Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Article
  • Published:

Frontal neurons driving competitive behaviour and ecology of social groups

Nature volume 603, pages 661–666 (2022)Cite this article

Subjects

Abstract

Competitive interactions have a vital role in the ecology of most animal species1,2,3 and powerfully influence the behaviour of groups4,5. To succeed, individuals must exert effort based on not only the resources available but also the social rank and behaviour of other group members2,6,7. The single-cellular mechanisms that precisely drive competitive interactions or the behaviour of social groups, however, remain poorly understood. Here we developed a naturalistic group paradigm in which large cohorts of mice competitively foraged for food as we wirelessly tracked neuronal activities across thousands of unique interactions. By following the collective behaviour of the groups, we found neurons in the anterior cingulate that adaptively represented the social rank of the animals in relation to others. Although social rank was closely behaviourally linked to success, these cells disambiguated the relative rank of the mice from their competitive behaviour, and incorporated information about the resources available, the environment, and past success of the mice to influence their decisions. Using multiclass models, we show how these neurons tracked other individuals within the group and accurately predicted upcoming success. Using neuromodulation techniques, we also show how the neurons conditionally influenced competitive effort—increasing the effort of the animals only when they were more dominant to their groupmates and decreasing it when they were subordinate—effects that were not observed in other frontal lobe areas. Together, these findings reveal cingulate neurons that serve to adaptively drive competitive interactions and a putative process that could intermediate the social and economic behaviour of groups.

This is a preview of subscription content, access via your institution

Access options

Buy this article

  • Purchase on Springer Link
  • Instant access to full article PDF
Buy now

Prices may be subject to local taxes which are calculated during checkout

Fig. 1: Naturalistic group foraging paradigm and competitive behaviour.
Fig. 2: Single-neuronal representations of group behaviour.
Fig. 3: Modulation and neural population predictions of competitive success.
Fig. 4: Selective control over competitive behaviour through ACC excitation and inhibition.

Similar content being viewed by others

Data availability

Additional behavioural and neuronal data that support the findings of the study are available from the corresponding author upon reasonable request. Source data are provided with this paper.

Code availability

All software packages used in this study are listed in the Reporting Summary along with their versions. The custom MALAB codes used to perform data and statistical analyses that support the findings of this study are available from the corresponding author upon reasonable request.

References

  1. Darwin, C. On the Origin of Species by Means of Natural Selection, or, the Preservation of Favoured Races in the Struggle for Life (John Murray, 1859).

  2. Huntingford, F. A. & Turner, A. K. In Animal Conflict (eds Huntingford, F. A. & Turner, A. K.) 227–250 (Springer, 1987).

  3. Zink, C. F. et al. Know your place: neural processing of social hierarchy in humans. Neuron 58, 273–283 (2008).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  4. Hoshaw, B. A., Evans, J. C., Mueller, B., Valentino, R. J. & Lucki, I. Social competition in rats: cell proliferation and behavior. Behav. Brain Res. 175, 343–351 (2006).

    Article  PubMed  PubMed Central  Google Scholar 

  5. Nagy, M., Akos, Z., Biro, D. & Vicsek, T. Hierarchical group dynamics in pigeon flocks. Nature 464, 890–893 (2010).

    Article  ADS  CAS  PubMed  Google Scholar 

  6. Stephens, D. W. In Encyclopedia of Ecology (eds Jorgensen, S. E. & Fath, B.) 284–289 (Elsevier, 2008).

  7. Clark, C. W. & Mangel, M. The evolutionary advantages of group foraging. Theor. Popul. Biol. 30, 45–75 (1986).

    Article  MathSciNet  MATH  Google Scholar 

  8. Sapolsky, R. M. The influence of social hierarchy on primate health. Science 308, 648–652 (2005).

    Article  ADS  CAS  PubMed  Google Scholar 

  9. Waite, T. A. The Bible of Social Foraging Theory. Ecology 82, 906–907 (2001).

    Article  Google Scholar 

  10. Zhou, T. et al. History of winning remodels thalamo-PFC circuit to reinforce social dominance. Science 357, 162–168 (2017).

    Article  ADS  CAS  PubMed  Google Scholar 

  11. Wang, F. et al. Bidirectional control of social hierarchy by synaptic efficacy in medial prefrontal cortex. Science 334, 693–697 (2011).

    Article  ADS  CAS  PubMed  Google Scholar 

  12. Levy, D. R. et al. Dynamics of social representation in the mouse prefrontal cortex. Nat. Neurosci. 22, 2013–2022 (2019).

    Article  CAS  PubMed  Google Scholar 

  13. Allsop, S. A. et al. Corticoamygdala transfer of socially derived information gates observational learning. Cell 173, 1329–1342 (2018).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  14. Haroush, K. & Williams, Z. M. Neuronal prediction of opponent’s behavior during cooperative social interchange in primates. Cell 160, 1233–1245 (2015).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  15. Kingsbury, L. et al. Correlated neural activity and encoding of behavior across brains of socially interacting animals. Cell 178, 429–446 (2019).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  16. Lee, D. K. et al. Reduced sociability and social agency encoding in adult Shank3-mutant mice are restored through gene re-expression in real time. Nat. Neurosci. 24, 1243–1255 (2021).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  17. Dal Monte, O., Chu, C. C. J., Fagan, N. A. & Chang, S. W. C. Specialized medial prefrontal-amygdala coordination in other-regarding decision preference. Nat. Neurosci. 23, 565–574 (2020).

    Article  CAS  PubMed  Google Scholar 

  18. Fan, Z. et al. Using the tube test to measure social hierarchy in mice. Nat. Protoc. 14, 819–831 (2019).

    Article  CAS  PubMed  Google Scholar 

  19. Rushworth, M. F. & Behrens, T. E. Choice, uncertainty and value in prefrontal and cingulate cortex. Nat. Neurosci. 11, 389–397 (2008).

    Article  CAS  PubMed  Google Scholar 

  20. Ruff, C. C. & Fehr, E. The neurobiology of rewards and values in social decision making. Nat. Rev. Neurosci. 15, 549–562 (2014).

    Article  CAS  PubMed  Google Scholar 

  21. Cowen, S. L., Davis, G. A. & Nitz, D. A. Anterior cingulate neurons in the rat map anticipated effort and reward to their associated action sequences. J. Neurophysiol. 107, 2393–2407 (2012).

    Article  PubMed  Google Scholar 

  22. Hillman, K. L. & Bilkey, D. K. Neural encoding of competitive effort in the anterior cingulate cortex. Nat. Neurosci. 15, 1290–1297 (2012).

    Article  CAS  PubMed  Google Scholar 

  23. Pereira, T. D., Shaevitz, J. W. & Murthy, M. Quantifying behavior to understand the brain. Nat. Neurosci. 23, 1537–1549 (2020).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  24. Desjardins, C., Maruniak, J. A. & Bronson, F. H. Social rank in house mice: differentiation revealed by ultraviolet visualization of urinary marking patterns. Science 182, 939–941 (1973).

    Article  ADS  CAS  PubMed  Google Scholar 

  25. Kareem, A. M. & Barnard, C. J. The importance of kinship and familiarity in social interactions between mice. Anim. Behav. 30, 594–601 (1982).

    Article  Google Scholar 

  26. Calhoun, J. B. The social aspects of population dynamics. J. Mammal. 33, 139–159 (1952).

    Article  Google Scholar 

  27. Williamson, C. M., Lee, W. & Curley, J. P. Temporal dynamics of social hierarchy formation and maintenance in male mice. Anim. Behav. 115, 259–272 (2016).

    Article  Google Scholar 

  28. Marlin, B. J., Mitre, M., D’Amour J, A., Chao, M. V. & Froemke, R. C. Oxytocin enables maternal behaviour by balancing cortical inhibition. Nature 520, 499–504 (2015).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

  29. Baez-Mendoza, R., Harris, C. J. & Schultz, W. Activity of striatal neurons reflects social action and own reward. Proc. Natl Acad. Sci. USA. 110, 16634–16639 (2013).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

  30. Snyder-Mackler, N. et al. Social determinants of health and survival in humans and other animals. Science 368, eaax9553 (2020).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  31. Manning, C. J., Wakeland, E. K. & Potts, W. K. Communal nesting patterns in mice implicate MHC genes in kin recognition. Nature 360, 581–583 (1992).

    Article  ADS  CAS  PubMed  Google Scholar 

  32. Yang, M., Weber, M. D. & Crawley, J. N. Light phase testing of social behaviors: not a problem. Front. Neurosci. 2, 186–191 (2008).

    Article  PubMed  PubMed Central  Google Scholar 

  33. Lindzey, G., Winston, H. & Manosevitz, M. Social dominance in inbred mouse strains. Nature 191, 474–476 (1961).

    Article  ADS  CAS  PubMed  Google Scholar 

  34. Xu, H. et al. A disinhibitory microcircuit mediates conditioned social fear in the prefrontal cortex. Neuron, 102, 668–682 (2019).

    Article  CAS  PubMed  Google Scholar 

  35. Murugan, M. et al. Combined social and spatial coding in a descending projection from the prefrontal cortex. Cell 171, 1663–1677 (2017).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  36. Olsson, I. A. S. et al. Understanding behaviour: the relevance of ethological approaches in laboratory animal science. Appl. Anim. Behav. Sci. 81, 245–264 (2003).

    Article  Google Scholar 

  37. Dewsbury, D. A. Comparative psychology, ethology, and animal behavior. Annu. Rev. Psychol. 40, 581–602 (1989).

    Article  Google Scholar 

  38. Musallam, S., Bak, M. J., Troyk, P. R. & Andersen, R. A. A floating metal microelectrode array for chronic implantation. J. Neurosci. Methods 160, 122–127 (2007).

    Article  PubMed  Google Scholar 

  39. Prasad, A. et al. Abiotic-biotic characterization of Pt/Ir microelectrode arrays in chronic implants. Front. Neuroeng. 7, 2 (2014).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  40. Gremel, C. M. & Costa, R. M. Orbitofrontal and striatal circuits dynamically encode the shift between goal-directed and habitual actions. Nat. Commun. 4, 2264 (2013).

    Article  ADS  CAS  PubMed  Google Scholar 

  41. Han, W. et al. Integrated control of predatory hunting by the central nucleus of the amygdala. Cell 168, 311–324 (2017).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  42. Scheggia, D. et al. Somatostatin interneurons in the prefrontal cortex control affective state discrimination in mice. Nat. Neurosci. 23, 47–60 (2020).

    Article  CAS  PubMed  Google Scholar 

  43. Gehrlach, D. A. et al. Aversive state processing in the posterior insular cortex. Nat. Neurosci. 22, 1424–1437 (2019).

    Article  CAS  PubMed  Google Scholar 

  44. Klavir, O., Prigge, M., Sarel, A., Paz, R. & Yizhar, O. Manipulating fear associations via optogenetic modulation of amygdala inputs to prefrontal cortex. Nat. Neurosci. 20, 836–844 (2017).

    Article  CAS  PubMed  Google Scholar 

  45. Hung, C. P., Kreiman, G., Poggio, T. & DiCarlo, J. J. Fast readout of object identity from macaque inferior temporal cortex. Science 310, 863–866 (2005).

    Article  ADS  CAS  PubMed  Google Scholar 

  46. Wasserman, L. All of Statistics: a Concise Course in Statistical Inference (Springer Texts, 2004).

Download references

Acknowledgements

We thank M. Mejdell, N. Occidental and S. Folz for their help with data collection and behavioural scoring; and A. Khanna, Y. Cohen, Y. Kfir, M. Mustroph, M. Jamali, N. Padilla, K. Tye, D. Rosene, K. Rockland and L. Toth for feedback. S.W.L. is supported by the Autism Science Foundation; R.B.-M. is funded by an MGH-ECOR Fund for Medical Discovery Fellowship and a NARSAD Young Investigator Grant from the Brain & Behavior Research Foundation; and Z.M.W. is supported by NIH R01HD059852, NIH R01NS091390 and NIH U01NS123130.

Author information

Authors and Affiliations

  1. Department of Neurosurgery, Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA

    S. William Li, Omer Zeliger, Leah Strahs, Raymundo Báez-Mendoza, Lance M. Johnson, Aidan McDonald Wojciechowski & Ziv M. Williams

  2. Department of Anatomy and Neurobiology, Boston University School of Medicine, Boston, MA, USA

    S. William Li

  3. Northeastern University, Boston, MA, USA

    Omer Zeliger, Leah Strahs & Aidan McDonald Wojciechowski

  4. Harvard College, Harvard University, Boston, MA, USA

    Lance M. Johnson

  5. Harvard–MIT Division of Health Sciences and Technology, Boston, MA, USA

    Ziv M. Williams

  6. Program in Neuroscience, Harvard Medical School, Boston, MA, USA

    Ziv M. Williams

Authors
  1. S. William Li
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Omer Zeliger
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Leah Strahs
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Raymundo Báez-Mendoza
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Lance M. Johnson
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Aidan McDonald Wojciechowski
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Ziv M. Williams
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

S.W.L. and Z.M.W. conceived and designed the study. S.W.L., O.Z., L.S., L.M.J. and A.M.W. performed the experiments. S.W.L., O.Z. and R.B.-M. performed the analyses. Z.M.W. directed and supervised all aspects of the research.

Corresponding author

Correspondence to Ziv M. Williams.

Ethics declarations

Competing interests

The authors declare no competing financial or non-financial interests.

Additional information

Peer review information Nature thanks Steve Chang and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Extended data figures and tables

Extended Data Fig. 1 Group foraging task and dominance assays.

a, A 3-D representation of the custom-designed arena with automated gates and food dispenser. Measurements are in cm. b, A tube test assay was used to evaluate the linear and transitive dominance hierarchies of the animals (n=7 unique groups totaling 49 animals over 6 weeks). The absolute ranks of the animals were confirmed to be stable over time (p>0.2 across all ranks and experimental weeks; Signed-rank). Error bars denote mean ± s.e.m. c, A urine marking assay was used to confirm the robustness of social dominance hierarchy across assays. The ratio of pixels with urine are displayed for each pairings of animals arranged based on their rank in the dominance hierarchies as determined by the tube test assay (mean ± s.e.m.; n=7 unique pairs across permutations). d, Schematic representation of the main task conditions and their permutation. The different primary economic (reward sizes), environmental (distance from staging area to reward zone) and social (relative rank or presence of social agents) conditions are displayed along the margins. Timeline of an example session where trials were run in a pseudo-randomized block design is shown in the middle. e, Example trials from one session depicting trajectories from all four possible staging areas (Fig. 1b). f, Heat maps showing increased latency (Left) and increased order (Right) to reach the reward zone with decreasing absolute hierarchical and relative ranks. Mean ± s.e.m. g, Heat maps showing increased time (Left) and increased order (Right) to exit the staging area with decreasing absolute hierarchical and relative ranks. Mean ± s.e.m. h, Spearman correlation between reaction time and competitive success. Heat map showing decreased correlation between reaction time (i.e. order to exit the staging area) and competitive success (i.e. order to reach the reward zone) with decreasing absolute hierarchical and relative ranks. For panels f–h, n=7 animals per absolute rank, across n=63 sessions. Dots represent session averages.

Source data

Extended Data Fig. 2 Confirming the relation between relative rank and competitive success across different metrics.

a, To evaluate the independent effect that reward amount or travel distance may have on the animals’ behaviour, we varied both the staging area from which the animals started and amount of food. Left, The animals reached the reward zone faster on high reward trials (**F(1,1751)=9.13, p=0.01) and on trials where they started from the near staging areas (***F(1,1751)=24.9, p=2.51x10-5). However, there was no interaction between terms describing the animals rank, reward amount or distance (Frank:reward(6,1751)=0.25, p=0.96; Frank:stagingarea (6,1751)=0.33, p=0.92; Freward:stagingarea (1,1751)=2.58, p=0.11; Frank:reward:stagingarea(6,1751)=0.4, p=0.88; three-way ANOVA). Right, Similar findings were also made when performing a within-session spearman correlation values across all trial conditions (F(3,251)=0.51, p=0.67; one-way ANOVA). b, There was no significant difference in reaction time when animals ran the task alone on control trials prior to the start of group trials (χ2(6,440)=9.42, p=0.15; Kruskal-Wallis). Error bars denote mean±95%CI. c, The mid-ranked animal was more likely to react faster (i.e. leaving the staging area faster than others) in the group competition task at lower relative ranks (1 and 2; *rs=-0.55, p=1.4x10-21, **rs=-0.34, p=2.2x10-8) and to react slower at higher relative rank (4; ***rs=0.68, p=2.3x10-35) but not at rank 3 (rs=-0.04, p=0.5). Error bars denote mean±95%CI. d, Hierarchical rank within a group was correlated with reaction time across all trial conditions. Spearman correlation calculated across sessions (n=63). Inset, There was no difference in within-session spearman correlation values across all trial conditions (F(3,251)=1.11, p=0.35; one-way ANOVA). Shaded areas denote mean±95%CI. e, Left, Graphical depiction of high vs low rank variance groupings. Right, Although relative rank was positively correlated with competitive success in both high and low rank variance trials, there was a significant interaction between group rank variance and the animals’ relative rank in influencing the animals’ competitive success (n=63 total sessions per high vs low rank variance; *Frankvar:relrank(3,265)=3.94, p=0.009; two-way ANOVA). Together, these findings suggested that, even when controlling for the animals’ rank relative to conspecific competitors, the specific rank of others played a significant role in adjusting the animal’s competitive behaviour. Error bars denote mean ± s.e.m. For all panels, n=7 animals per absolute rank, across n=63 sessions. Dots represent session averages.

Source data

Extended Data Fig. 3 Relation between social rank and competitive success.

a, Group-dropping procedure demonstrated a consistent behavioural correlation between hierarchical rank and competitive success across all groups (F(7,440)=0.68, p=0.69; one-way ANOVA), suggesting that no particular groups(s) disproportionately affected the main behavioural results. Error bars denote mean±95%CI. b, There was no difference in the animals’ competitive success between the first and second half of each session (Frank(6,872)=0.025, p=0.70; two-way RM-ANOVA) and no interaction between terms that defined the animal’s rank and time period (Frank*time (6,872)=0.21, p=0.27; two-way RM-ANOVA); suggesting that the animals’ overall competitive success was stable within individual sessions. Error bars denote mean±95%CI. c, Comparing the animals’ original dominance rank to the order in which the same animals reached reward during foraging demonstrated that competitive success did not significantly differ over consecutive sessions (n=7 groups of 7 mice per session; FsessionNumber(7,391)=0.21, p=0.98; two-way RM-ANOVA). There was no interaction between terms that defined the animal’s rank and session number (FsessionNumber*rank(42,391)=0.103, p=0.80; two-way RM-ANOVA). Error bars denote mean ± s.e.m. d, Left, The mid-rank (recorded) animals’ latency to the reward zone was stable across the duration of the entire session (n=63 sessions; p>0.1 for all time points; one-sample t-test), indicating that motivation for food was consistent throughout the session. Right, Representative behavioural trajectories of the recorded animal across all group trials in one session (same session as shown in Fig. 1b). There were no differences in motoric behaviour between the recorded animal (the mid-ranked animal having the head stage) compared to its groupmates. Squares indicate the instantaneous position of the animal when the first animal in the group entered the reward zone (RZ), coloured by the recorded animal’s relative rank. e, Heat map showing the percentage of trials where the relative ranking of animals exactly matched their competitive success. The animals’ relative ranks and the order in which they reached reward matched exactly on 38.5% of the foraging trials and at a proportion that was significantly higher than expected from chance (p<0.05, Permutation tests; across all comparisons). f, Latency to reach the reward zone was dependent on the animal’s relative rankings when foraging in groups (*rs=0.8, p=3.1x10-58). The mid-ranked animals spent more time occupying the reward zone compared to competitors when they were more dominant relative to others (g, *rs=-0.46, p=1.78x10-10; Spearman correlation) as well as when they reached the reward zone prior to others (h, *rs=-0.80, p=2.46x10-51; Spearman correlation). i, The animals’ relative ranks and the order in which they reached reward were significantly correlated on a session-by-session basis (*rs=0.77, p=2.17x10-47; Spearman correlation). j, Competitive success was stable following electrode implantation (n=4 animals, n=2 sessions pre-implant and first 2 sessions post-implant; Fimplant(1,63)=0.16, p=0.69; one-way RM-ANOVA). Error bars denote mean ± s.e.m. For panels b, f–i, dots represent session averages (n=63). Box-plot edges represent 25th/75th percentiles with center=median and whiskers=1st-99th percentile range.

Source data

Extended Data Fig. 4 Task controls and behavioural performances.

a, A 3-D representation of the custom-designed arena with automated gates and food dispenser (Extended Data Fig. 1) but with dividers now placed in the staging area (SA) for the pre-trial partitioning control. b, Spatial trajectories of all animals across all trials within one representative session with SA dividers. Squares indicate the instantaneous position of the animals when the first mouse reaches the reward zone (RZ). c, Higher ranked animals displayed greater success when compared to their subordinates both in trials with and without SA partitioning (rs=0.419, p=9.1x10-7 for divider trials; rs=0.593, p=1.22x10-16 for open trials; Spearman correlation). The presence of dividers, however, significantly affected the effect that the animals’ rank had on their competitive success (*Fdivider*rank(6,223)=2.34, p=0.032; two-way ANOVA). N=16 sessions across n=3 unique groups of 7 mice. Shaded areas denote mean±95%CI. d, Left, For trials with SA partitioning, the dominant animals were more likely to be the first to reach the reward zone compared to subordinate partners (*rs=-0.702, p=6.74x10-18; Spearman correlation). However, there was no relation between dominance rank and leaving the staging area (rs=-0.095, p=0.12). Right, In trials without SA partitioning, the dominant animals were more likely to be the first to exit the staging area (*rs=-0.43, p=2.41x10-6) and reach the reward zone (*rs=-0.742, p=8.15x10-21). There was a significant interaction between the animal’s absolute dominance in their respective hierarchies and the presence of dividers in the staging area (***Fdivider*rank(6,447)=2.58, p=0.018; two-way ANOVA), together suggesting that the animals took into account information about the relative ranks of the other animals both prior to and after trial start during competition. N=16 sessions across n=3 unique groups of 7 mice. e, Graphic depicting a dynamic non-social context control where inanimate totems were pulled from the staging area until the entrance point of the reward zone using an automated retractor while one mouse foraged for a food pellet (see Methods). f, Spatial trajectories of all moving totems and mouse across all trials within one representative session. Squares indicate the instantaneous position of the animals and totems when the mouse reached the entrance point. g, The mice and moving totems reached the entrance point at approximately the same time (n=54 sessions across n=14 mice; F(3,197)=0.27, p=0.85; one-way ANOVA). h, There were no main effects of hierarchical rank (Frank(6,107)=0.89, p=0.50; two-way RM-ANOVA) or interaction effects of rank and totem movement (Frank*totemtype(6,107)=0.68, p=0.67; two-way RM-ANOVA) on the animals’ latency to reach reward during the totem trials; together suggesting that presence of the moving totems did not influence the animals’ behaviour. There was no difference in latency to reach the reward zone based on absolute dominance rank when the animals foraged alone. N=54 total sessions across n=2 mice per rank. Error bars denote mean±95%CI. i, To evaluate whether the more dominant animals were stronger or were more perseverant, the animals were required to move a mass of variable weight at the reward entrance point. There was no significant difference in latency for animals to reach the entrance point between different weighted blockers (F(4,255)=2.16, p=0.07) but did display an expected difference in for the different staging areas (*F(1,255)=35.2, p≈0). There was no interaction between terms describing the weight amount and staging area location (F(4,255)=0.23, p=0.92; Two-way ANOVA). N=26 mice per blocker weight. Error bars denote mean±95%CI. j, There was a significant difference in latency to enter the reward zone based on the weight of the blockers (*F(4,255)=27.9, p≈0), but no difference between staging area locations (F(1,255)=0.08, p=0.78). There was no interaction between terms describing the weight amount and staging area location (F(4,255)=0.25, p=0.91; Two-way ANOVA). N=26 mice per blocker weight. Error bars denote mean±95%CI. k, There was no difference in latency to enter the reward zone based on ranks across any of the weights (F(24,255)=7.9, p=0.32; Two-way ANOVA); suggesting that hierarchical rank did not significantly affect the animals’ strength or perseverance. N=26 mice per blocker weight (n=3,3,4,4,5,3,4 for absolute ranks 1-7, respectively). Error bars denote mean ± s.e.m. Dots represent session averages. Box-plot edges represent 25th/75th percentiles with centre=median and whiskers=1st-99th percentile range.

Source data

Extended Data Fig. 5 Temporal dynamic of neural population response.

a, Electrode localization for all recorded animals (n=7), colour-coded for each mouse. cg1/cg2=cingulate areas 1/2; prL=prelimbic cortex; iL=infralimbic cortex. b, Peri-event histogram (PETH) and spike rasters a neuron that displayed a selective change in firing rate for only high vs low competitive success trials. PETHs are aligned to when the recorded animal enters the reward zone. Grey dots represent gate opening (i.e. trial start). c, PETH and spike rasters of a neuron that did not display selective changes in firing rate change. PETHs are aligned to when the recorded animal enters the reward zone. Grey dots represent gate opening. d, Recruitment of ACC neurons over the course of the trials. For each epoch, each Venn diagram depicts the distribution of neurons within the recorded population that responded differentially to the three primary factors that described the animals’ competitive interactions: competitive success (CSS), reward size (REW), and relative hierarchical rank (RR) across all task-relevant epochs. For each main factor, embedded pie charts show the proportion of overlapping cells based on their encoding properties. e, For each epoch, all recorded neurons are highlighted and labelled with colours corresponding to their specific encoding properties on the same t-SNE space as shown in Fig. 2f. Dots represent each recorded neuron (n=1049 from 7 mice). Grey dots represent neurons that displayed no task-related modulation. f, Scatter plots illustrating the absolute difference in neuronal activities per neuron across the three main task conditions. Here, dots represent each recorded neuron (n=1049 from 7 mice) and are colour-coded based on whether they each displayed significant differences in response to relative rank, reward and success. Primary comparisons were made between competitive success vs. reward (*Z=10.8, p=3.4x10-27; Rank-sum), competitive success vs. relative rank (**Z=7.48, p=7.7x10-14; Rank-sum) and relative rank vs. reward (***Z=10.29, p=8.14x10-25; Rank-sum). g, Polar plots illustrating the relative tuning of neurons that responded to differences in the animal’s relative rank, reward size and competitive success. For comparison, polar plots are also provided for neurons that responded to differences in speed (SP); travel distance (TD), social context (SO), mixture of controls (MC) and mixture of all task conditions (M). Polar plot for cells that responded to none of these task features is shown in grey. Dashed boxes represent significance limits for each condition (p<0.01; Kruskal-Wallis with one-sided Holm-Sidak correction for post hoc comparisons). The s.e.m. for each polar plot is given as dashed lines. h, Table displaying the average number of putative neurons and active electrode channels per recorded animal. i, Left, firing rates for neurons recorded from left and right hemispheres (t1047=1.64, p=0.10; two-sided t-test). Right, there was no difference in the encoding proportions for any of the main features of the group competitive foraging task between neurons recorded from the left versus right hemispheres (p>0.05, Chi-square tests). Box-plot edges represent 25th/75th percentiles with centre=median and whiskers=1st-99th percentile range.

Source data

Extended Data Fig. 6 Stability of neuronal encoding across animals, sessions and assay.

a, Normalized proportion of neuronal encoding competitive success, reward size, and relative rank across animals. b, An animal-dropping procedure revealed no difference in peak decoding performance for competitive success, reward size, and relative rank across animals (F(6,1399)=1.27, p=0.20 for competitive success, F(6,1399)=0.89, p=0.69 for reward size, and F(6,1399)=0.44, p=0.98 for relative rank; one-way ANOVA). c, Population firing rates and proportion of neurons encoding competitive success were stable across trials within sessions (n=63 sessions across 7 animals; F(1,116)=1.24, p=0.27 for firing rate; F(1,125)=0.30, p=0.59 for proportion of encoding neurons; two-way RM-ANOVA). d, Population firing rates and proportion of neurons encoding relative rank were stable across trials within sessions (n=63 sessions across 7 animals; F(1,116)=0.11, p=0.74 for firing rate; F(1,125)=0.013, p=0.909 for proportion of encoding neurons; two-way RM-ANOVA). e, Population firing rates and proportion of neurons encoding competitive success were stable across sessions (n=63 sessions across 7 animals; F(7,55)=0.67, p=0.76 for firing rate; F(7,55)=0.758, p=0.679 for proportion of encoding neurons; two-way RM-ANOVA). f, Population firing rates and proportion of neurons encoding relative rank were stable across sessions (n=63 sessions across 7 animals; F(7,55)=0.79, p=0.64 for firing rate; F(7,55)=1.709, p=0.098 for proportion of encoding neurons; two-way RM-ANOVA). g, Top, Graphic showing the two-chamber arena in which the recorded animal was placed with cagemates following each group foraging session. Bottom, Heat map and trajectories of both animals during a representative trial. h, Animals spent similar amount of time investigating the other animal regardless of their hierarchical rank relative to the recorded animal (χ2(5,304)=5.63, p=0.34; Kruskal-Wallis), but spent significantly less time investigating inanimate totems (Z=4.13, p=3.64x10-5; Rank-sum). Dots represent trials (n=376) across n=58 sessions. Box-plot edges represent 25th/75th percentiles with centre=median and whiskers=1st-99th percentile range. i, PETH and spike raster plots of two representative neurons that displayed changes in firing rate based on the animals’ relative rank (high vs. low social rank in relation to the other animal). PETHs are aligned to the time point at which the recorded animal initiated an investigation of the other animal. Purple dots represent the end of an investigation. Shaded area denotes s.e.m. j, Top, Representative neuron during wireless recordings of the group foraging task (10 min after start of a session) compared to that of the two-chamber assay (3 h after start). Shaded area denotes s.e.m. Bottom, Venn diagram depicting the number of neurons encoding relative rank (p<0.01; Wilcoxon signed-rank) during the two-chamber assay and their overlap with those encoding relative rank during the group foraging task (n=115 overlap; χ2(1)=42.17, p=8.35x10-11; Chi-Square test). k, Correlation of normalized firing rates for neurons that encoded relative rank during the group foraging vs. two-chamber assay on a cell-by-cell basis (n=115 RR-encoding of n=174 total rank-encoding neurons; r=0.673, p=2.85x10-24; Pearson correlation). Grey line depicts linear line of best fit. l, Decoding accuracy of relative rank during the two-chamber assay (n=942) was significantly higher than chance (p=0.0024, Permutation test) but lower than the peak decoding accuracy during the group foraging task (*p=0.013, Permutation test). m, SVM models trained on neuronal activity recorded during group foraging was used to decode validation data recorded during the two-chamber setup (i.e., switch model) or vice versa. Decoding accuracy for both switch models were significantly higher than expected from chance (p=0.0031 for group foraging model used to decode two-chamber data, p=0.021 for two-chamber model used to decode group foraging data; Permutation test). Overall, the group foraging model was better at decoding two-chamber rank rather than vice versa (p=0.026, Permutation test). Error bars denote mean±95%CI. N=500 bootstrapped samples for all decoding results.

Source data

Extended Data Fig. 7 Neuronal responses to physical interaction between animals during competitive bouts.

a, Graphic depicting an overtaking event where an animal (red, overtaking) overtakes another (yellow, overtaken) in the middle of running the trial. Right, overtaking animals were more likely to be higher ranking than the overtaken animals (n=4105 total overtaking events; *Z=-3.95, p=7.75x10-5; Signed-rank). b, Plot demonstrating the likelihood of being the overtaking animal based on absolute rank and position within the arena (n=4105 total overtaking events; Top, close to entrance point; Bottom, close to the staging area). c, Cumulative distribution function (CDF) demonstrating that overtaking events for higher ranked animals were more likely to occur farther away from the entrance point (closer to the staging area). N=4105 total overtaking events. d, The animals being overtaken were more likely to pause after an overtaking even when they were lower in rank than their competitors (*rs=0.14, p=1.21x10-10; Spearman correlation). Dots represent one overtaking/pausing bout (n=2074 total). Based on GLMs that further took into account the animals’ previous trial performance (Methods), there was a significant effect on the probability of a pausing event after an overtaking event based on rank difference (t3145=-2.77, p=0.0088), velocity difference (t3145=-11.02, p=2.99x10-28) proximity between animals (t3145=-5.68, p=1.32x10-8), race position (t3145=-6.53, p=2.99x10-13) and distance from reward (t3145=9.91, p=3.78x10-23). e, The mid-ranked (recorded) animals were more likely to be ranked higher than another when overtaking them and more likely to be ranked lower when being overtaken (n=1221 overtaking events involving mid-ranked mouse; *Z=-4.08, p=4.43x10-5; Signed-rank). f, Left, PETH and raster plots illustrating two representative cells that displayed a difference in their activities based on whether the recorded animal were overtaking another animal or being overtaken. Right, Most neurons that responded to differences in the animals’ competitive success across groupings displayed little response to physical factors such as proximity to the other animals (χ2(1)=232.3, p=1.88x10-52; Chi-Square test) or overtaking events at which time the recorded animals overtook another animal in their group (χ2(1)=214, p=1.82x10-48; Chi-Square test). g, Graphic depicting pausing behaviour, where an animal (yellow) pauses in the middle of running a trial. h, Lower ranked animals were more likely to pause while foraging with higher ranked animals no matter the spatial location. i, There was no difference in pause durations based on absolute rank (χ2(6,7129)=2.04, p=0.067; Kruskal-Wallis). j, Left, mid-rank (recorded) animals paused more often when they were lower in rank than their competitors (*rs=0.17, p=0.007) but displayed no difference in pausing behaviour when running with totems compared to group trials (Z=-2.28 p=0.26; Signed-rank). Right, The total duration of pauses by the mid-ranked animals were not influence by relative rank (rs=-0.02, p=0.85) but was significantly shorter when running with totem trials compared to group trials (Z=-4.76 p=1.9x10-6; Signed-rank). Dots represent session averages (n=63). k, Graphic depicting physical race positions based on the animals’ closeness to the reward zone in relation to others on trials that had an overtaking event. Right, the dominant animals were more likely to be in a higher race position. l, Left, Bar plots of two representative neurons that were tuned to the animals’ instantaneous position during a trial. Error bars denote mean ± s.e.m. Right, Most neurons that responded to differences in competitive success across groupings displayed little response to the animal’s physical position in relation to others (χ2(1)=205, p=1.66x10-46; Chi-Square test). m, Graphic depicting events in which the animals are crowded outside the entrance point, where the target animal (yellow) is in proximity with either two (Top) or one (Bottom) other competitor. n, Animals that were lower in relative rank were more likely to crowd with one or two others compared to animals that were higher in relative rank. N=3819 total crowding events. o, Heat map demonstrating that, during crowding events, animals of lower relative rank were more commonly partnered with other animals of lower relative rank. Error bars and shaded areas denote mean±95%CI.

Source data

Extended Data Fig. 8 Effect of past interactions, relative rank and reward on neuronal response.

a, Venn diagram depicting the number of neurons encoding the animals’ ordinal (i.e., first vs. second vs. third vs. fourth to reach reward entry zone) competitive success overlapping with neurons encoding binary (i.e., first two vs last two to reach reward entry zone) competitive success (p<0.01, FDR corrected for multiple epochs). Most neurons that encoded the animals’ ordinal and binary competitive success overlapped (n=120; χ2(1)=30.7, p=3x10-8; Chi-Square test). b, Decoding accuracy for ordinal competitive success for all task-modulated neurons (n=560; p<0.05; Permutation tests). Decoding accuracy gradually increased to a peak of 63.2±1.5% prior to reward zone entry (p<0.001, Permutation test). c, Venn diagram depicting the degree of overlap between neurons encoding the animals’ ordinal relative rank (i.e., first vs. second vs. third vs. fourth) and binary (i.e., two highest vs two lowest) relative rank (p<0.01, FDR corrected for multiple epochs). Neurons encoding the animals’ ordinal relative rank almost entirely overlapped with those that encoded the animal’s binary relative rank (n=122 overlap; χ2(1)=29.3, p=6.24x10-8; Chi-Square test). d, Using all task-modulated neurons (n=560), decoding accuracy for ordinal relative rank were significantly higher than chance over the course of the trials (p<0.0001; Permutation tests). e, Decoding accuracies for relative rank and reward size using only neurons that encoded relative rank (Left, n=87) or reward size (Right, n=156), respectively. f, Recent history effects. Top, example trial sequence and count of cumulative matching trials in succession. Bottom, number of ‘matching’ trials in succession when comparing wins vs losses at trial t (Z=1.013, p=0.31; Rank-sum) suggesting transient, short-lasting successions of wins and losses. N=4966 total trials. g, Venn diagrams depicting the number of neurons encoding competitive success in the past trial t-1 and their overlap with neurons encoding competitive success (Top) or relative rank (Bottom) in the present trial t (p<0.01, two-way ANOVA, FDR corrected for multiple epochs). Most neurons that encoded the animals past success (5.1%, n=54) overlapped with those that encoded their current success (n = 34; χ2(1)=1.31, p=0.25; Chi-Square test) but were largely distinct from those that encoded the animals’ relative rank (n=9; χ2(1)=12.2, p=0.00046; Chi-Square test). h, Using all task-modulated neurons (n=560), decoding accuracies for the previous trial competitive outcome (t-1) were not significantly different than chance (p>0.05; Permutation tests). Peak decoding accuracy for past success was 57.2±2.3% (H0=50% chance performance; p>0.05; Permutation tests). i, Using all task-modulated neurons (n=560), decoding accuracies for the animal’s current success (t) contingent on the previous trial’s competitive outcome (t-1) was significantly higher than chance (p<0.01; Permutation tests). These neurons predict the animal’s current success contingent on their past outcome with accuracy of up to 41.3±3.1% prior to trial onset (H0=25%, p<0.001; Permutation test). j, Using all task-modulated neurons (n=560), considering succession of wins or losses (behavioural states), decoding accuracy for the animals’ behavioural state prior to gate opening was 66.2±2.0% (H0=50% chance performance; p<0.01, Permutation test). k, Using all task-modulated neurons (n=560), peak decoding accuracy for the animals upcoming success contingent on their prior behavioural state prior to gate opening was 64.7±3.7% (H0=25% chance performance; p<0.001; Permutation test). l, GLMs were used to quantify the contribution of past behaviour to neural population response for competitive success (Methods). Time points in which the fraction of explained variance for the interaction between terms was higher than expected from chance (p<0.01; Permutation test). Trials are aligned to the gate opening or time point at which the recorded animal reached the reward zone. Shaded areas denote mean±95%CI. N=500 bootstrapped samples for all decoding results.

Source data

Extended Data Fig. 9 DREADD manipulation of the ACC selectively influences competitive effort but not motivation or reward-seeking behaviour.

a, b, Overlay of viral expression areas for a. hM3D(Gq)-mCherry across 13 animals and for b. hM4D(Gi)-mCherry across 12 animals. cg1/cg2=cingulate areas 1/2; prL=prelimbic cortex; iL=infralimbic cortex. c, Novel intruder assay for social aggression. Although the animals displayed an increase in social inspection behaviours following CNO compared saline (Left, *Z=1.77, p=0.038, one-sided rank sum), they displayed no change in attack or defensive behaviours (Right, p>0.3, two-sided Rank-sum). N=52 trials across n=15 animals. d, The likelihood of competitive success was higher for ACC excitation (Left, n=13, compared to saline) in trials where they were of lower relative rank (relative rank 3 or 4) than their competitors (*t(12)=2.63, p=0.011; Paired t-test), but not when they were higher in relative rank (relative rank 1 or 2) than their competitors (t(12)=-0.24, p=0.59; Paired t-test). Right, The likelihood of competitive success was decreased for ACC inhibition (n=12, compared to saline) in trials where they were of higher relative rank (relative rank 1 or 2) than their competitors (**t(11)=-2.11, p=0.028; Paired t-test), but not when they were of lower relative rank (relative rank 3 or 4) than their competitors (t(11)=-0.25, p=0.4; Paired t-test; Fig. 4e). e, Animals with ACC excitation reached the reward zone faster (n=13, *F(1,109)=3.98,p=0.049) in high compared to low reward trials (**F(1,109)=4.33,p=0.04) and when starting in the near compared to far staging areas (***F(1,109)=27,p=1.04x10-17), but there were no interactions between any of the conditions (Fdrug:reward(1,109)=0.23, p=0.63; Fdrug:stagingarea(1,109)=0.66, p=0.42; Freward:stagingarea(1,109)=0.52, p=0.47; three-way ANOVA). f, Animals with ACC inhibition reached the reward zone faster in high compared to low reward trials (n=12, *F(1,95)=5.1, p=0.026) and when starting in the near compared to far staging areas (**F(1,95)=4.1,p=0.046), but there were no interactions between any of the conditions (Fdrug:reward(1,95)=0.13, p=0.72; Fdrug:stagingarea(1,95)=0.043, p=0.84; Freward:stagingarea(1,95)=0.51, p=0.33; three-way ANOVA). g, Mice foraging alone with inanimate totems. Left, There was no difference in latency to reach reward for either ACC excitation (n=13 mice, t(12)=-0.058, p=0.96; Paired t-test) or inhibition (n=12 mice, t(11)=0.051, p=0.96; Paired t-test). Right, There was no difference in average path error to reaching the reward zone for either ACC excitation (t(12)=-0.79, p=0.44; Paired t-test) or inhibition (t(11)=0.47, p=0.65; Paired t-test). h, Mice moving a mass of variable weight to receive reward. Left, there was no difference in the latency to reach the reward zone based on differences in weight (low vs. high) for either ACC excitation (n=14 sessions across n=13 mice, F(2,77)=0.07, p=0.93; Two-way ANOVA) or inhibition (n=12 sessions across n=12 mice, F(2,67)=0.58, p=0.56; Two-way ANOVA). Right, There was no difference in the latency to push past the weight to reach the reward zone for either ACC excitation (F(2,77)=0.85, p=0.8; Two-way ANOVA) or inhibition (F(2,67)=0.01, p=0.99; Two-way ANOVA). Error bars denote mean ± s.e.m. Dots represent session averages. N=6 mice for Exc group and n=6 mice for Inh group.

Source data

Extended Data Fig. 10 Nonselective effects of vmPFC and NAc inhibition on competitive behaviour.

a, Schematic illustrating reversable musicmol inactivation in either the ventromedial prefrontal cortex (vmPFC) or the nucleus accumbens (NAc; Methods). b, Representative histological images displaying the muscimol injection sites for the vmPFC (Left) and NAc (Right). c, Animals injected with muscimol did not display a change in absolute social rank in the tube test compared to saline for either the vmPFC (Left, n=6 mice, Z=1, p=0.32, Signed-rank) or the NAc (Right, n=6 mice, Z=-1, p=0.32, Signed-rank). Error bars denote mean±95%CI. d, Group competition. Left, animals injected with muscimol in the vmPFC did not display a difference in the competitive order to reaching reward when compared to saline (p>0.2; Paired t-tests). Right, animals injected with muscimol in the NAc displayed a decrease in the competitive order to reaching reward when compared to saline, but this effect was observed both when the animals’ relative rank was either higher (*t(5)=2.51, p=0.027; Paired t-test) or lower (**t(5)=3.12, p=0.013; Paired t-test) than their competitors. Error bars denote mean ± s.e.m. e, Group competition. Left, animals injected with muscimol in the vmPFC did not display a difference in reaching reward first or last when compared to saline (>0.05; Paired t-tests). Right, animals injected with muscimol in the NAc displayed a decrease in likelihood of reaching reward first when compared to saline, but this effect was observed both when the animals’ relative rank was either higher t(5)=-2.17, p=0.041; Paired t-test) or lower (**t(5)=-2.8, p=0.019; Paired t-test) than their competitors. Error bars denote mean ± s.e.m. f, Foraging with totems. Left, animals injected with muscimol in the vmPFC did not display a difference in latency to reaching the reward zone (t(5)=-0.85, p=0.42; Paired t-test). Animals injected with muscimol in the NAc displayed an increase in latency (*t(5)=0.2.08, p=0.046; Paired t-test). Right, animals injected with muscimol in the vmPFC did not display a difference in the average path error to the reward zone (t(5)=0.17, p=0.87; Paired t-test). Animals injected with muscimol in the NAc also did not display a difference in the average path error to the reward zone NAc (t(5)=0.44, p=0.34; Paired t-test) together suggesting that NAc inactivation decreased the animals’ effort across both social and non-social conditions without affecting their overall motoric ability20. Error bars denote mean ± s.e.m. For panels c–f, n=6 mice for vmPFC group and n=6 mice for NAc group.

Source data

Supplementary information

Reporting Summary

Supplementary Video 1

Competitive foraging apparatus design. Video displaying a three-dimensional render of the competitive foraging apparatus. See Methods, Fig. 1a and Extended Data Fig. 1a for more details.

Supplementary Video 2

Example group competitive foraging trial. A representative trial depicting a group of four mice engaging the competitive foraging task. In this trial, the recorded animal (green fur) is the most dominant (red circle) relative to others (squares). The animals started from a common staging area and competed for food within the reward zone (grey). Kinematic position labels were created using custom-adapted automated tracking software (see methods). Timestamp shows the elapsing time from gate opening.

Source data

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Li, S.W., Zeliger, O., Strahs, L. et al. Frontal neurons driving competitive behaviour and ecology of social groups. Nature 603, 661–666 (2022). https://doi.org/10.1038/s41586-021-04000-5

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/s41586-021-04000-5

This article is cited by

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Search

Advanced search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing