Selective approach behavior toward context-dependent ultrasonic vocalizations in male mice

This study demonstrates that male mice selectively approach conspecifics based on specific acoustic and temporal features of ultrasonic vocalizations recorded during social chasing, particularly responding to the calls of chased individuals rather than those of the chasers.

The Gist

Imagine a world where mice don't just squeak; they sing complex songs to tell each other stories. For a long time, scientists thought these high-pitched "ultrasonic vocalizations" (USVs) were mostly for moms to find their babies or for males to woo females. But what about when two male mice meet? Do they just ignore each other's songs, or is there a secret language of their own?

This study by Takahashi and colleagues acts like a detective story, investigating the hidden meanings behind the squeaks of male mice during a very specific, high-stakes scenario: a chase.

The Setup: A High-Tech "Whispering Gallery"

The researchers set up a special, soundproof maze (shaped like the letter "I") that acts like a whispering gallery. They used high-tech cameras and microphones to record the conversations of male mice.

They observed two main characters in these interactions:

1. The Chaser (The Resident): The mouse who lives in the cage and is doing the chasing.
2. The Chased (The Intruder): The mouse who was just let in and is running away.

The Discovery: Two Different "Songs" for Two Different Roles

The team discovered that these mice aren't just making noise; they are singing two completely different types of songs depending on whether they are the hunter or the hunted.

• The Chaser's Song: When the resident mouse chases the intruder, it sings a song that is lower in pitch and has a specific rhythm. Think of this like a driving beat or a march. It's assertive and territorial.
• The Chased's Song: When the intruder is running for its life, it sings a song that is higher-pitched, sharper, and full of "upward frequency-modulated" notes (sounds that swoop up like a siren). Think of this like a high-pitched plea or a distress call that says, "I'm not a threat! Let's just talk!"

The Experiment: The "Speaker Test"

To see if other mice understand these songs, the researchers played recordings of these two different "songs" to a third mouse (the listener) in a quiet room. They used a clever trick: they played the real songs and also played "scrambled" versions of the same sounds (like taking a sentence, shuffling the words, and playing it back so it sounds like gibberish).

The Results were surprising:

• When they played the "Chaser's Song": The listener mice didn't care. They walked right past the speaker. It was like hearing a boring news report; they ignored it.
• When they played the "Chased's Song": The listener mice immediately ran toward the speaker. They didn't just walk; they sprinted with a synchronized, almost robotic precision. It was as if the song was a magnet pulling them in.

The "Why": A Secret Code of Peace

Why would a mouse run toward a sound of someone being chased? The researchers suggest a fascinating theory:

In the animal kingdom, running away while screaming usually means "I am scared and I surrender." The "Chased" song (with its high-pitched, upward-swooping notes) seems to be a peace offering. It's a way for the intruder to say, "I'm not here to fight; I'm just passing through."

When other male mice hear this specific "peace song," they interpret it as a signal that it's safe to approach and investigate. It's like hearing a specific handshake code that says, "We are friends, come over here."

The Takeaway: It's Not Just About the Noise, It's About the Rhythm

The study found that it wasn't just the pitch of the sound that mattered, but also the timing. The "peace song" had a specific rhythm that the listeners recognized instantly. Even if the mice were used to the room (habituated), they still responded to the rhythm of the "Chased" song by moving in unison.

In simple terms:
Male mice have a secret language where the sound of someone running away actually invites others to come closer. It turns out that in the mouse world, a "distress call" isn't always a signal to run away; sometimes, it's a signal to come together.

This research changes how we see mouse communication. It's not just random squeaking; it's a sophisticated system where the context (who is chasing whom) changes the meaning of the song, and other mice are smart enough to understand the difference and react accordingly.

Technical Summary

1. Problem Statement

While Ultrasonic Vocalizations (USVs) are well-established as a primary mode of rodent communication, the functional significance of male-to-male vocal interactions remains poorly understood. Previous research has largely focused on:

• Pup calls: Eliciting maternal retrieval.
• Courtship songs: Eliciting approach in females but typically ignored by males.
• Affective states: 50 kHz calls (positive) vs. 22 kHz calls (negative) in rats.

Although it is known that male mice emit distinct USVs depending on whether they are chasing or being chased, it is unclear if these context-specific vocalizations serve as meaningful social signals that influence the behavior of conspecific males. Specifically, do male mice selectively respond to the acoustic features of USVs emitted during specific social roles (e.g., a chased intruder vs. a chasing resident)?

2. Methodology

Animals and Strain:

• Subjects: 46 male CBA/J mice (2–6 months old). This strain was selected to avoid age-related hearing loss common in C57BL/6 mice, ensuring high-frequency sensitivity.
• Groups: 18 mice for USV recording; 28 mice for playback experiments (18 in a novel environment, 9 in a habituated environment).

USV Recording and Classification:

• Setup: A custom "USVCAM" system with four microphones and infrared video tracking was used to localize sound sources during resident-intruder interactions.
• Behavioral Contexts: Interactions were categorized into four groups based on emitter identity (Resident vs. Intruder) and action (Chasing vs. Being Chased):
  • ResCing: Resident chasing intruder.
  • ResCed: Resident being chased.
  • IntCing: Intruder chasing resident.
  • IntCed: Intruder being chased.
• Analysis: Syllables were segmented using the USVSEG algorithm. Acoustic features (duration, max/min frequency, amplitude, frequency variation) were analyzed via Principal Component Analysis (PCA) and Fisher's exact tests to determine syllable distribution differences.

Playback Experiments:

• Apparatus: A custom "I-maze" (linear, 10cm x 43cm) designed to minimize spatial bias and accommodate the high directivity of ultrasonic speakers.
• Stimuli:
  • Target: 30-second sequences of USVs from IntCed (chased intruder) and ResCing (chasing resident).
  • Controls: "Noised" stimuli (phase-randomized USVs preserving amplitude spectrum but destroying temporal structure) and "Silent" conditions.
• Experimental Conditions:
  • Experiment 1 (Novel Context): Mice placed in the maze without prior acclimation to maximize exploratory drive. Stimuli triggered when the mouse was at the distal end.
  • Experiment 2 (Habituated Context): Mice acclimated to the maze for 10 minutes prior to testing to isolate stimulus-specific responses from general exploration.
• Behavioral Tracking: DeepLabCut was used for markerless pose estimation (head, ears, tail) to quantify approach latency, exploratory behavior, and movement synchrony.
• Statistical Analysis: Kaplan-Meier survival analysis (log-rank test) for approach latency; Pearson correlation coefficients for inter-individual synchrony; PCA and bootstrap tests for acoustic features.

3. Key Contributions

1. Context-Dependent Vocal Signaling: Demonstrated that male mice produce acoustically distinct USV repertoires based on their social role (chaser vs. chased), specifically distinguishing between IntCed and ResCing contexts.
2. Selective Male-to-Male Attraction: Provided the first robust evidence that male mice exhibit selective approach behavior toward USVs emitted by chased conspecifics (IntCed), but not toward those emitted by chasing conspecifics (ResCing).
3. Acoustic and Temporal Determinants: Identified that the approach response is driven by specific acoustic features (high mean frequency, high amplitude, and prevalence of upward frequency-modulated [up-fm] syllables) and the temporal organization of these syllables.
4. Stereotyped Synchrony: Revealed that the approach response to IntCed USVs is not only faster but also exhibits high temporal synchrony across individuals, suggesting a hard-wired, stereotyped behavioral program.

4. Key Results

Acoustic Differences:

• Syllable Composition: IntCed vocalizations contained a significantly higher proportion of up-fm and chevron syllables compared to ResCing.
• Acoustic Features: PCA and bootstrap testing confirmed that IntCed USVs have significantly higher mean frequency and maximum amplitude than ResCing USVs.
• Statistical Significance: The difference in syllable distribution between IntCed and ResCing was highly significant ($p < 0.001$).

Behavioral Responses (Novel Context):

• Approach Latency: Mice approached the speaker significantly faster when IntCed USVs were played compared to the "noised" control ($p = 0.018$). No significant difference was observed for ResCing USVs ($p = 0.140$).
• Synchrony: In the IntCed group, the approach trajectories of individual mice were highly correlated ($p = 0.0001$), indicating a synchronized, stereotyped response. This synchrony was absent in the ResCing and "noised" groups.

Behavioral Responses (Habituated Context):

• Exploratory Modulation: Even in habituated mice (where general exploration is reduced), IntCed USVs significantly increased exploratory behavior toward the sound source compared to controls ($p = 0.015$).
• Velocity Bias: Habituated mice showed a significant bias toward approach velocity over withdrawal velocity specifically during IntCed playback, confirming the response is a targeted reaction to the stimulus rather than random movement.

Syllable-Specific Triggers:

• Analysis of the specific time windows where approach occurred revealed that these windows were dominated by up-fm syllables with higher mean frequencies and amplitudes.
• ResCing sequences, which lacked these specific features, failed to elicit approach even when mice had the spatial opportunity to move.

5. Significance

• Redefining Male Social Communication: This study challenges the notion that male mice ignore male-male USVs. It establishes that males can decode complex social contexts from vocalizations and respond with specific, adaptive behaviors (approach).
• Functional Interpretation: The authors hypothesize that IntCed USVs (high-frequency up-fm calls) function as contact calls or de-escalation signals. Similar to 50-kHz calls in rats, these vocalizations may signal non-aggression or a desire to maintain social contact, prompting the receiver to approach and potentially diffuse a conflict or establish a hierarchy.
• Multimodal Integration: The findings suggest that social decisions in mice rely on a combination of acoustic features (pitch, amplitude) and temporal structure (rhythm, syllable order).
• Methodological Advancement: The study highlights the utility of the USVCAM system for source localization and the "I-maze" for precise playback control, offering a robust framework for future studies on rodent social neuroethology.

In conclusion, the paper demonstrates that male mice possess a sophisticated ability to distinguish between social contexts via USVs, selectively approaching the sounds of a "chased" conspecific. This behavior is driven by specific acoustic signatures (up-fm syllables) and is characterized by high inter-individual synchrony, suggesting a fundamental mechanism for social cohesion and conflict resolution in male mice.