RESEARCH ARTICLE

Undergraduate Tutorial for Simulating

Flocking with the Vicsek Model
A. Pasha Tabatabai 1,
*, MacQuarrie Thomson1,§, Reece Keller1,§

1
Seattle University, Department of Physics, Seattle, WA, USA

Downloaded from http://meridian.allenpress.com/the-biophysicist/article-pdf/4/1/30/3253757/i2578-6970-4-1-30.pdf by University of California Santa Barbara user on 17 November 2023
ABSTRACT There are many instances of collective behaviors in the natural
world. For example, eukaryotic cells coordinate their motion to heal wounds;
bacteria swarm during colony expansion; defects in alignment in growing bacterial
populations lead to biofilm growth; and birds move within dynamic flocks. Although
the details of how these groups behave vary across animals and species, they share
the same qualitative feature: they exhibit collective behaviors that are not simple
extensions of details associated with the motion of an individual. To learn more
about these biological systems, we propose studying these systems through the
lens of the foundational Vicsek model. Here, we present the process of building this
computational model from scratch in a tutorial format that focuses on building the
appropriate skills of an undergraduate student. In doing so, an undergraduate
student should be able to work alongside this article, the corresponding tutorial,
and the original manuscript of the Vicsek model to build their own model. We
conclude by summarizing some of the current work involving computational
modeling of flocking with Vicsek-type models.

KEY WORDS collective behavior; flocking; Vicsek model; computational

physics

I. INTRODUCTION
The biological world is full of fascinating collective behaviors across
an enormous range of length scales. At the nanometer-micrometer
length scale, cytoskeletal proteins organize and exhibit long-range
order while coordinating motion leading to cell division (1). In vitro
mixtures of cytoskeletal proteins also develop fascinating long-range
patterns (2–4).
Ordering and collective motion also occur within expanding
bacterial colonies on agar plates (5, 6). Group motion of bacteria
within bacterial suspensions is capable of affecting mechanical
properties, such as the viscosity of a fluid (7). Furthermore, bacterial
motility is coupled to biofilm growth (8, 9).
Eukaryotic cells coordinate with their neighbors in both their
motion and the forces they generate when closing wounds (10–12).
Within epithelial sheets, cells exhibit interesting and complex patterns
of motion while maintaining a mechanically robust cell layer (13–15).
Although those coordinate behaviors contribute positively to human
“§” equal contribution health, coordinated behavior in the form of cancer metastasis does
not (16, 17).
“*” corresponding author
Collective behaviors reach significantly larger length scales as well.
Received: 28 September 2022 Most notably, collective behaviors are observed in the flocking of birds
Accepted: 17 May 2023 and the schooling of fish. Order and complex dynamics of humans
Published: 8 August 2023
within dense crowds (18) and animals within migrating mammal herds
© 2023 Biophysical Society. (19) are also observed.

Tabatabai et al. The Biophysicist 2023; 4(1). DOI: 10.35459/tbp.2022.000227 30

 Vicsek model flocking simulation

The emergence of group motion from indi- and Reece Keller (RK), to build their own active
vidual motion, which is present from micrometer particle simulations, which included the Vicsek
length scales to hundreds of meters, is common model. A unique feature of this tutorial is that MT
to the biological world. Although the specific and RK played an integral role in identifying
details associated with collective motion likely particular skills that they needed to learn to build
differ at different length scales, a comprehensive their models independently (i.e., skills that may
understanding of these behaviors would be not be reinforced in a standard undergraduate
helpful in defining the effective interactions curriculum). In response, we highlight and provide

Downloaded from http://meridian.allenpress.com/the-biophysicist/article-pdf/4/1/30/3253757/i2578-6970-4-1-30.pdf by University of California Santa Barbara user on 17 November 2023
between individuals that lead to these interest- support in these areas by generating specific
ing behaviors. problems within the tutorial that address these
The Vicsek model (20), proposed in 1995, is a skills. It is our hope that the discrete nature of the
simplified computational model that captures questions within this tutorial with clear actionable
the emergence of collective behavior. In this items makes it accessible to all students regardless
model, objects move through space and tend of their background in computational physics or
to orient themselves with their neighbors. Thus, collective motion. Additionally, we provide de-
the details of the interaction between individ- tailed discussion in other sections for the topics of
uals change as an object accumulates more periodic boundary conditions (IV.B.3), trouble-
neighbors—a hallmark of an emergent system. shooting as a skill (IV.B.4), and defining steady
The Vicsek model is a valuable tool to study state (IV.B.5). A conservative estimate is that this
collective behaviors at any length scale because tutorial can be completed within an academic
it makes no assumptions about the origin of quarter or semester.
the interaction between individuals. We conclude the manuscript with a discus-
The utility of this model is also to help identify sion of some current work with flocking models
new characteristics of collective behavior. For and how an undergraduate reader would
example, short-range interactions between indi- implement these alterations to perform novel
viduals that lead to network-level coordination research in emergent behaviors.
must transfer information. As such, the Vicsek
model is a tangible method to study information II. PEDAGOGICAL BACKGROUND
flow (21, 22). Additionally, comparisons between
The Vicsek model is a valuable pedagogical
group-level properties in the model and exper-
tool for many reasons.
imental systems can elucidate the true interac-
tions that likely deviate from the simplified (a) An interdisciplinary approach to science
Vicsek interaction (23, 24). encourages students to build a well-rounded
The Vicsek model also predicts a phase skill set making them better prepared for
transition from a disordered state (individuals scientific careers. Although the basic Vicsek
acting independently) to an ordered state (collec- model obscures biologically relevant details
tive behavior). Most phase transitions that students in the interactions between objects, it is
learn about in a typical undergraduate curriculum possible to adapt this model to better reflect
are thermodynamic in origin, such as crossing of true biological systems.
solid/liquid or liquid/gas phase boundaries by (b) This model provides an approachable strat-
changing the temperature or pressure of water. egy to connect an extremely complicated
However, this example is a different type of observation in the real world (i.e., flocking)
transition—a kinetic phase transition. with basic mathematical and physical rules.
In this manuscript, we present a worked- This model is computationally simpler and
through tutorial for building the Vicsek model easier to integrate than other active simula-
from scratch. This example is the product of a tions where a true equation-of-motion with
research project in the lab of A. Pasha Tabatabai interparticle interactions is required (25, 26).
(APT), where APT led undergraduate physics (c) This model provides an example of a phase
majors, coauthors MacQuarrie Thomson (MT) transition. It would be beneficial for students

Tabatabai et al. The Biophysicist 2023; 4(1). DOI: 10.35459/tbp.2022.000227 31

Vicsek model flocking simulation

to see this model after taking a chemistry or code accompanied by blog-style commentary
thermodynamics course. In this way, com- (28). Although resources like these are helpful,
parisons can be made to a thermodynamic we believe that students will more deeply
phase diagram. Within a physics course, this understand the Vicsek model by building their
could be compared with the phase transition own code from scratch as opposed to refer-
in the ferromagnetic two-dimensional (2D) encing the prebuilt code of others.
Ising model. Introducing the Vicsek model (a
non-thermodynamic phase transition) pro- III. MATERIALS

Downloaded from http://meridian.allenpress.com/the-biophysicist/article-pdf/4/1/30/3253757/i2578-6970-4-1-30.pdf by University of California Santa Barbara user on 17 November 2023
vides an opportunity to explore the transition
Although it is possible to write these
through simulation. Implementing this within
simulations in any computer language, we
a noncomputational course could be accom-
chose to write in Python 3 because it is free,
plished by providing code to students and
open source, and taught in the introductory
encouraging them to change variables.
computer science course at our institution. We
(d) Computational skills are increasingly impor-
downloaded the Anaconda distribution and
tant for students interested in a career in
worked within a Jupyter notebook.
science. Little computational background is
We provide a tutorial in the Supplemental
needed to build this model, particularly with
Material that describes building the Vicsek model
the assistance of the tutorial. As such,
from scratch by using questions that develop and
exploring the Vicsek model can be imple-
test the model step-by-step. We provide solutions
mented as part of a thermodynamics course,
to each question so that students can be sure
statistical physics course, or a biological that they are moving in the right direction at
physics course to learn new physics. Alter- each step.
natively, this example could be implemented
as a module within a computational physics
course to emphasize computational skills in IV. RESULTS
the context of a physics problem. Depending A. The Vicsek model
on the course goal, students can be In the canonical Vicsek model, the time-
provided with sections of prebuilt code. varying position of each individual !
r i ðtÞ evolves
(e) The ability to troubleshoot issues in re- discretely as
search is a skill that translates across all
disciplines and careers. In this manuscript
!
r i ðtÞ þ !
r i ðt þ DtÞ ¼ ! v i ðtÞDt ð1Þ
and associated tutorial, we discuss trouble- where Dt is the time step between measure-
shooting strategies and break down the ments. The convention is to set Dt ¼ 1. The
problem into small components that build velocity of each individual ! v i ðtÞ is assumed to be
on each other. Providing pre-built code to of constant magnitude v and pointed in the
students, as mentioned above, may limit direction hi(t) (Fig 1). This orientation also evolves
the troubleshooting growth of students.
hi ðt þ DtÞ ¼ hhðtÞir þ Dh ð2Þ
Often times students learning challenges are
defined by the instructor, which may or may
not be aligned with actual student challenges.
As such, MT and RK identified skills that they
needed to learn to implement a computational
model. We hope that this style of tutorial is
helpful for future undergraduate students to
work independently through the manuscript.
Other pedagogical tools based on the Vicsek
Fig 1. Schematic of a single object’s position at time t and t + Dt.
model exist. For example, the authors are aware The object travels a distance vDt before the orientation is
of a video describing the original paper (20) in reevaluated. The difference between the orientation at t + Dt
detail (27), as well as fully worked example and the dashed line is the Dh.

Tabatabai et al. The Biophysicist 2023; 4(1). DOI: 10.35459/tbp.2022.000227 32

 Vicsek model flocking simulation

Fig 2. Schematic of interaction range between objects. The object

within the center of the left (red) circle has an interaction radius r0

Downloaded from http://meridian.allenpress.com/the-biophysicist/article-pdf/4/1/30/3253757/i2578-6970-4-1-30.pdf by University of California Santa Barbara user on 17 November 2023
which includes the leftmost object. The orientation of this object
will be affected by the orientation of the other object within the
interaction range. The object within the rightmost (blue) circle does
not have any neighbors within its interaction radius. As a
consequence, the object in this snapshot has an orientation that is
unaffected by the other objects.

where hhðtÞir is the average position of all

orientations within some radius r0 of an individual
at time t (Fig 2). This average is calculated as

 Fig 3. The position of a single particle moving with constant speed
arctan hsinðhðt ÞÞir =hcosðhðtÞÞir ð3Þ v ¼ 1 in a box of length L ¼ 10 superimposed over four instances
in time each separated by Dt ¼ 1. Lighter shades represent object
and includes the orientation of the individual itself.
position further in the past. g ¼ 3.
The average in Eq. 3 may seem unnecessarily
complex. As an example, consider two objects each
with angle h1 ¼ p/4 and h2 ¼ 7p/4. If h ¼ 0 is This order parameter measures the extent to
considered to be pointed to the right, the average which all objects have velocities in the same
orientation of these two objects should be to the directions. For example, if all objects are
right. However, calculating the geometric mean of moving in the same direction, va ! 1, whereas
these angles gives (h1 + h2)/2 ¼ p. By taking the
average from Eq. 3, we correctly calculate the
circular average h ¼ 0.
The angular noise Dh varies for each object at
each time step and is chosen from a uniformly
distributed random angle within the interval
[–g/2, g/2]. Therefore, g sets the possible values
of the random noise Dh. The effect of this
angular noise is to add some uncertainty in the
trajectory of a given particle, and its effects can
be seen by looking at an isolated particle where
hhðtÞir becomes hi(t) (Fig 3).
In a system with many objects, the term hhðtÞir
in Equation 2 acts to align neighboring objects,
whereas Dh acts to randomize orientations. When
g is small, the ordering effect dominates, and
long-range order is present (Fig 4).
The overall extent of collective behavior is
calculated by measuring the order parameter
  Fig 4. Snapshot of N ¼ 200 objects in a box of length L ¼ 10
1  X
N 
 interacting with radius r0 ¼ 1 after t ¼ 250. Noise is set to g ¼ 0.1
va ¼  ~v i : ð4Þ and Dt ¼ 1. Color coding of arrows represents object orientation.
Nv i 
Objects were initialized with random positions and orientations.

Tabatabai et al. The Biophysicist 2023; 4(1). DOI: 10.35459/tbp.2022.000227 33

Vicsek model flocking simulation

(29) and the transition occurs at a noise level that

is also a function of system density gc(ρ) (30).
One reason that the Vicsek model is interesting
to study is that the long-range order associated
with large values of the order parameter va are
not possible in 2D for equilibrium systems. The
Mermin–Wagner theorem states that a system
with short-range interactions at thermal equilib-

Downloaded from http://meridian.allenpress.com/the-biophysicist/article-pdf/4/1/30/3253757/i2578-6970-4-1-30.pdf by University of California Santa Barbara user on 17 November 2023
rium cannot lead to a long-range order (31, 32). In
the Vicsek model, the interactions are short range
(r0 ,, L), yet the steady state is capable of
achieving long-range order (va ~ 1) because self
propulsion (v 6¼ 0) is a nonequilibrium feature. An
exercise in the tutorial demonstrates that mea-
surements of va for systems with and without
swimming are very different despite all other
parameters being the same.

Fig 5. Steady state order parameter va as a function of g for L ¼ 5 B. Skills and concepts
(black), L ¼ 15 (red), and L ¼ 25 (blue) at a density of ρ ¼ 8
1. Python familiarity
with r0 ¼ 1 and Dt ¼ 1. Computational abilities vary widely across an
undergraduate cohort partially because compu-
many noninteracting, randomly oriented ob- tational courses are introduced at different points
jects lead to va ! 0. The order parameter va can in the undergraduate curriculum, and different
be calculated at every instance in time and departments have different major requirements
therefore evolves until the system reaches for computational courses. As such, many of the
“steady state” (section IV.B.5); the steady state challenges in developing this model are rooted in
values of va are quoted in Figure 5. the learning curve associated with Python.
Thus we see that by incorporating these rules, The tutorial assumes students know the
we capture an essential result of the Vicsek model: following Python skills before starting the tutorial:
that the noise g controls a dynamic first-order
phase transition from ordered flocking (high va) to (a) How to download and install the Anaconda
disorder (va ! 0), even when the number of distribution (www.anaconda.com) of Python
objects per unit area (i.e., the density) is kept 3 and work within a Python environment. We
recommend Spyder or Jupyter notebooks.
constant (Fig 5).
(b) How to import libraries such as matplotlib
When fully describing a phase transition, it is
and NumPy.
common to determine exactly how an order
(c) How to use PyPlot from matplotlib to
parameter changes at this transition point. In the
display and save plots.
case of the Vicsek model, a noise level gc (d) How to write a function.
separates the regime of collective motion from
the disordered regime. In fact, this noise value is Students who are unfamiliar with these items
also a function of system size gc(L), as seen in can find myriad online resources to help them
Figure 5. In the original publication, Vicsek et al. prepare for this tutorial.
(20) treated this transition as continuous (i.e., a 2. Python usage
second-order phase transition like the 2D Ising The tutorial is designed to approach partic-
model). However, later works with larger systems ular computational problems step-by-step. The
argue that the canonical Vicsek model is a tutorial covers the following computational
discontinuous (i.e., a first-order) phase transition topics in the context of problem solving:

Tabatabai et al. The Biophysicist 2023; 4(1). DOI: 10.35459/tbp.2022.000227 34

 Vicsek model flocking simulation

(a) Importing and using built-in Python functions. the research process. Typical methods used to
(b) Random number generators. confirm that the model is implemented correctly
(c) Using loops and if statements. (e.g., visual checks, building toy models, and taking
(d) Generating and displaying images and plots. limits of results) are demonstrated in the tutorial.
(e) Calling multiple functions and using return 5. Steady state
statements within functions. In a typical simulation, particles are distributed
3. Periodic boundary conditions in a region of space with random positions and

Downloaded from http://meridian.allenpress.com/the-biophysicist/article-pdf/4/1/30/3253757/i2578-6970-4-1-30.pdf by University of California Santa Barbara user on 17 November 2023
The use of periodic boundary conditions in orientations. The system is allowed to evolve,
computational simulations is common. When and in this evolution process the positions and
simulating something, you want to avoid the orientations may become more correlated across
effects of a boundary. If a boundary exists, you objects depending on the experimental condi-
have to consider new rules for interaction with tions. The goal is to determine the effect of
that boundary. In fact, what you really want is parameters on the ultimate behavior of the
an infinitely large simulation so that the effects system that is independent of the initial condi-
of these boundaries are minimal, but that tions and configuration.
would require infinite computing resources As such, we separate the behavior of our
(and take infinitely long to run). system into two regimes: the “transient” and
We use periodic boundary conditions such that the “steady state.” In the steady state behavior,
objects leaving the bounding area reappear on fluctuations in a measured quantity such as the
the opposite side of the simulation area. As such, order parameter occur, but the mean value is
objects simply pass through the boundary and constant. In the preceding transient regime, a
do not interact. Implementing periodic boundary measured order parameter fluctuations and the
conditions is an explicit step in the tutorial. mean are changing in a way that depended on
the initial configuration. Here, we only make
4. Troubleshooting measurements in steady state. A strategy to
A primary challenge in building these mod- determine whether the simulation has reached
els, and research in general, is finding trouble- the steady state regime is to compare averages
shooting methods for suspected errors. Within of the order parameter over different time
the typical structure of an undergraduate physics windows. If the average is unchanged, then
course, students tend to determine whether their each time window is in steady state. Note that
solutions to problems are correct on the basis of reaching steady state is not necessarily the
a comparison with a known solution (either given same as reaching thermodynamic equilibrium.
or from a peer). However, in research scenarios, a
known solution may not exist. As such, students
must find ways to convince themselves and their
V. A PERSPECTIVE ON CURRENT
advisor that their results are correct, determine VICSEK MODEL USE
how to correct their results, or both. Although the simplicity of the Vicsek model is
This process deviates significantly from the sufficient to capture this order–disorder transi-
learning process within a traditional physics tion, the model is still being adapted in current
course. In these courses, students tend to research. We highlight a few recent works.
submit assignments and receive a grade—end The primary assumption in the standard
of transaction. However research is iterative, Vicsek model is that objects interact with all
and progress relies on reflecting on the work other objects within a given radius. However,
done and implementing improvements. the visual information collected by a bird, for
As a consequence of this difference between example, is not isotropic (i.e., the bird cannot
research and coursework, troubleshooting is not see behind itself). To this end, nonisotropic
an explicit skill learned in courses. Undergraduate interactions with limited fields of view, as
research is important for this reason. We empha- shown in Figure 6, have recently been inte-
size here that troubleshooting is an active part of grated into the Vicsek model (33–35).

Tabatabai et al. The Biophysicist 2023; 4(1). DOI: 10.35459/tbp.2022.000227 35

Vicsek model flocking simulation

stems from local interactions between active

objects. We provide discussion about concepts
that are necessary to build this model that may
be new intellectual territory for undergraduate
students, and we accompany a tutorial to help
students build the model step-by-step. Finally,
we have discussed a few ways in which current
researchers are adapting the Vicsek model for

Downloaded from http://meridian.allenpress.com/the-biophysicist/article-pdf/4/1/30/3253757/i2578-6970-4-1-30.pdf by University of California Santa Barbara user on 17 November 2023
their own research.
It is our hope that undergraduate students
feel comfortable approaching this model and
Fig 6. Schematic of nonisotropic interaction between objects. The that our provided resources, in addition to the
primary object is shown to have an interaction that is limited original manuscript (20), are sufficient to assist in
to a range of p/2. Only the red arrow is within the interaction an undergraduate student building their own
range of this object. model from scratch.

It is important to consider that true interac-

SUPPLEMENTAL MATERIAL
tions between animals may not be governed by
metric interactions (i.e., how far the animals are A supplemental tutorial that builds the Vicsek model is
available at https://doi.org/10.35459/tbp.2022.000227.s1.
apart). For example, starlings in a flock have
been shown to interact topologically (individ-
uals interact with some number of nearest AUTHOR CONTRIBUTIONS
neighbors, regardless of distance) (23), and fish APT designed and conceived the project and wrote the
manuscript. MT, RK, and APT performed simulations and analysis.
in a school use line-of-sight interactions (24). MT and RK identified and defined learned skills and concepts.
Another principle assumption of the standard
Vicsek model is that objects move with a
constant speed at all times. This model has been
ACKNOWLEDGMENTS
APT was supported by an M. J. Murdock Charitable Trust
adapted to better mimic the variability of fish Award (FSU-201913717) and a National Science Foundation
speed, which has been determined to depend on Award (2137509). The authors declare no conflicts of interests.
the local alignment of neighbors (36).
Additionally, most real systems exist in three REFERENCES
dimensions, whereas this manuscript and the 1. Brugués, J., and D. Needleman. 2014. Physical basis of spindle self-organization. Proc
tutorial are explicitly done for the 2D Vicsek Natl Acad Sci U S A, 111:18496–18500. https://doi.org/10.1073/pnas.1409404111.
2. Sanchez, T., D. T. N. Chen, S. J. DeCamp, M. Heymann, and Z. Dogic. 2012.
model. This can be extended to three dimensions. Spontaneous motion in hierarchically assembled active matter. Nature 491:431–
Furthermore, self-propelled particle models 434. https://www.nature.com/articles/nature11591.
3. Linsmeier, I., S. Banerjee, P. W. Oakes, W. Jung, T. Kim, and M. P. Murrell. 2016.
such as the Vicsek model are used as a reference Disordered actomyosin networks are sufficient to produce cooperative and
to other active models like the vertex model when telescopic contractility. Nat Commun 7:12615. https://www.nature.com/articles/
ncomms12615.
investigating the order–disorder transition (37). 4. Seara, D. S., V. Yadav, I. Linsmeier, A. P. Tabatabai, P. W. Oakes, S. M. A. Tabei, S.
Banerjee, and M. P. Murrell. 2018. Entropy production rate is maximized in non-
contractile actomyosin. Nat Commun 9:4948. https://www.nature.com/articles/
VI. DISCUSSION AND s41467-018-07413-5.
CONCLUSION 5. Czirók, A., E. Ben-Jacob, I. Cohen, and T. Vicsek. 1996. Formation of complex
bacterial colonies via self-generated vortices. Phys Rev E 54:1791–1801. https://
We have motivated the reasons to study the doi.org/10.1103/PhysRevE.54.1791.
6. Zhang, H. P., A. Be’er, E.-L. Florin, and H. L. Swinney. 2010. Collective motion and
Vicsek model in the undergraduate curriculum density fluctuations in bacterial colonies. Proc Natl Acad Sci U S A 107:13626–
both from the perspective of a student 13630. https://doi.org/10.1073/pnas.1001651107.
7. Sokolov, A., and I. S. Aranson. 2012. Physical properties of collective motion in
researcher and as an educator. We have defined suspensions of bacteria. Phys Rev Lett 109:248109. https://doi.org/10.1103/
the details of the Vicsek model and implemented PhysRevLett.109.248109.
8. Copenhagen, K., R. Alert, N. S. Wingreen, and J. W. Shaevitz. 2021. Topological
this model to show a canonical result—a kinetic defects promote layer formation in Myxococcus xanthus colonies. Nat Phys 17:211–
phase transition between order and disorder that 215. https://doi.org/10.1038/s41567-020-01056-4.

Tabatabai et al. The Biophysicist 2023; 4(1). DOI: 10.35459/tbp.2022.000227 36

 Vicsek model flocking simulation

9. Qin, B., C. Fei, A. A. Bridges, A. A. Mashruwala, H. A. Stone, N. S. Wingreen, and Interaction ruling animal collective behavior depends on topological rather than
B. L. Bassler. 2020. Cell position fates and collective fountain flow in bacterial metric distance: evidence from a field study. Proc Natl Acad Sci U S A 105:1232–
biofilms revealed by light-sheet microscopy. Science 369:71–77. https://doi.org/ 1237. https://doi.org/10.1073/pnas.0711437105.
10.1126/science.abb8501. 24. Rosenthal, S. B., C. R. Twomey, A. T. Hartnett, H. S. Wu, and I. D. Couzin. 2015.
10. Trepat, X., M. R. Wasserman, T. E. Angelini, E. Millet, D. A. Weitz, J. P. Butler, and Revealing the hidden networks of interaction in mobile animal groups allows
J. J. Fredberg. 2009. Physical forces during collective cell migration. Nat Phys prediction of complex behavioral contagion. Proc Natl Acad Sci U S A 112:4690–
5:426–430. https://doi.org/10.1038/nphys1269. 4695. https://doi.org/10.1073/pnas.1420068112.
11. Basan, M., J. Elgeti, E. Hannezo, W.-J. Rappel, and H. Levine. 2013. Alignment of 25. Mishra, S., A. Baskaran, and M. C. Marchetti. 2010. Fluctuations and pattern
cellular motility forces with tissue flow as a mechanism for efficient wound healing. formation in self-propelled particles. Phys Rev E 81:061916. https://doi.org/10.
Proc Natl Acad Sci U S A 110:2452–2459. https://doi.org/10.1073/pnas.1219937110. 1103/PhysRevE.81.061916.
12. Ajeti, V., A. P. Tabatabai, A. J. Fleszar, M. F. Staddon, D. S. Seara, C. Suarez, M. S.

Downloaded from http://meridian.allenpress.com/the-biophysicist/article-pdf/4/1/30/3253757/i2578-6970-4-1-30.pdf by University of California Santa Barbara user on 17 November 2023
26. Marchetti, M. C., Y. Fily, S. Henkes, A. Patch, and D. Yllanes. 2016. Minimal model
Yousafzai, D. Bi, D. R. Kovar, S. Banerjee, and M. P. Murrell. Wound healing
of active colloids highlights the role of mechanical interactions in controlling the
coordinates actin architectures to regulate mechanical work. Nat Phys 15:696–
emergent behavior of active matter. Curr Opin Colloid Interface Sci 21:34–43.
705. https://doi.org/10.1038/s41567-019-0485-9.
https://doi.org/10.1016/j.cocis.2016.01.003.
13. Giavazzi, F., C. Malinverno, S. Corallino, F. Ginelli, G. Scita, and R. Cerbino. 2017.
Giant fluctuations and structural effects in a flocking epithelium. J Phys D Appl 27. Lian, M., T. E. Bate, and K.-T. Wu. 2022. Collective motion of active matter. PICUP
Phys 50:384003. https://doi.org/10.1088/1361-6463/aa7f8e. Faculty Commons. Accessed 1 March 2023. https://www.compadre.org/PICUP/
14. Angelini, T. E., E. Hannezo, X. Trepat, M. Marquez, J. J. Fredberg, and D. A. Weitz. exercises/exercise.cfm?I¼488&A¼activematter.
2011. Glass-like dynamics of collective cell migration. Proc Natl Acad Sci U S A 28. Turci, F. 2020. Minimal Vicsek model in Python. Accessed 1 March 2023. https://
108:4714–4719. https://doi.org/10.1073/pnas.1010059108. francescoturci.net/2020/06/19/minimal-vicsek-model-in-python/.
15. Khalilgharibi, N., J. Fouchard, N. Asadipour, R. Barrientos, M. Duda, A. Bonfanti, A. 29. Grégoire, G., and H. Chaté. 2004. Onset of collective and cohesive motion. Phys
Yonis, A. Harris, P. Mosaffa, Y. Fujita, A. Kabla, Y. Mao, B. Baum, J. J. Muñoz, M. Rev Lett 92:025702. https://doi.org/10.1103/PhysRevLett.92.025702.
Miodownik, and G. Charras. 2019. Stress relaxation in epithelial monolayers is 30. Chaté, H., F. Ginelli, G. Grégoire, F. Peruani, and F. Raynaud. 2008. Modeling
controlled by the actomyosin cortex. Nat Phys 15:839–847. https://doi.org/10. collective motion: variations on the Vicsek model. Eur Phys J B 64:451–456.
1038/s41567-019-0516-6. https://doi.org/10.1140/epjb/e2008-00275-9.
16. Yousafzai, M. S., V. Yadav, S. Amiri, M. F. Staddon, Y. Errami, G. Jaspard, S. 31. Mermin, N. D., and H. Wagner. 1966. Absence of ferromagnetism or
Banerjee, and M. Murrell. 2022. Cell-matrix elastocapillary interactions drive antiferromagnetism in one- or two-dimensional isotropic Heisenberg models. Phys
pressure-based wetting of cell aggregates. Phys Rev X 12:031027. https://doi.org/ Rev Lett 17:1133–1136. https://doi.org/10.1103/PhysRevLett.17.1133.
10.1103/PhysRevX.12.031027. 32. Toner, J., Y. Tu, and S. Ramaswamy. 2005. Hydrodynamics and phases of flocks.
17. Deisboeck, T. S., and I. D. Couzin. 2009. Collective behavior in cancer cell Ann Phys 318:170–244. https://doi.org/10.1016/j.aop.2005.04.011.
populations. BioEssays 31:190–197. https://doi.org/10.1002/bies.200800084. 33. Durve, M., and A. Sayeed. 2016. First-order phase transition in a model of self-
18. Silverberg, J. L., M. Bierbaum, J. P. Sethna, and I. Cohen. 2013. Collective motion propelled particles with variable angular range of interaction. Phys Rev E
of humans in mosh and circle pits at heavy metal concerts. Phys Rev Lett
93:052115. https://doi.org/10.1103/PhysRevE.93.052115.
110:228701. https://doi.org/10.1103/PhysRevLett.110.228701.
34. Durve, M., A. Saha, and A. Sayeed. 2018. Active particle condensation by non-
19. Ferdinandy, B., K. Ozogány, and T. Vicsek. 2017. Collective motion of groups of
reciprocal and time-delayed interactions. Eur Phys J E 41:49. https://doi.org/10.
self-propelled particles following interacting leaders. Physica A Stat Mech Appl
1140/epje/i2018-11653-4.
479:467–477. https://doi.org/10.1016/j.physa.2017.03.025.
20. Vicsek, T., A. Czirók, E. Ben-Jacob, I. Cohen, and O. Shochet. 1995. Novel type of 35. Roy, S., M. J. Shirazi, B. Jantzen, and N. Abaid. 2019. Effect of visual and auditory
phase transition in a system of self-driven particles. Phys Rev Lett 75:1226–1229. sensing cues on collective behavior in Vicsek models. Phys Rev E 100:062415.
https://doi.org/10.1103/PhysRevLett.75.1226. https://doi.org/10.1103/PhysRevE.100.062415.
21. Toner, J., and Y. Tu. 1998. Flocks, herds, and schools: a quantitative theory of 36. Mishra, S., K. Tunstrøm, I. D. Couzin, and C. Huepe. 2012. Collective dynamics of
flocking. Phys Rev E 58:4828–4858. https://doi.org/10.1103/PhysRevE.58.4828. self-propelled particles with variable speed. Phys Rev E 86:011901. https://doi.org/
22. Brown, J., T. Bossomaier, and L. Barnett. 2020. Information flow in finite flocks. Sci 10.1103/PhysRevE.86.011901.
Rep 10:3837. https://doi.org/10.1038/s41598-020-59080-6. 37. Bi, D., J. H. Lopez, J. M. Schwarz, and M. L. Manning. 2015. A density-independent
23. Ballerini, M., N. Cabibbo, R. Candelier, A. Cavagna, E. Cisbani, I. Giardina, V. rigidity transition in biological tissues. Nat Phys 11:1074–1079. https://doi.org/10.
Lecomte, A. Orlandi, G. Parisi, A. Procaccini, M. Viale, and V. Zdravkovic. 2008. 1038/nphys3471.

Tabatabai et al. The Biophysicist 2023; 4(1). DOI: 10.35459/tbp.2022.000227 37