Binary Mixtures of Intelligent Active Brownian Particles with Visual Perception

Gerhard Gompper; Rajendra Singh Negi; Roland G. Winkler

REVIEW 4 major objections 4 minor 72 references

Binary Mixtures of Intelligent Active Brownian Particles with Visual Perception

T0 review · 4 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Nine sign combinations of vision-based steering turn a binary active mixture into dimers, predator-prey chases, and honeycomb lattices.

desk verdict A solid, well-specified simulation catalogue of binary vision-based active particles; the predator-prey optimum and the enhanced-diffusion claim need qualification before I'd trust them. read the letter →

arxiv 2506.09698 v1 pith:YFNZMEEP submitted 2025-06-11 cond-mat.soft cond-mat.stat-mechphysics.bio-ph

Rajendra Singh Negi , Roland G. Winkler , Gerhard Gompper This is my paper · ORCID

classification cond-mat.softcond-mat.stat-mechphysics.bio-ph

keywords activeBrownianparticlesvisualperceptionself-steeringnonreciprocalinteractionsbinarymixturespredator-preydynamicshoneycomblatticehoppertransport

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper claims that a minimal model of two species of self-steering active Brownian particles, each turning toward or away from neighbors it sees in a forward vision cone, is enough to generate a broad catalog of nonequilibrium structures. The paper sweeps all nine principal sign combinations of the four steering couplings and reports that they organize the mixture into dimers and multimer aggregates, encapsulated clusters, honeycomb-like lattices, and predator-prey pursuit. Two results stand out: with charge-like steering (same types avoid, opposite types attract), excess particles hop between dimers in a way the paper compares to the Grotthuss proton-shuttle mechanism, and off-stoichiometric mixtures diffuse faster than ordinary active Brownian particles at intermediate activity; with nonreciprocal steering, an A predator chasing B prey hunts best with a focused vision half-angle near $\pi/4$. If the model is right, these structures are robust emergent outcomes of the stated rules, which could guide the design of micro-robots and help interpret biological swarming and hunting.

What carries the argument

The central object is the intelligent active Brownian particle (iABP), a self-propelled disk whose orientation obeys a cognitive torque from every neighbor inside its forward vision cone of half-angle $\theta_\alpha$ and finite range $R_v$. The torque is proportional to the maneuverability $\Omega_{\alpha\gamma}$ and is normalized by the weighted number of visible particles, so dense surroundings shorten the effective vision range by blocking the view. Because the couplings $\Omega_{AB}$ and $\Omega_{BA}$ need not be equal, the steering can be nonreciprocal, which is the mechanism behind the predator-prey phase. The sign pattern of the four maneuverabilities, together with the Péclet number (the dimensionless activity), vision angle, vision range, and composition, is the control parameter that organizes the system into dimers, hoppers, aggregates, honeycomb lattices, and chases.

What would settle it

A direct test is to rerun the predator-prey simulations with the predator's Péclet number twice the prey's while keeping the vision cone and maneuverabilities fixed, and to check whether the front-accumulation and back-depletion profile and the optimum near $\theta_A=\pi/4$ survive; a second test is to compute a quantitative hexatic order parameter for the honeycomb phase over many independent realizations to see whether the visually assigned lattice persists.

Watch

Extended reading notes

Core claim

The central claim is that vision-based steering alone, with no physical forces between the two species beyond repulsion at contact, produces a wide variety of emergent collective behaviors characterized by the signs of the maneuverabilities $\Omega_{\alpha\gamma}$ that set whether type $\alpha$ turns toward or away from type $\gamma$. With same-type repulsion and opposite-type attraction, the system forms stable A-B dimers, and any stoichiometric excess particles act as hoppers that travel between dimers, replace a partner, and leave as a new hopper, in a caged-then-diffusive dynamics reminiscent of the Grotthuss mechanism. With a nonreciprocal chase interaction ($\Omega_{AB}>0$, $\Omega_{BA}<0$, both like-type couplings attractive), predator-prey pursuit emerges in which prey deplete behind the predator and accumulate in front, with the strongest front accumulation at a predator vision half-angle around $\pi/4$. With one species aggregating, the other dispersing, and mutual avoidance, a honeycomb-like lattice forms whose cluster size and spacing grow with the vision cutoff range $R_v$. The paper further finds that off-stoichiometric charge-like mixtures at intermediate Péclet number diffuse faster than non-steering active Brownian particles.

Load-bearing premise

The load-bearing premise is that predator and prey swim at the same speed throughout the predator-prey study, and that the phase catalog is read from single final snapshots by eye rather than from quantitative order parameters.

Editorial extensions

If this is right

Charge-like steering produces stable heterodimers, and a slight excess of one species creates hopper particles that exchange into and out of dimers, giving subdiffusive short-time and diffusive long-time motion.
In off-stoichiometric charge-like mixtures at intermediate activity, the effective long-time diffusion coefficient exceeds that of non-steering active Brownian particles because temporary clusters with inhomogeneous orientations increase persistent motion.
For nonreciprocal steering with A chasing B and B fleeing A, prey accumulate ahead of the predator and deplete behind it, and the predator's front prey density is highest for a vision half-angle near $\theta_A=\pi/4$.
Increasing the vision cutoff range $R_v$ in the honeycomb phase grows the average B-cluster size and the lattice spacing, with A particles at hexagon boundaries acting as a barrier that halts coarsening.
The nine-sign sweep yields a systematic phase overview in which the Péclet number and vision angle control transitions among mixed aggregates, segregated aggregates, dimers, encapsulated clusters, and honeycomb structures.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

If the equal-speed constraint on predator and prey were relaxed, the prey density profile and the claimed $\pi/4$ optimum would likely change; a faster predator should thin the prey ahead and sharpen the depletion behind, a case the paper does not simulate.
A quantitative test of the phase labels is still open: computing bond-orientational order parameters or cluster-size distributions over many realizations would show whether the honeycomb and aggregate phases are sharply defined or depend on the chosen snapshot.
The Grotthuss analogy suggests a coarse-grained kinetic model of hopper exchange in which dimers act as reactive sites with exchange rates set by vision range, stoichiometric excess, and Péclet number; the paper does not formulate such a model.
The honeycomb result implies a practical tuning rule for programmable microrobots: lattice wavelength can be set by choosing the visual interaction range, provided the A-barrier coarsening arrest observed here persists in experimental implementations.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Worth a read. This is a genuine extension of the authors' iABP model to binary mixtures with nonreciprocal steering, and the paper does two things well: it lays out the equations of motion cleanly (Eqs. 1-9) and it maps out a broad phase catalog across the nine sign combinations of the maneuverabilities. The hopper exchange dynamics (Grotthuss analogy) is a nice, new observation, and the honeycomb-lattice formation controlled by vision range is a concrete, testable prediction. I don't see any post-hoc fitting; the structures are direct outputs of the stated model.

The soft spots are real but mostly fixable. First, the phase taxonomy is assigned by eye from end-of-run snapshots. For a paper claiming 'systematic characterization,' I'd want at least one quantitative order parameter per phase and error bars on the diffusion coefficients, hopper statistics, and prey-density profiles. Up to 10 realizations isn't enough for some of the fine claims (e.g., the difference between θ_A=π/4 and neighboring angles in Fig. 11). Second, the abstract's enhanced-diffusion claim is tied to 'non-stoichiometric composition,' but the diffusion data in Fig. 4 is for a nearly 1:1 mixture. That's a mismatch: either show the non-stoichiometric data or rephrase. Third, the predator-prey optimum at π/4 is only tested with equal Péclet numbers for predator and prey. The paper states this in the text, but the conclusion and abstract drop the caveat. The stress-test note is right: speed asymmetry could shift or erase that optimum. It's a limitation, not a contradiction, but it needs to be flagged wherever the design principle is advertised. Finally, the Fig. 4 caption contains an editorial artifact ('Will (a) be removed?')—that's a clear sign the manuscript needs a cleanup pass.

The math and simulation protocol look sound, and the citation pattern is appropriate; the authors are extending their own prior work, which is fine. This paper will be useful to people working on cognitive active matter and swarm robotics, as it gives a concrete map of what to expect from this class of models.

My recommendation: do send it to peer review. The referee should ask for quantitative phase measures, error bars, a qualified predator-prey claim, and a cleanup of the abstract and caption. After that, it's a solid addition.

Referee Report

4 major / 4 minor

Summary. This paper proposes a binary-mixture model of 'intelligent' active Brownian particles (iABPs) with visual perception: each particle steers toward or away from same- or other-species neighbors within a vision cone, with species-dependent maneuverabilities that may be nonreciprocal. The authors simulate all nine principal sign combinations of the four maneuverabilities and report a catalog of emergent structures, including dimers and multimers, encapsulated aggregates, honeycomb-like lattices, and predator-prey pursuit. The main quantitative analyses concern hopper transport in charge-like mixtures, the dependence of prey-density profiles on predator vision angle and maneuverability, and the role of the vision cutoff radius in honeycomb-lattice formation. The central advertised results are an optimal predator vision angle near pi/4, an enhanced diffusion at intermediate activity for charge-like systems, and the robustness of the honeycomb phase to changes in vision range.

Significance. If the results hold, the paper provides a useful minimal simulation framework and a systematic phase catalog for nonreciprocal cognitive active matter, with direct implications for the design of micro-robotic swarms and for interpreting nonreciprocal interactions in biological collectives. The paper's strengths are that the equations of motion and integration protocol are explicit, the data are openly deposited, and the central structures are emergent outputs of the stated model rather than outcomes of parameter fitting. The predator-prey density asymmetry and the hopper-exchange mechanism are qualitatively interesting and could guide future experiments. The main limitations are that the predator-prey optimum is established only for equal bare speeds of predator and prey, that several quantitative claims are made without uncertainty quantification, and that the phase taxonomy rests largely on visual inspection of single snapshots.

major comments (4)

[Section IV / Section VI] The predator-prey analysis is performed under the explicitly stated restriction that A and B particles have the same Peclet number, hence the same bare speed. The claimed optimal predator vision angle near pi/4 in Section VI, and the general discussion of predator-prey pursuit in the abstract, are based solely on this equal-speed case. In most natural and engineered predator-prey systems the predator moves faster than the prey, and the density depletion/accumulation profiles in Figs. 10 and 11 are measured in a reference frame in which the prey can escape sidewise. I request either a systematic variation of the speed ratio (e.g., Pe_A/Pe_B between 1 and 5) or a clear restriction of the optimum claim to the equal-speed case, with the abstract and conclusions revised accordingly.
[Section IV, Figs. 10-11] The existence of the 'eagle's eye' optimum at theta_A = pi/4 is inferred from prey-density profiles that are described only as averages over 'several encounters'; no error bars, number of encounters, or statistical test are provided. In Fig. 11(a) the difference between theta_A = pi/4 and the neighboring angles appears modest, and without uncertainty quantification it is not established that the optimum is significant. Please provide error bars or bootstrap confidence intervals for the profiles and for the extracted optimal angle.
[Section III.A, Figs. 2-3] The phase labels used throughout the paper (dimers, encapsulated aggregates, segregated aggregates, honeycomb lattices, and so forth) are assigned by visual inspection of single end-of-run snapshots, with no quantitative order parameters or ensemble reproducibility check. Since the paper claims to 'systematically characterize' all nine sign combinations, the taxonomy should be supported by at least a small set of quantitative measures, such as cluster-size distributions, species-mixing ratios, bond-orientational order parameters, or time-averaged structure factors, applied to the representative states in Figs. 2 and 3.
[Section III.B, Fig. 4(b)] The abstract's claim of enhanced diffusion relative to non-steering active Brownian particles is not directly demonstrated. Fig. 4(b) plots a Peclet-scaled effective diffusion coefficient without overlaying the non-steering ABP baseline or showing statistical uncertainties. The text states that the theta = pi/16 results are consistent with the non-steering value, but the intermediate-angle enhancement should be quantified by an explicit comparison to the baseline over the reported parameter range, with error bars from the independent realizations.

minor comments (4)

[Fig. 4 caption] The caption contains the editorial query '(Will (a) be removed?)'; this manuscript artifact must be deleted before publication.
[Fig. 12 caption] The caption reads 'Omega_aa = -12.5, Omega_aa = 12.5', which is presumably a typo for 'Omega_aa = -12.5, Omega_bb = 12.5'; please correct it.
[Section VI, final paragraph] There is a typo: 'the large space of of self-organization behavior' should read 'the large space of self-organization behavior'.
[Section III.C] The statement that 'passivated' particles in clusters diffuse with an effective coefficient ~1/N_e is asserted without derivation or a direct test; a brief scaling argument or a reference would improve the readability.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the paper is a simulation study whose equations of motion are fully restated, and all reported structures and dynamics are direct outputs of the stated model rather than fits or imported conclusions.

full rationale

The paper's central claims — multimers, hopper transport, predator-prey pursuit with an optimal predator vision angle near pi/4, and honeycomb-lattice formation — are extracted from simulations of the model defined by Eqs. (1)–(8), which are stated completely in the manuscript. The model is inherited from the authors' earlier works (Refs. 44 and 48), but the governing equations are reproduced in full, so a reader can implement and test the model without consulting those papers; the citations are therefore provenance, not load-bearing self-reference. No parameter is fitted to reproduce a target emergent structure, and the reported phase snapshots and correlation functions are direct measurements rather than predictions derived from fitted inputs. The predator-prey optimal-angle claim is obtained under the explicitly stated constraint that both species have the same Peclet number and hence the same speed, which limits the generality of the claim but is a transparent scope condition, not a circular reduction. Likewise, the honeycomb-lattice dependence on vision range R_v is a controlled parameter sweep, not a quantity defined in terms of the outcome. The reader's concern about phase classification by visual inspection of snapshots is a methodological caveat about robustness, not circularity. Overall, the derivation chain is self-contained and the emergent behaviors are genuine outputs of the stated equations.

Assumptions & free parameters 5 free parameters · 3 assumptions · 1 invented entities

The paper's results are generated by a hand-posed model: the steering torque (Eqs. 3 to 5) is postulated, and all interaction parameters are chosen by hand rather than fitted or measured, with no experimental calibration. Scanned control parameters (Péclet number, vision angle, packing fraction, number ratio) are deliberately varied to map the phase space and are not listed as free parameters; the entries above are the hand-chosen interaction and analysis parameters that the headline structures directly depend on.

free parameters (5)

Maneuverability magnitude Omega_0 (all four Omega_alpha_beta set to +/-Omega_0) = 12.5, 62.5, and up to 200 in predator-prey runs, in units of D_R
Steering strength is a hand-chosen input that directly controls all reported phases (Figs. 2 and 3); no calibration to biological or robotic data is offered.
Visual perception range R_0 = 1.5 sigma
Sets the exponential decay of the steering torque in Eq. (4) and hence the effective vision range in dense systems; chosen by hand, not measured.
Vision cutoff radius R_v = 4 R_0 = 6 sigma, varied up to 14 sigma
The honeycomb-lattice phase appears only for large R_v, and cluster size and lattice structure grow with R_v (Section V, Figs. 12 and 13), making this a load-bearing hand-chosen parameter.
Repulsion strength epsilon = (1 + P_e) k_B T
Chosen to keep particle overlap nearly constant at all activities (Section II.B); affects collision dynamics throughout.
Hopper and cluster distance cutoffs = h_d = 1.5 sigma for hoppers; r <= 1.5 sigma for clusters
Operational definitions that set which particles count as hoppers and clusters (Sections III.C and V); results such as N_hopp and C(t) depend on these hand-chosen thresholds.

assumptions (3)

standard math The velocity-Verlet Langevin integrator of Grønbech-Jensen and Farago (Ref. 61) correctly samples the stochastic dynamics of Eqs. (1) and (8) over 2 x 10^7 steps.
All reported statistics (MSDs, pair correlations, cluster sizes) inherit this assumption; no verification or error analysis is presented.
ad hoc to paper The N_alpha_gamma-normalized steering torque of Eqs. (4) and (5) is an appropriate model of vision-limited perception for animals and robots.
The normalization is motivated by 'blocking the view' and Ref. 58, but it is a postulated interaction kernel with no quantitative validation against biological or robotic sensing.
domain assumption Periodic 2D boxes with 625 to 1000 particles, packing fractions 0.00785 to 0.0785, and D_R t approximately 1600 reach steady state, and snapshot-based phase assignments represent distinct phases rather than transients.
No finite-size, finite-time, or convergence analysis is provided; phase diagrams rest on visual inspection of single end-of-run snapshots (Figs. 2, 3, and 12).

invented entities (1)

Species-asymmetric vision-cone steering interaction (nonreciprocal maneuverabilities Omega_alpha_beta defined by Eqs. 3 to 5)
purpose: Generates all reported emergent behaviors: multimers, encapsulated aggregates, honeycomb lattices, hopper transport, and predator-prey pursuit.
A postulated minimal interaction, not derived from or calibrated to any measured sensory-motor system; the only external handles are qualitative analogies (bee swarms, reef sharks) and qualitative design suggestions for micro-robots, none tested here.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Binary Mixtures of Intelligent Active Brownian Particles with Visual Perception." pith.science (2026). https://pith.science/paper/YFNZMEEP

@misc{pith2026250609698,
  author       = {Pith},
  title        = {Pith review of: Binary Mixtures of Intelligent Active Brownian Particles with Visual Perception},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/YFNZMEEP}},
  note         = {Machine review of arXiv:2506.09698}
}

read the original abstract

The collective properties of a binary mixture of A- and B-type self-steering particles endowed with visual perception are studied by computer simulations. Active Brownian particles are employed with an additional steering mechanism, which enables them to adjust their propulsion direction relative to the instantaneous positions of neighboring particles, depending on the species, either steering toward or away from them. Steering can be nonreciprocal between the A- and B-type particles. The underlying dynamical and structural properties of the system are governed by the strength and polarity of the maneuverabilities associated with the vision-induced steering. The model predicts the emergence of a large variety of nonequilibrium behaviors, which we systematically characterize for all nine principal sign combinations of AA, BB, AB and BA maneuverabilites. In particular, we observe the formation of multimers, encapsulated aggregates, honeycomb lattices, and predator-prey pursuit. Notably, for a predator-prey system, the maneuverability and vision angle employed by a predator significantly impacts the spatial distribution of the surrounding prey particles. For systems with electric-charge-like interactions and non-stochiometric composition, we obtain at intermediate activity levels an enhanced diffusion compared to non-steering active Brownian particles.

Figures

Figures reproduced from arXiv: 2506.09698 by the authors.

**Figure 2.** FIG. 2. Snapshots of phases for 9 combinations of the maneuverabilities and the P´eclet numbers 1 [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗

**Figure 3.** FIG. 3. Snapshot of phases for the maneuverabilites Ω [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗

Figures from the paper (7 more)

**Figure 5.** Figure 5: FIG. 5. Snapshots of the hopper-pair exchange dynamics. [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]

**Figure 6.** Figure 6: FIG. 6. Activity dependence of the number of hoppers. Num [PITH_FULL_IMAGE:figures/full_fig_p008_6.png]

**Figure 8.** Figure 8: FIG. 8. Dynamics of [PITH_FULL_IMAGE:figures/full_fig_p009_8.png]

**Figure 9.** Figure 9: FIG. 9. Snapshots of a predator-prey pursuit. Series of snapshots (zoomed) progressing in time from (a)-(d), which illustrate [PITH_FULL_IMAGE:figures/full_fig_p011_9.png]

**Figure 10.** Figure 10: FIG. 10. Distribution of the number of prey particles around [PITH_FULL_IMAGE:figures/full_fig_p011_10.png]

**Figure 12.** Figure 12: FIG. 12. Snapshots of honeycomb-type lattices. The various [PITH_FULL_IMAGE:figures/full_fig_p012_12.png]

**Figure 13.** Figure 13: FIG. 13. (a) Pair correlation-function of [PITH_FULL_IMAGE:figures/full_fig_p012_13.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

72 extracted references · 61 canonical work pages

[1]

Interestingly, at smallP e= 1.25 the number of hoppers is larger compared to higherP e, and increases only slightly as the number differenceN d in- creases

Hopper number, encounter distance, displacement, and hopping time Figures 6(a) and (b) show the average numberN hopp of hoppers present in the system at various activities and packing fractions. Interestingly, at smallP e= 1.25 the number of hoppers is larger compared to higherP e, and increases only slightly as the number differenceN d in- creases. This ...

work page
[2]

hunting packs

Mean-Square Displacement: Caging and Chasing To characterize the dynamics of iABPs in the ”hop- ping phase”, we evaluate the mean-square displacement (MSD) of theAandBparticles. The characterization of the dynamics of hoppers itself is only possible for rather short times due to frequent recombination and exchange withA-Bpairs. The presence of the majorit...

work page
[3]

H. C. Berg, Bacterial behaviour, Nature254, 389 (1975)

work page 1975
[4]

Be’er, B

A. Be’er, B. Ilkanaiv, R. Gross, D. B. Kearns, S. Heiden- reich, M. B¨ ar, and G. Ariel, A phase diagram for bacterial swarming, Commun. Phys.3, 66 (2020)

work page 2020
[5]

I. S. Aranson, Bacterial active matter, Rep. Prog. Phys. 85, 076601 (2022)

2022
[6]

Cavagna and I

A. Cavagna and I. Giardina, Bird flocks as condensed matter, Annu. Rev. Cond. Mat. Phys.5, 183 (2014)

work page 2014
[7]

D. J. Pearce, A. M. Miller, G. Rowlands, and M. S. Turner, Role of projection in the control of bird flocks, Proc. Natl. Acad. Sci. USA111, 10422 (2014). 14

work page 2014
[8]

Vicsek, A

T. Vicsek, A. Czir´ ok, E. Ben-Jacob, I. Cohen, and O. Shochet, Novel type of phase transition in a system of self-driven particles, Phys. Rev. Lett.75, 1226 (1995)

1995

Show all 72 references

[9]

Elgeti, R

J. Elgeti, R. G. Winkler, and G. Gompper, Physics of mi- croswimmers—single particle motion and collective be- havior: a review, Rep. Prog. Phys.78, 056601 (2015)

2015
[10]

Bechinger, R

C. Bechinger, R. Di Leonardo, H. L¨ owen, C. Reichhardt, G. Volpe, and G. Volpe, Active particles in complex and crowded environments, Rev. Mod. Phys.88, 045006 (2016)

2016
[11]

Gompperet al., The 2020 motile active matter roadmap, J

G. Gompperet al., The 2020 motile active matter roadmap, J. Phys. Condens. Matter32, 193001 (2020)

2020
[12]

Gompperet al., The 2025 motile active matter roadmap, J

G. Gompperet al., The 2025 motile active matter roadmap, J. Phys. Condens. Matter37, 143501 (2025)

2025
[13]

B.-W. Park, J. Zhuang, O. Yasa, and M. Sitti, Multifunc- tional bacteria-driven microswimmers for targeted active drug delivery, ACS nano11, 8910 (2017)

2017
[14]

B¨ auerle, A

T. B¨ auerle, A. Fischer, T. Speck, and C. Bechinger, Self- organization of active particles by quorum sensing rules, Nat. Commun.9, 1 (2018)

2018
[15]

Bastos-Arrieta, A

J. Bastos-Arrieta, A. Revilla-Guarinos, W. E. Uspal, and J. Simmchen, Bacterial biohybrid microswimmers, Front. Robot. AI5, 97 (2018)

2018
[16]

S. Yan, X. Zeng, Y. Wang, and B.-F. Liu, Biomineraliza- tion of bacteria by a metal–organic framework for thera- peutic delivery, Adv. Healthc. Mater.9, 2000046 (2020)

2020
[17]

Sitti, H

M. Sitti, H. Ceylan, W. Hu, J. Giltinan, M. Turan, S. Yim, and E. Diller, Biomedical applications of un- tethered mobile milli/microrobots, Proc. IEEE103, 205 (2015)

2015
[18]

Y. Fu, H. Yu, X. Zhang, P. Malgaretti, V. Kishore, and W. Wang, Microscopic swarms: From active mat- ter physics to biomedical and environmental applications, Micromachines13, 295 (2022)

2022
[19]

J. Hu, S. Zhou, Y. Sun, X. Fang, and L. Wu, Fabrication, properties and applications of janus particles, Chem. Soc. Rev.41, 4356 (2012)

2012
[20]

Z. You, A. Baskaran, and M. C. Marchetti, Nonreciproc- ity as a generic route to traveling states, Proc. Natl. Acad. Sci.117, 19767 (2020)

2020
[21]

S. Saha, J. Agudo-Canalejo, and R. Golestanian, Scalar active mixtures: The nonreciprocal cahn-hilliard model, Phys. Rev. X10, 041009 (2020)

2020
[22]

Osat and R

S. Osat and R. Golestanian, Non-reciprocal multifarious self-organization, Nat. Nanotechnol.18, 79 (2023)

2023
[23]

C. H. Meredith, P. G. Moerman, J. Groenewold, Y.-J. Chiu, W. K. Kegel, A. van Blaaderen, and L. D. Zarzar, Predator–prey interactions between droplets driven by non-reciprocal oil exchange, Nat. Chem.12, 1136 (2020)

2020
[24]

Natan, V

G. Natan, V. M. Worlitzer, G. Ariel, and A. Be’er, Mixed- species bacterial swarms show an interplay of mixing and segregation across scales, Sci. Rep.12, 16500 (2022)

2022
[25]

Q. Wang, J. G. Frye, M. McClelland, and R. M. Harshey, Gene expression patterns during swarming in salmonella typhimurium: genes specific to surface growth and puta- tive new motility and pathogenicity genes, Mol. Micro- biol.52, 169 (2004)

2004
[26]

S. E. Burriel and R. Colin, Active density fluctuations in bacterial binary mixtures, bioRxiv , 2023 (2023)

2023
[27]

Tanget al., A predator-prey interaction between a marine pseudoalteromonas sp

B.-L. Tanget al., A predator-prey interaction between a marine pseudoalteromonas sp. and gram-positive bacte- ria, Nat. Commun.11, 285 (2018)

2018
[28]

J. R. Seymour, D. R. Brumley, R. Stocker, and J.-B. Raina, Swimming towards each other: the role of chemo- taxis in bacterial interactions, Trends Microbiology.32, 640 (2024)

2024
[29]

Fetecau and A

R. Fetecau and A. Guo, A mathematical model for flight guidance in honeybee swarms, Bull. Math. Biol.74, 2600 (2012)

2012
[30]

Beekman, R

M. Beekman, R. L. Fathke, and T. D. Seeley, How does an informed minority of scouts guide a honeybee swarm as it flies to its new home?, Anim. Behav.71, 161 (2006)

2006
[31]

Sengupta, T

A. Sengupta, T. Kruppa, and H. L¨ owen, Chemotactic predator-prey dynamics, Phys. Rev. E83, 031914 (2011)

2011
[32]

N. J. Cira, A. Benusiglio, and M. Prakash, Vapour- mediated sensing and motility in two-component droplets, Nature519, 446 (2015)

2015
[33]

Bialk´ e, T

J. Bialk´ e, T. Speck, and H. L¨ owen, Crystallization in a dense suspension of self-propelled particles, Phys. Rev. Lett.108, 168301 (2012)

2012
[34]

Wysocki, R

A. Wysocki, R. G. Winkler, and G. Gompper, Co- operative motion of active Brownian spheres in three- dimensional dense suspensions, EPL (Europhys. Lett.) 105, 48004 (2014)

2014
[35]

M. E. Cates and J. Tailleur, Motility-induced phase sepa- ration, Annu. Rev. Condens. Matter Phys.6, 219 (2015)

2015
[36]

Digregorio, D

P. Digregorio, D. Levis, A. Suma, L. F. Cugliandolo, G. Gonnella, and I. Pagonabarraga, Full phase diagram of active Brownian disks: From melting to motility-induced phase separation, Phys. Rev. Lett.121, 098003 (2018)

2018
[37]

Peruani, A

F. Peruani, A. Deutsch, and M. B¨ ar, Nonequilibrium clustering of self-propelled rods, Phys. Rev. E74, 030904 (2006)

2006
[38]

Theers, E

M. Theers, E. Westphal, K. Qi, R. G. Winkler, and G. Gompper, Clustering of microswimmers: interplay of shape and hydrodynamics, Soft Matter14, 8590 (2018)

2018
[39]

Stenhammar, R

J. Stenhammar, R. Wittkowski, D. Marenduzzo, and M. E. Cates, Activity-induced phase separation and self- assembly in mixtures of active and passive particles, Phys. Rev. Lett.114, 018301 (2015)

2015
[40]

Wysocki, R

A. Wysocki, R. G. Winkler, and G. Gompper, Propagat- ing interfaces in mixtures of active and passive Brownian particles, New J. Phys.18, 123030 [1 (2016)

2016
[41]

Wittkowski, J

R. Wittkowski, J. Stenhammar, and M. E. Cates, Nonequilibrium dynamics of mixtures of active and pas- sive colloidal particles, New J. Phys.19, 105003 (2017)

2017
[42]

Kolb and D

T. Kolb and D. Klotsa, Active binary mixtures of fast and slow hard spheres, Soft Matter16, 1967 (2020)

2020
[43]

D. P. Singh, U. Choudhury, P. Fischer, and A. G. Mark, Non-equilibrium assembly of light-activated col- loidal mixtures, Adv. Mater.29, 1701328 (2017)

2017
[44]

Jhajhria, S

M. Jhajhria, S. Sahoo, T. Biswas, and S. Thakur, Activ- ity induced non-monotonic aggregation in a mixture of chemically active and passive particles, Soft Materials , 1 (2023)

2023
[45]

Gao and I

Z. Gao and I. D. Couzin, Swarming transitions of self- propelled particles with anisotropic social interactions, Phys. Rev. Research7, 023144 (2025)

2025
[46]

R. S. Negi, R. G. Winkler, and G. Gompper, Collec- tive behavior of self-steering active particles with velocity alignment and visual perception, Phys. Rev. Research6, 013118 (2024)

2024
[47]

Saavedra, G

R. Saavedra, G. Gompper, and M. Ripoll, Swirling due to misaligned perception-dependent motility, Phys. Rev. Lett.132, 268301 (2024)

2024
[48]

I. D. Couzin, J. Krause, R. James, G. D. Ruxton, and N. R. Franks, Collective memory and spatial sorting in animal groups, J. Theor. Biol.218, 1 (2002). 15

2002
[49]

Barberis and F

L. Barberis and F. Peruani, Large-scale patterns in a minimal cognitive flocking model: Incidental leaders, ne- matic patterns, and aggregates, Phys. Rev. Lett.117, 248001 (2016)

2016
[50]

R. S. Negi, R. G. Winkler, and G. Gompper, Emergent collective behavior of active Brownian particles with vi- sual perception, Soft Matter18, 6167 (2022)

2022
[51]

Liu and M

Z. Liu and M. Dijkstra, Collective dynamics of intelligent active brownian particles with visual perception and ve- locity alignment in 3d: Spheres, rods, and worms, Soft Matter (2025)

2025
[52]

Chatterjee, M

S. Chatterjee, M. Mangeat, C.-U. Woo, H. Rieger, and J. D. Noh, Flocking of two unfriendly species: The two- species vicsek model, Phys. Rev. E107, 024607 (2023)

2023
[53]

N. K. Agrawal and P. S. Mahapatra, Alignment-mediated segregation in an active-passive mixture, Phys. Rev. E 104, 044610 (2021)

2021
[54]

Agudo-Canalejo and R

J. Agudo-Canalejo and R. Golestanian, Active phase sep- aration in mixtures of chemically interacting particles, Phys. Rev. Lett.123, 018101 (2019)

2019
[55]

Maity and A

S. Maity and A. Morin, Spontaneous demixing of binary colloidal flocks, Phys. Rev. Lett.131, 178304 (2023)

2023
[56]

S. Das, G. Gompper, and R. G. Winkler, Confined active Brownian particles: theoretical description of propulsion- induced accumulation, New J. Phys.20, 015001 (2018)

2018
[57]

S. Goh, R. G. Winkler, and G. Gompper, Noisy pur- suit and pattern formation of self-steering active parti- cles, New J. Phys.24, 093039 (2022)

2022
[58]

R. S. Negi, P. Iyer, and G. Gompper, Controlling inter- particle distances in crowds of motile, cognitive, active particles, Sci. Rep.14, 9443 (2024)

2024
[59]

P. Iyer, R. S. Negi, A. Schadschneider, and G. Gompper, Directed motion of cognitive active agents in a crowded three-way intersection, Commun. Phys.7, 379 (2024)

2024
[60]

Chepizhko, D

O. Chepizhko, D. Saintillan, and F. Peruani, Revisiting the emergence of order in active matter, Soft Matter17, 3113 (2021)

2021
[61]

See Supplemental Material at [URL by publisher] for sim- ulation movies
[62]

Mandal, B

S. Mandal, B. Liebchen, and H. L¨ owen, Motility-induced temperature difference in coexisting phases, Phys. Rev. Lett.123, 228001 (2019)

2019
[63]

Grønbech-Jensen and O

N. Grønbech-Jensen and O. Farago, A simple and effec- tive verlet-type algorithm for simulating langevin dynam- ics, Mol. Phys.111, 983 (2013)

2013
[64]

Agmon, The grotthuss mechanism, Chem

N. Agmon, The grotthuss mechanism, Chem. Phys. Lett. 244, 456 (1995)

1995
[65]

See https://iso.500px.com/parting-of-the-feast-behind- this-remarkable-photo-of-a-reef-shark-parting-a-school- of-fish/

A picture of a reef shark in a school of fish was taken in 2012 in Australia by Scott Carr. See https://iso.500px.com/parting-of-the-feast-behind- this-remarkable-photo-of-a-reef-shark-parting-a-school- of-fish/

2012
[66]

Touahmi, N

Y. Touahmi, N. Burlutskiy, K. Lee, and B. Lee, Conges- tion avoidance for multiple micro-robots using the be- haviour of fish schools, Int. J. Adv. Robotic Sy.9, 67 (2012)

2012
[67]

A. C. Tsang, E. Demir, Y. Ding, and O. S. Pak, Roads to smart artificial microswimmers, Adv. Intell. Syst.2, 1900137 (2020)

2020
[68]

J. Chen, X. Lei, Y. Xiang, M. Duan, X. Peng, and H. P. Zhang, Emergent chirality and hyperuniformity in an ac- tive mixture with nonreciprocal interactions, Phys. Rev. Lett.132, 118301 (2024)

2024
[69]

Fruchart, R

M. Fruchart, R. Hanai, P. B. Littlewood, and V. Vitelli, Non-reciprocal phase transitions, Nature592, 363 (2021)

2021
[70]

Suchanek, K

T. Suchanek, K. Kroy, and S. A. M. Loos, Irreversible mesoscale fluctuations herald the emergence of dynamical phases, Phys. Rev. Lett.131, 258302 (2023)

2023
[71]

K. L. Kreienkamp and S. H. L. Klapp, Nonreciprocal alignment induces asymmetric clustering in active mix- tures, Phys. Rev. Lett.133, 258303 (2024)

2024
[72]

Addi- tional data can be provided by the authors upon rea- sonable request

Exemplary simulation datasets corresponding to all dis- cussed parameter sets are available in the Zenodo repos- itory: https://doi.org/10.5281/zenodo.15208145. Addi- tional data can be provided by the authors upon rea- sonable request

doi

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.

[1] [1]

Interestingly, at smallP e= 1.25 the number of hoppers is larger compared to higherP e, and increases only slightly as the number differenceN d in- creases

Hopper number, encounter distance, displacement, and hopping time Figures 6(a) and (b) show the average numberN hopp of hoppers present in the system at various activities and packing fractions. Interestingly, at smallP e= 1.25 the number of hoppers is larger compared to higherP e, and increases only slightly as the number differenceN d in- creases. This ...

work page

[2] [2]

hunting packs

Mean-Square Displacement: Caging and Chasing To characterize the dynamics of iABPs in the ”hop- ping phase”, we evaluate the mean-square displacement (MSD) of theAandBparticles. The characterization of the dynamics of hoppers itself is only possible for rather short times due to frequent recombination and exchange withA-Bpairs. The presence of the majorit...

work page

[3] [3]

H. C. Berg, Bacterial behaviour, Nature254, 389 (1975)

work page 1975

[4] [4]

Be’er, B

A. Be’er, B. Ilkanaiv, R. Gross, D. B. Kearns, S. Heiden- reich, M. B¨ ar, and G. Ariel, A phase diagram for bacterial swarming, Commun. Phys.3, 66 (2020)

work page 2020

[5] [5]

I. S. Aranson, Bacterial active matter, Rep. Prog. Phys. 85, 076601 (2022)

2022

[6] [6]

Cavagna and I

A. Cavagna and I. Giardina, Bird flocks as condensed matter, Annu. Rev. Cond. Mat. Phys.5, 183 (2014)

work page 2014

[7] [7]

D. J. Pearce, A. M. Miller, G. Rowlands, and M. S. Turner, Role of projection in the control of bird flocks, Proc. Natl. Acad. Sci. USA111, 10422 (2014). 14

work page 2014

[8] [8]

Vicsek, A

T. Vicsek, A. Czir´ ok, E. Ben-Jacob, I. Cohen, and O. Shochet, Novel type of phase transition in a system of self-driven particles, Phys. Rev. Lett.75, 1226 (1995)

1995