 13-syllable type model. For this, we counted jumps in the first half of the syllable and jumps in the second half of the syllable, applying the same criteria as above. We also differentiated between jumps going up and jumps going down. This resulted in seven different jump syllable types, depending on whether there was a jump in the first (early jumps), in the second (late jumps) or in both halves and if these jumps were going up or down, or if there were more than two jumps in one syllable. If, for example, a jump to a higher frequency occurred in the first half of the syllable, this syllable was assigned to the Jump-Early-Up (JEU) type; if a jump to a higher frequency occurred in the second half of the syllable this would be a Jump-Late-Up (JLU) syllable. The same principle applies for syllables with two frequency jumps. A syllable in which the first jump is upwards and the second jump downwards would be a Jump-Up-Down (JUD) syllable. To distinguish between different types of turn syllables, we differentiated between syllables with a U-shaped turn, syllables with a turn in the opposite direction and syllables with more than one turn. For the resulting syllable types see figure 2 and table 2.


A role for ultrasonic vocalisation in social communication and divergence of natural populations of the house mouse (Mus musculus domesticus).

von Merten S, Hoier S, Pfeifle C, Tautz D - PLoS ONE (2014)

Spectrograms of the 13 syllable types.Spectrograms were generated with 256(FFT) using the software Selena (Department of Animal Physiology, University of Tübingen; Germany). For abbreviations see table 2.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4016290&req=5

pone-0097244-g002: Spectrograms of the 13 syllable types.Spectrograms were generated with 256(FFT) using the software Selena (Department of Animal Physiology, University of Tübingen; Germany). For abbreviations see table 2.
Mentions: To find a reasonable number of syllable types, we tested a 5-syllable-type model against a 13-syllable type model (Figure 2). We defined 5 or 13 syllable types, respectively, using the extracted parameters and comparable criteria that have been used so far (see e.g. [12], [41]). We assigned each syllable to one of those types for both models. Our types and criteria for the 5-type model were as follows: Jumps, syllables that have one or more frequency jumps (at least 20 kHz change in less than 4 instantaneous FFT bins); Turns, syllables that have one or more frequency turns (one turn consisting of two frequency changes, each at least 0.8 kHz in less than 3 ms); Up, syllables with an upward frequency modulation (at least 0.05 kHz per 1 ms); Down, syllables with an downward frequency modulation (at least 0.05 kHz per 1 ms); Simple, all other syllable types. As visual inspection of syllables suggests that especially turns and jumps are much more variable, we additionally generated the 13-syllable type model. For this, we counted jumps in the first half of the syllable and jumps in the second half of the syllable, applying the same criteria as above. We also differentiated between jumps going up and jumps going down. This resulted in seven different jump syllable types, depending on whether there was a jump in the first (early jumps), in the second (late jumps) or in both halves and if these jumps were going up or down, or if there were more than two jumps in one syllable. If, for example, a jump to a higher frequency occurred in the first half of the syllable, this syllable was assigned to the Jump-Early-Up (JEU) type; if a jump to a higher frequency occurred in the second half of the syllable this would be a Jump-Late-Up (JLU) syllable. The same principle applies for syllables with two frequency jumps. A syllable in which the first jump is upwards and the second jump downwards would be a Jump-Up-Down (JUD) syllable. To distinguish between different types of turn syllables, we differentiated between syllables with a U-shaped turn, syllables with a turn in the opposite direction and syllables with more than one turn. For the resulting syllable types see figure 2 and table 2.

Bottom Line: We have analysed song frequency and duration, as well as spectral features of songs and syllables.Using a statistical syntax analysis we find complex temporal sequencing patterns that could suggest that the syntax conveys meaningful information to the receivers.We conclude that wild mice use USV for complex social interactions and that USV patterns can diverge fast between populations.

View Article: PubMed Central - PubMed

Affiliation: Department Evolutionary Genetics, Max Planck Institute for Evolutionary Biology, Plön, Germany.

ABSTRACT
It has long been known that rodents emit signals in the ultrasonic range, but their role in social communication and mating is still under active exploration. While inbred strains of house mice have emerged as a favourite model to study ultrasonic vocalisation (USV) patterns, studies in wild animals and natural situations are still rare. We focus here on two wild derived mouse populations. We recorded them in dyadic encounters for extended periods of time to assess possible roles of USVs and their divergence between allopatric populations. We have analysed song frequency and duration, as well as spectral features of songs and syllables. We show that the populations have indeed diverged in several of these aspects and that USV patterns emitted in a mating context differ from those emitted in same sex encounters. We find that females vocalize not less, in encounters with another female even more than males. This implies that the current focus of USVs being emitted mainly by males within the mating context needs to be reconsidered. Using a statistical syntax analysis we find complex temporal sequencing patterns that could suggest that the syntax conveys meaningful information to the receivers. We conclude that wild mice use USV for complex social interactions and that USV patterns can diverge fast between populations.

Show MeSH