Rank- and threat-dependent social modulation of innate defensive behaviors
- Department of Neurobiology, School of Basic Medical Sciences, Capital Medical University, China
- Beijing Institute for Brain Research, Chinese Academy of Medical Sciences and Peking Union Medical College, China
- Chinese Institute for Brain Research, China
Figures
Figure 1 with 1 supplement
Social presence attenuates looming-evoked defensive behavior in a rank-dependent manner.
(A) Schematic of the looming exposure paradigm. (B) Heatmaps of average time allocation in nest, edge, and center zones during different phases under single exposure, n=40 mice. (C) Time spent in each zone per minute, normalized by zone area. n=40 mice, paired t-test. (D) Heatmaps of average locomotion speed under single exposure. Arrows indicate mean velocity vectors, n=40 mice. (E) Locomotion speed in each zone. n=40 mice, paired t-test. (F) Change in zone occupancy (paired - single). n=20 mice for both dominant (Dom) and subordinate (Sub) groups, paired t-test. (G) Change in locomotion speed (paired - single). n=20 mice for both groups, paired t-test. (H) Representative raster plots of behaviors during single and paired looming exposures. D: dominant; S: subordinate. (I) Distribution of social time inside and outside of the nest. n=20 mice for both groups, paired t-test. (J) Representative trajectories showing distinct behavioral decisions within 5 s after looming onset. (K) Distribution of behavioral decisions following looming exposure. n=89, 85, 88, 96 trials for sD (single dominant), pD (pair dominant), sS (single subordinate), and pS (pair subordinate) groups, respectively; chi-square test. (L) Left, expected behavior co-occurrence probability. Right, observed minus expected co-occurrence probability assuming independence. n=20 mice for both groups. (M) Time spent in defensive behaviors (freezing, escape, tail-rattling, and rearing/up-stretch) during different phases. n=20 mice for both groups, two-way ANOVA with post hoc Tukey’s range test. (N) Peak escape speed. n=16, 18, 15, 15 mice, paired t-test. (O) Duration of the first looming-evoked freezing bout. n=16, 19, 15, 19 mice, paired t-test. (P) Time spent on self-grooming. n=20 mice per group, two-way ANOVA with post hoc Tukey’s range test. (Q) Time spent on rearing and up-stretch. n=20 mice per group, two-way ANOVA with post hoc Tukey’s range test. *p<0.05; **p<0.01; ***p<0.001.
Figure 1—figure supplement 1
Characterization of looming-evoked defensive behaviors.
(A) Time in the edge zone (orange) and transitions between edge and center zones (gray) plotted against edge width. The orange circle marks the elbow point, which was selected as the edge width for analysis in Figure 1. n=40 mice. (B) Time spent in each zone across different time windows under single exposure. n=40 mice; paired t-test. (C) Representative raster plots of behaviors during the habituation phase in single and paired looming exposures. (D) Distribution of co-triggered and single-triggered trials across ranks. n=65 bouts (Dom) and 71 bouts (Sub). (E) Time allocation for three types of grooming. n=20 mice for all groups. *p<0.05; **p<0.01; ***p<0.001.
Figure 2 with 1 supplement
Social presence alleviates sustained predatory stress and fine-tunes defensive strategies across ranks.
(A) Schematic of the rat exposure paradigm. (B) Heatmap of average time allocation in far, middle, and near zones during rat exposure, n=22 mice. (C) Time spent in each zone per minute, normalized by zone area. n=22 mice, paired t-test. (D) Heatmap of average locomotion speed, with arrows indicating average velocity vectors, n=22 mice. (E) Speed in each zone. n=22 mice, paired t-test. (F) Difference (paired – single) in time spent across zones during rat exposure. n=11 mice for both dominant and subordinate groups, paired t-test. (G) Difference (paired – single) in speed across zones during rat exposure. n=11 mice for each group, paired t-test. (H) Representative raster plots of behaviors during single and paired live rat exposure. D: dominant; S: subordinate. (I) Distribution of social time in far and other zones. n=11 mice for both dominant and subordinate groups. (J) Representative trajectories of example mice displaying passive (left) and active (right) defense during rat exposure. (K) Distribution of behavioral categories during rat exposure. n=11 mice for each group, chi-square test. (L) Left: expected behavior co-occurrence probability. Right: observed minus expected co-occurrence probability assuming independence. n=11 mice for each group. (M) Time spent on passive defense across time windows. n=11 mice for each group, two-way ANOVA with post hoc Tukey’s range test. (N) Freezing frequency. n=11 mice for each group, paired t-test. (O) Average freezing duration. n=11 mice for each group, paired t-test. (P) Time spent on active defense across time windows. n=11 mice for each group, two-way ANOVA with post hoc Tukey’s range test. (Q–S) Total stretch-attend time (Q), frequency (R), and average duration (S). n=11 mice for each group, paired t-test. (T–V) Total AIW (approach-investigation-withdrawal) time (T), frequency (U), and approach speed (V). n=11 mice for each group, paired t-test. *p<0.05; **p<0.01; ***p<0.001.
Figure 2—figure supplement 1
Characterization of rat-evoked defensive behaviors.
(A) Time in the far zone (blue) and transitions between the far and the non-far zones (gray) plotted against far zone width. The blue circle marks the elbow point, which was selected as the far zone width for analysis in Figure 2. n=22 mice. (B) Time in the near (orange) and middle (green) zones and transitions between near and middle zones (gray) plotted against near zone width. The gray circle marks the stable point, which was selected as the near zone width for analysis in Figure 2. n=22 mice. (C) Percentage of time spent in each zone during rat exposure. n=22 mice, paired t-test. (D) Representative raster plots of behaviors during the habituation phase in single and paired rat exposures. (E) Total freezing time. n=11 mice for each group, paired t-test. (F) Longest freezing duration. n=11 mice for each group, paired t-test. (G) Average AIW duration. n=11 mice for each group. (H) Withdrawal speed. n=11 mice for each group. *p<0.05; **p<0.01; ***p<0.001.
Figure 3 with 1 supplement
Rank- and threat-dependent organization of defensive behavioral transitions.
(A) Behavioral transition maps during single (left) and paired (right) looming exposure in dominant mice. Node size scales with total time allocated to each behavior; line thickness scales with transition probability. n=20 mice, paired t-test. (B) Matrix of behavioral transition differences between single and paired conditions (paired – single). n=20 mice, paired t-test. (C, D) Same as (A, B) for subordinate mice. n=20 mice, paired t-test. (E) Distribution of freezing bouts during looming exposure. Colors indicate freezing bouts that overlapped with the partner’s sniffing-partner onset (pSPon, dark green), approaching-partner offset (pAPoff, light green), freezing offset (pFoff, orange), and all other behaviors (Others, blue). n=111 bouts (Dom) and 129 bouts (Sub). (F) Change in freezing time (pre – post) within a 1 s window aligned to partner’s behavior during the 0–5 s post-looming period in dominant (left) and subordinate (right) mice. Dominant: n=1 (pAPoff), 3 (pSPon), and 18 (pFoff); subordinate: n=2 (pAPoff), 3 (pSPon), and 16 (pFoff), permutation test. (G) Same as (E) for the 5–60 s post-looming period. n=153 bouts (Dom) and 170 bouts (Sub). (H) Same as (F) for the 5–60 s post-looming period. Dominant: n=2 (pAPoff), 3 (pSPon), and 1 (pFoff); subordinate: n=2 (pAPoff), 4 (pSPon), and 2 (pFoff), permutation test. (I–L) Same as (A–D) for rat exposure. n=11 mice per rank, paired t-test. (M) Distribution of freezing bouts during rat exposure. Color codes as in (E). n=421 bouts (Dom) and 578 bouts (Sub). (N) Change in freezing time (pre – post) within a 1 s window aligned to partner’s behavior during rat exposure in dominant (left) and subordinate (right) mice. Dominant: n=18 (pAPoff), 22 (pSPon), and 164 (pFoff); subordinate: n=5 (pAPoff), 11 (pSPon), and 153 (pFoff), permutation test. (O) Distribution of withdrawal bouts during rat exposure. Colors indicate withdrawal bouts that overlapped with partner’s sniffing-partner onset (pSPon, dark green), approaching-partner offset (pAPoff, light green), withdrawal onset (pWon, orange), and all other behaviors (Others, blue). n=165 bouts (Dom) and 123 bouts (Sub). (P) Change in investigation time (pre – post) within a 1 s window aligned to partner’s behavior during rat exposure in dominant (left) and subordinate (right) mice. Dominant: n=7 (pAPoff), 11 (pSPon), and 7 (pWon); subordinate: n=5 (pAPoff), 8 (pSPon), and 13 (pWon), permutation test. (Q) Transition maps of behavioral categories for single (left) and paired (right) looming exposure in dominant mice. n=20 mice. (R) Same as (Q) for subordinate mice. n=20 mice. (S, T) Same as (Q, R) for rat exposure. n=11 mice per rank. *p<0.05; **p<0.01; ***p<0.001.
Figure 3—figure supplement 1
Transition probabilities between behavior pairs and behavior occupancy aligned to partner’s associated behaviors.
(A, B) Transition probability matrices for single (A) and paired (B) exposure in the looming paradigm for dominant mice. n=20 mice. Green squares indicate transitions observed in more than five mice. (C, D) Same plot as (A, B) for subordinate mice. n=20 mice. (E–H) Same plot as (A–D) for rat exposure paradigm. n=11 mice for both dominant and subordinate groups. (I, J) Freezing occupancy aligned to partner’s associated behavioral events (pAPoff, top; pSPon, middle; pFoff, bottom) for dominant and subordinate mice during different time periods in looming exposure (I: 0–5 s post-looming; J: 5–60 s post-looming). n=20 mice. (K) Same plot as (I, J) for freezing occupancy in rat exposure. n=11 mice. (L) Investigation occupancy aligned to partner’s associated behavioral events (pAPoff, top; pSPon, middle; pWon, bottom) for dominant and subordinate mice during rat exposure. n=11 mice.
Figure 4 with 1 supplement
Threat exposure reinforces social roles and promotes cohesive behavior.
(A) Time spent on social behaviors within a 1 min window before and after looming exposure. n=20 mice for both groups, paired t-test. (B) Average bout duration of social behaviors before and after looming exposure. n=20 mice for both groups, paired t-test. (C) Frequency of social behaviors before and after looming exposure. n=20 mice for both groups. (D) Time spent on social behaviors during a 5 min window before and during rat exposure. n=11 mice for both groups, paired t-test. (E) Average bout duration of social behaviors before and during rat exposure. n=11 mice for both groups, paired t-test. (F) Frequency of social behaviors before and during rat exposure. n=11 mice for both groups, paired t-test. (G) Huddling time expressed as a percentage of total social behavior time during a 5 min window before and during rat exposure. n=11 mice for both groups, paired t-test. (H) Schematic illustration of social interactions. Asterisks (*) denote proactive social behaviors. Bouts 1, 2, 4, and 5 represent proactive bouts that elicited a response from the partner, whereas bouts 3 and 6 represent the ones that failed to elicit a partner’s response. (I–L) Social interactions in the looming paradigm: (I) Percentage of proactive social bouts across ranks. (J) Average duration of proactive bouts. (K) Response rate to proactive bouts. (L) Average duration of reactive bouts. n=20 mice for each group, paired t-test. (M–P) Same plots as (I–L) for the rat paradigm. n=11 mice for each group. (Q) Left: behavior transition maps during baseline for paired looming exposure for dominant mice. Right: matrix of behavior transition differences between threat and baseline conditions (threat – baseline). n=20 mice, paired t-test. (R) Same plot for subordinate mice in looming exposure. n=20 mice, paired t-test. (S) Escape bouts categorized by termination status: CA, corner alone; CP, corner paired; NA, nest alone; NP, nest paired. ‘Alone’ indicates the partner was absent from the destination location at bout termination, whereas ‘paired’ indicates the partner was present. n=32 bouts (Dom) and 42 bouts (Sub). (T) Distribution of the first behavior following escape. (D-E) and (S-E) denote escape bouts performed by dominant and subordinate mice, respectively. n=6, 7, 22, 32 bouts, Fisher’s exact test. (U) Time allocation of different behaviors during a 10 s time window following looming exposure. n=32 bouts (D-E) and 42 bouts (S-E), Fisher’s exact test. (V, W) Same plot as (Q, R) for rat exposure. n=11 mice, paired t-test. (X–Z) Same plot as (S–U) for withdrawal bouts in rat exposure. (X) n=124 bouts (Dom), 76 bouts (Sub), chi-square test. (Y) n=47, 17, 77, 59 bouts, Fisher’s exact test. (Z) n=160 bouts (D–W) and 120 bouts (S–W), Fisher’s exact test. *p<0.05; **p<0.01; ***p<0.001.
Figure 4—figure supplement 1
Threat exposure reshapes social interactions and dynamics.
(A–C) Total huddling time (A), huddling frequency (B), and average huddling duration (C) during the 5 min periods before and during rat exposure. n=11 mice for both groups, paired t-test. (D–G) Percentage of proactive huddling bouts (D), proactive huddling duration (E), response rate (F), and reactive huddling duration (G) during the 5 min periods before and during rat exposure. n=11 mice for both groups, paired t-test. (H–J) Total sniffing duration (H), sniffing frequency (I), and average sniffing duration (J) during the 5 min periods before and during rat exposure. n=11 mice for both groups, paired t-test. (K–M) Same plot as (H–J) during a 1 min period before and after looming exposure. n=22 mice for both groups, paired t-test. (N) Escape bouts categorized by partner’s location at escape offset during paired looming exposure. n=32 bouts (Dom) and 42 bouts (Sub). (O) Withdrawal bouts categorized by partner’s location at withdrawal offset in paired rat exposure. n=160 bouts (Dom) and 120 bouts (Sub), chi-square test. (P) Behavior occupancy aligned to dominant escape onset in paired looming exposure. Left, dominant mouse; right, subordinate mouse. n=32 bouts. (Q) Behavior occupancy aligned to subordinate escape onset in paired looming exposure. Left, subordinate mouse; right, dominant mouse. n=42 bouts. (R) Behavior occupancy aligned to dominant withdrawal onset in paired rat exposure. Left, dominant mouse; right, subordinate mouse. n=160 bouts. (S) Behavior occupancy aligned to subordinate withdrawal onset in paired rat exposure. Left, subordinate mouse; right, dominant mouse. n=120 bouts.
Videos
Video 1
Example video showing an escape decision in response to a looming stimulus in the paired condition.
Video 2
Example video showing an escape-after-assessment decision in response to a looming stimulus in the paired condition.
Video 3
Example video showing a freezing decision in response to a looming stimulus in the paired condition.
Video 4
Example video showing a no-response decision in response to a looming stimulus in the paired condition.
Video 5
Example video showing active defense against a live rat in the paired condition: one mouse approaches while the other remains in the far zone.
Video 6
Example video showing active defense against a live rat in the paired condition: sequential approach, with one mouse leading and the other following.
Video 7
Example video showing passive defense against a live rat in the paired condition: both mice freeze and huddle together in a single corner.
Video 8
Example video showing passive defense against a live rat in the paired condition: two mice freeze in different corners.
Tables
Key resources table
| Reagent type (species) or resource | Designation | Source or reference | Identifiers | Additional information |
|---|---|---|---|---|
| Strain, strain background (Mus musculus, male) | C57BL/6J | Chinese Institute for Brain Research, Beijing | RRID:IMSR_JAX:000664 | |
| Software, algorithm | MATLAB 2024b | MathWorks | RRID:SCR_001622 | |
| Software, algorithm | Python v3.9 | Python Software Foundation | RRID:SCR_008394 | |
| Software, algorithm | R v4.2.1 | R Project for Statistical Computing | RRID:SCR_001905 |
Additional files
Download links
A two-part list of links to download the article, or parts of the article, in various formats.
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
Rank- and threat-dependent social modulation of innate defensive behaviors
eLife 15:RP109571.
https://doi.org/10.7554/eLife.109571.3
