Pith. sign in

REVIEW 4 major objections 5 minor 54 references

Anomalous Diffusion and Emergent Universality in Coupled Memory-Driven Systems

T0 review · 4 major / 5 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read Two trail-following walkers form new diffusion classes.

desk verdict New model, plausible phenomenology, but the pseudonormal distributional scaling is non-normalizable as written and the universality claims need tighter analysis. read the letter →

arxiv 2411.09092 v2 pith:A4CTHWJ7 submitted 2024-11-13 q-bio.PE cond-mat.stat-mechphysics.bio-phphysics.comp-ph

Nick Dashti , M. N. Najafi , Debra J. Searles This is my paper · ORCID
classification q-bio.PEcond-mat.stat-mechphysics.bio-phphysics.comp-ph
keywords anomalousdiffusioncoupledrandomwalkstrueself-avoidingwalkuniversalityclassespheromone-guidednavigationnon-Gaussianpositiondistributionsencounterstatisticsmeansquareddisplacement
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 introduces a minimal model of two random walkers on a lattice, each leaving a trail of debris at every site it visits. Each walker is biased against returning to its own trail, with strength $\beta$, and toward the other walker's trail, with strength $\beta'$. The paper claims that the balance of these two biases organizes the long-time behavior into distinct universality classes: superdiffusion with exponent $\alpha = 4/3$ when $\beta' \le \beta$, a pseudonormal subdiffusive regime with $\alpha$ approaching 1 from below under strong attraction, and $\alpha = 1/2$ when self-avoidance is absent. In two dimensions it reports a new subdiffusive class with $\alpha = 9/10$. If these results hold, simple pairwise trail-following rules would generate collective transport regimes that neither self-avoiding nor self-attracting walks produce on their own.

What carries the argument

The carrying mechanism is the coupled transition probability $p^{(X)}_{i \to j} = e^{-\beta h^{(X)}_j + \beta' h^{(X')}_j}/Z$, defined on nearest-neighbor sites, with $h^{(X)}_j$ the cumulative visit count (debris field) of walker $X$. This single formula encodes the competition: self-repulsion lowers the weight of sites the walker has visited, while cross-attraction raises the weight of sites marked by the other walker. The paper uses this kernel in large-scale lattice simulations, reports the nearest fractional exponents that fit the data, and identifies the phase boundaries $B(\beta)$ and $B_{2D}$ numerically as the points where fitted exponents change.

What would settle it

Run the one-dimensional model to $t = 10^9$ for $\beta = 1, \beta' = 5$ and measure $R^2_t/t$ directly: the pseudonormal claim predicts that this ratio keeps slowly decaying (roughly as $t^{-0.02}$) while the bulk position width grows as $t^{2/3}$; if instead the ratio plateaus at a nonzero constant with Gaussian bulk scaling, the class is not asymptotic. For the two-dimensional claim, compare $R^2_t/t^{9/10}$ with $R^2_t/t$ beyond $t = 10^7$; if the ratio grows logarithmically, the $\alpha = 9/10$ class collapses into the true self-avoiding walk regime.

Watch

Extended reading notes

Core claim

The central claim is that two coupled walkers with the transition probability $p^{(X)}_{i \to j} \propto \exp(-\beta h^{(X)}_j + \beta' h^{(X')}_j)$, where $h^{(X)}_j$ is the number of visits walker $X$ has made to site $j$, show new universality classes. For $\beta' \le \beta$ the model recovers the true self-avoiding walk regime: $\langle R^2_t \rangle \sim t^{4/3}$ in one dimension and $\sim t(\ln t)^{1/2}$ in two dimensions. For $\beta' > B$ with $\beta > 0$ in one dimension, the mean-squared displacement exponent approaches $\alpha = 1$ from below while the position distribution is fat-tailed and non-Gaussian, a regime the paper calls pseudonormal. For $\beta = 0$ and large $\beta'$, the exponent becomes $\alpha = 1/2$. In two dimensions, $\beta' > B_{2D}$ gives a subdiffusive class with $\alpha = 9/10$ and no logarithmic correction. The encounter and meeting-duration statistics are also non-Gaussian, and the paper states that these coupled classes have not been reported before and that the exponents are independent of the initial separation $D_0$.

Load-bearing premise

The load-bearing premise is that exponents fitted from simulations run to $t = 10^7$, with reported uncertainty near $0.05$, are the true large-time limits; the new classes are defined by differences of that same size, such as $\alpha$ slightly below 1 versus $\alpha = 1$, or $\alpha = 9/10$ versus the logarithmic true self-avoiding walk law.

Editorial extensions

If this is right

  • In one dimension the model has three asymptotic regimes: true self-avoiding walk superdiffusion ($\alpha = 4/3$), pseudonormal subdiffusion ($\alpha \to 1^-$ with fat-tailed positions), and $\alpha = 1/2$ when self-repulsion is switched off.
  • In two dimensions, strong cross-attraction yields a subdiffusive class with $\alpha = 9/10$ and no logarithmic correction, distinct from the true self-avoiding walk class with $\alpha = 1, \hat\alpha = 1/2$.
  • Encounter and meeting-duration statistics in the superdiffusive regime follow compressed exponentials ($\exp(-a z^{4/3})$) rather than Gaussians, implying more frequent long meetings than independent random walkers would produce.
  • The pseudonormal regime combines a mean-squared displacement exponent near 1 with a bulk position width growing as $t^{2/3}$, so rare, long excursions dominate the second moment.
  • The reported exponents are independent of the initial separation $D_0$ between the agents.

Reading between the lines

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

  • The boundaries $B(\beta)$ and $B_{2D}$ are fitted numerically, not derived; because the claimed phases differ by roughly the reported uncertainty, an analytical treatment such as a renormalization-group or exact-enumeration study of the transition kernel could either sharpen or dissolve the distinctions between classes.
  • In the pseudonormal regime the fat tails imply strong sample-to-sample fluctuations in squared displacement; measuring the distribution of single-trajectory $R^2_t$, which the paper does not report, would directly test whether the ensemble exponent is stable.
  • The reciprocal-attraction setup suggests an optimization reading: encounter frequency should peak somewhere near $\beta' \approx \beta$, where the compressed-exponential encounter distribution gives way to exponential; mapping encounter rate across the phase diagram could connect the scaling classes to search efficiency.
  • The two-dimensional $\alpha = 9/10$ class is close enough to the true self-avoiding walk logarithmic law that first-passage or record statistics, rather than mean-squared displacement alone, would be a sharper distinguishing observable.
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.

Referee Report

4 major / 5 minor

Summary. The paper introduces a two-agent lattice random-walk model in which each agent avoids its own trail (self-avoidance strength β) and is attracted to the other agent's trail (attraction strength β'). Using Monte Carlo simulations up to t = 10^7 with 5–10 million samples per parameter set, the authors report phase diagrams in one and two dimensions. In 1D they identify three regimes: TSAW-like superdiffusion with α = 4/3 for β' ≤ β; a 'pseudonormal' subdiffusive regime with α approaching 1 from below for β' > B; and α = 1/2 for β = 0 with β' large. In 2D they report TSAW behavior for β' ≤ β and a novel subdiffusive regime with α = 9/10 for β' > B2D. The paper also presents scaling forms for position, encounter-number, and encounter-duration distributions, and claims these constitute new universality classes of coupled random walks.

Significance. If the reported new regimes are confirmed, the paper would extend the known taxonomy of self-interacting and mutually interacting random walks in a natural and interesting direction. The model is minimal and clearly motivated, the supplementary pseudo-code aids reproducibility, and the simulation campaign is substantial, with long times and large ensembles. The recovery of known TSAW exponents in the β' ≤ β regime provides a useful internal check. However, the central claims depend on exponent differences that are close to the stated numerical uncertainty and on distributional scaling forms that, as written, are not normalizable. These issues are load-bearing for the claim of new universality classes, and they require additional analysis before the conclusions can be accepted.

major comments (4)
  1. [Section IV, Table I (Eq. 10)] The pseudonormal entry P(x,t) ∼ t^{-1} f((x - D0/2)/t^{2/3}) with ζx = 1 and νx = 2/3 is not normalizable as a single-component scaling form. Substituting u = (x - D0/2)/t^{2/3} gives ∫P(x,t) dx = t^{νx - ζx} ∫f(u) du = t^{-1/3} ∫f(u) du, which tends to zero for any fixed integrable f. The paper's statement that ζ and ν may differ for fat-tailed distributions does not rescue a one-variable ansatz with a single fixed f; it requires a superposition of at least two scaling components, such as a bulk region and a separately scaling tail. The same defect appears in the same row for P(T,t) ∼ t^{-7/5} f(T/t), whose implied total mass is t^{-2/5}. Since the abstract and conclusion define the new universality classes by their 'unique scaling laws and distributional properties', the missing tail is not cosmetic: it must carry normalization and the second moment responsible for α ≈ 1.
  2. [Section III, Figs. 2(a), 6(c)] The two principal new exponents are claimed with a stated uncertainty of about 0.05, yet the 1D pseudonormal claim α → 1− corresponds to α ≈ 0.98 (the fit R2_t/t ∼ t^{-0.02} in Fig. 2a), a difference from normal diffusion that is smaller than the stated uncertainty. The 2D claim α = 9/10 differs from the TSAW value α = 1 with log correction by 0.10, again close to the resolution of the fits. The paper reports 'nearest fractional values that accurately represent the numerical results' but does not report fit ranges, confidence intervals, or systematic time-dependent corrections over the accessed window up to t = 10^7. Without a finite-time scaling analysis, a test against logarithmic corrections, or an independent analytical estimate, the existence of distinct asymptotic universality classes at these parameter values is not established to the precision claimed.
  3. [Section IV, mean-field argument] The reduction to TSAW for β' ≤ β assumes h(A)_i ≈ h(B)_i, and the authors state that for β' = β 'the fluctuations of h_i make this interpretation inapplicable'. Nevertheless the phase diagram and Table I include β' = β in the TSAW regime. This is load-bearing for the phase boundary, and in 2D the data for β = β' = 1 show a visible deviation from the claimed t(ln t)^{1/2} behavior (Fig. 6a), which is only argued away by the MSI plateau in Fig. 6b. The authors should provide a direct test for β' = β, such as a distribution collapse or a systematic sweep across the boundary, or temper the classification at this point.
  4. [Section III, Fig. 3(f) and Table I] The α = 1/2 claim for β = 0, β' > B relies on fits over the range shown in Fig. 2c, where only the largest β' values approach a plateau and the asymptotic regime for smaller β' is not quantified. In addition, the position distribution is collapsed only in the left region x ≲ -35; the right and central parts are not given a scaling form. Because the paper presents distributional properties as part of the universality class, this incomplete characterization weakens the claim for this regime, not just for the exponent value.
minor comments (5)
  1. [Figure 5(c)] The functions are written as exp(a1 z^{3/2}) and exp(a2 z^5) with positive coefficients, but the plotted curves decay; presumably the intended forms are exp(-a1 z^{3/2}) and exp(-a2 z^5).
  2. [Eq. (10) and Table I] The general scaling form writes P(u,t) ∼ t^{-ζu} f_u(u/t^{νu}), but Table I entries are written with explicit prefactors such as t^{-1} f(x/t^{2/3}); please clarify whether the prefactor exponent is ζu or νu and define the relation between the notation in Eq. (10) and the table entries.
  3. [Figure 3(e) caption] The caption says 'actually for all β' ≥ B' but panel (d) includes β' = 1.5 with B ≈ 1.5; please state whether β' = 1.5 lies in the pseudonormal or the transient region.
  4. [Section III, Fig. 2(d)] The text states that error bars are smaller than symbol sizes, but no error bars or fitting procedure are shown; please provide the fitting method, the fitting range, and residuals or confidence intervals for the exponents.
  5. [Supplementary Material] The pseudo-code is helpful; making the simulation code publicly available would further strengthen the reproducibility of the numerical universality-class claims.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the reported exponents and distributions are measured from simulations of an independently defined model, not inserted as inputs.

full rationale

The model is defined in Eq. (5) through transition probabilities that depend only on β, β′, and the pheromone histories h_i; no target exponent, scaling function, or universality class is used to define the dynamics. All claims about α, ζ, and ν are obtained by simulating this model and fitting scaling forms to the resulting MSD and distribution data, with the paper explicitly saying in Section III that 'we report the nearest fractional values that accurately represent the numerical results.' The mean-field mapping to the TSAW for β′≤β is a post-hoc rationalization of the observed exponent 4/3, not an input that produces that exponent. The only author-overlap citation is Ref. [48], used in the Introduction to motivate interest in multi-agent scenarios ('particularly when multiple interacting agents are involved [47, 48]'); it is not load-bearing for the central scaling claims and is not invoked as a uniqueness theorem or as justification of the scaling ansatz. There is a substantive non-circular correctness/consistency concern in Table I: the pseudonormal row lists P(x,t) ~ t^{-1} f((x-D0/2)/t^{2/3}) with ζx=1 and νx=2/3, which for a one-argument scaling form cannot normalize as t→∞; this suggests the distributional description needs a separate fat-tail scaling component, but this is an internal-consistency issue, not an instance of the paper deriving its conclusions from its own inputs.

Assumptions & free parameters 6 free parameters · 5 assumptions · 0 invented entities

The central claims (new scaling exponents and distribution shapes) are entirely determined by fitting the paper's own simulation data. Six classes of fitted numbers appear: regime diffusion exponents, tail exponents, collapse exponents, log corrections, prefactors, and the transition values B(beta), B2D. Model parameters beta and beta-prime are varied inputs, not fitted outputs. No new physical entities are postulated: the 'pheromone' fields h(X)_i are the memory counters defined in Eq. (4). The main unstated assumptions are convergence of the ensemble averages and the stability of the identified exponents beyond t = 10^7.

free parameters (6)
  • 1D diffusion exponent alpha (beta-prime > B, beta > 0) = about 1 (R2/t ~ t^(-0.02))
    Fitted to MSD data in Fig. 2a; within the stated +/-0.05 uncertainty of alpha = 1, so the 'pseudonormal' class is not resolved beyond normal diffusion.
  • 2D diffusion exponent alpha (beta-prime > B2D) = 9/10
    Fitted to MSD data in Fig. 6c; uncertainty 0.05 leaves the true exponent between 0.85 and 0.95.
  • 1D diffusion exponent alpha (beta = 0, beta-prime > 0) = 1/2
    Fitted to MSD data in Fig. 2c; equals the known 3D TSAW subdiffusive exponent.
  • Distribution tail exponents = 3.3, 3, 4/3, 3/2, 5
    Fitted to tails of P(x,t), P(m,t), P(T,t); quoted as clean numbers with prefactors a, a1, a2 depending on beta and beta-prime.
  • Collapse exponents (zeta_u, nu_u) and log corrections = e.g., zeta_x = 1, nu_x = 2/3; zeta_T = 7/5, nu_T = 1; 2D log terms 1/2 and 1/4
    Chosen so that curves at t = 5x10^5,...,10^7 collapse; calibrated on the same ensemble as the MSD exponents.
  • Transition values B and B2D = B about 1.5 (1D, beta = 1); B2D about 2 (2D, beta = 1)
    Read off where the apparent scaling changes; the paper states the exact form of B(beta) is unknown and the intermediate region is skipped.
assumptions (5)
  • standard math The scaling ansatz R2_t ~ t^alpha (ln t)^(alpha-hat) and the distribution form P(u,t) ~ t^(-zeta_u) f(u/t^(nu_u)) are assumed.
    Invoked in Section II (Eq. 8) and Section IV (Eq. 10); standard in the anomalous diffusion literature and not specific to this model.
  • domain assumption Ensemble averages over 5-10 million samples yield converged estimates of the quoted exponents.
    Section III gives sample counts but no convergence diagnostics; the authors admit rare-event sensitivity in the beta-prime > B regimes, which is exactly where the new exponents are claimed.
  • domain assumption Asymptotic results are independent of the initial separation D0.
    Asserted in Section III using D0 = 100 (1D) and D0 = 10 (2D); no D0 scan is presented.
  • ad hoc to paper For beta-prime <= beta the two debris fields are locally equal (h(A)_i about h(B)_i), reducing Eq. (5) to TSAW with effective coupling beta - beta-prime.
    Section IV; the authors themselves note this fails at beta-prime = beta and is unsuitable for beta-prime > B, so it only rationalizes the already-known regime.
  • ad hoc to paper In the pseudonormal regime, rare long excursions dominate the MSD while the bulk profile collapses with width t^(2/3).
    Section IV (strong sample-to-sample variability) and Fig. 3e; the coexistence of these two sectors at asymptotically large times is asserted, not derived, and is needed to reconcile P(x,t) scaling with alpha near 1.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Anomalous Diffusion and Emergent Universality in Coupled Memory-Driven Systems." pith.science (2026). https://pith.science/paper/A4CTHWJ7

@misc{pith2026241109092,
  author       = {Pith},
  title        = {Pith review of: Anomalous Diffusion and Emergent Universality in Coupled Memory-Driven Systems},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/A4CTHWJ7}},
  note         = {Machine review of arXiv:2411.09092}
}
read the original abstract

Understanding how simple local interactions give rise to emergent exploration patterns is a fundamental question in statistical physics. We introduce a minimal model of two coupled agents that avoid retracing their own paths while being attracted to the trails left by one another. This system is inspired by, but not limited to, pheromone-guided insect navigation. The coupling of self-avoidance and attraction generates rich emergent behavior, including distinct anomalous diffusion regimes, non-Gaussian position distributions, and compressed exponential encounter statistics. Most notably, we identify new universality classes for coupled random walks, characterized by unique scaling laws and distributional properties that, to our knowledge, have not been previously reported. These findings advance the theoretical understanding of coupled stochastic processes with memory and interaction feedback, providing a framework for exploring transport phenomena in a broad range of multi-agent systems beyond biological contexts.

Figures

Figures reproduced from arXiv: 2411.09092 by the authors.

Figure 1
Figure 1. FIG. 1 [PITH_FULL_IMAGE:figures/full_fig_p008_1.png] view at source ↗
Figure 2
Figure 2. a demonstrates the time dependence of an agent’s MSD for various values of β ′ , with a fixed coefficient β = 1 and D0 = 100. Figure 2b displays the mean-squared distance between two agents over time, again with varying β ′ values. These curves exhibit similar asymptotic trends compared to R2 t . The case of β = 0 is unique, as shown in Fig. 2c: for large values of β ′ , the exponent α = 1/2. Figure 2d illustrates h… view at source ↗
Figure 3
Figure 3. FIG. 3 [PITH_FULL_IMAGE:figures/full_fig_p011_3.png] view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: FIG. 4 [PITH_FULL_IMAGE:figures/full_fig_p012_4.png]
Figure 5
Figure 5. Figure 5: FIG. 5 [PITH_FULL_IMAGE:figures/full_fig_p014_5.png]
Figure 6
Figure 6. Figure 6: FIG. 6 [PITH_FULL_IMAGE:figures/full_fig_p015_6.png]
Figure 7
Figure 7. Figure 7: FIG. 7 [PITH_FULL_IMAGE:figures/full_fig_p016_7.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

54 extracted references · 53 canonical work pages

  1. [1]

    (c) P (T , t) for superdiffusion region with β′ = 1 .5, 5, and

    The inset demonstrates that rescaling T → T/t1/3 and P → t1/3P collapses the curves for different times onto a single curve. (c) P (T , t) for superdiffusion region with β′ = 1 .5, 5, and

    work page

  2. [2]

    E. A. Codling, M. J. Plank, and S. Benhamou, Random walk models in biology, J. R. Soc. Interface 5, 813 (2008)

    work page 2008

  3. [3]

    H. C. Berg, Random Walks in Biology(Princeton University Press, 1993)

    work page 1993

  4. [4]

    T. D. Wyatt, Pheromones, Curr. Biol. 27, R739 (2017)

    work page 2017

  5. [5]

    H. H. Shorey, Behavioral responses to insect pheromones, Annu. Rev. Entomol.18, 349 (1973)

    work page 1973

  6. [6]

    Two compressed exponential regimes are observed: exp( a1z3/2) for 0 .1 ≲ T /t ≲ 0.6 (black solid line) and exp( a2z5) for 0 .6 ≲ T /t ≲ 0.9 (black dashed line)

    Distributions exhibit high values for small T , indicating limited duration of meetings. Two compressed exponential regimes are observed: exp( a1z3/2) for 0 .1 ≲ T /t ≲ 0.6 (black solid line) and exp( a2z5) for 0 .6 ≲ T /t ≲ 0.9 (black dashed line). The coefficients a1 and a2 are functions of β and β′. (d) P (T , t= 10 7) for β = 1 , β′ = 5. The inset sho...

    work page

  7. [7]

    Dussutour, V

    A. Dussutour, V. Fourcassi´ e, D. Helbing, and J.-L. Deneubourg, Optimal traffic organization in ants under crowded conditions, Nature 428, 70 (2004)

    work page 2004

  8. [8]

    Metzler, J.-H

    R. Metzler, J.-H. Jeon, A. G. Cherstvy, and E. Barkai, Anomalous diffusion models and their properties: Non-stationarity, non-ergodicity, and ageing at the centenary of single particle tracking, Phys. Chem. Chem. Phys. 16, 24128 (2014)

    work page 2014

Show all 54 references
  1. [9]

    O. Vilk, E. Aghion, T. Avgar, C. Beta, O. Nagel, A. Sabri, R. Sarfati, D. K. Schwartz, M. Weiss, D. Krapf, et al., Unravelling the origins of anomalous diffusion: From molecules to migrating storks, Phys. Rev. Research 4, 033055 (2022)

    2022

  2. [10]

    Balakrishnan, Anomalous diffusion in one dimension, Physica A 132, 569 (1985)

    V. Balakrishnan, Anomalous diffusion in one dimension, Physica A 132, 569 (1985)

    1985

  3. [11]

    M. E. Newman and G. T. Barkema, Monte Carlo Methods in Statistical Physics(Clarendon 21 Press, 1999)

    1999

  4. [12]

    Bouchaud and A

    J.-P. Bouchaud and A. Georges, Anomalous diffusion in disordered media: Statistical mecha- nisms, models and physical applications, Phys. Rep. 195, 127 (1990)

    1990

  5. [13]

    Avin and B

    C. Avin and B. Krishnamachari, The power of choice in random walks: An empirical study, Comput. Netw. 52, 44 (2008)

    2008

  6. [14]

    Tsallis, Introduction to Nonextensive Statistical Mechanics: Approaching a Complex World (Springer, 2009)

    C. Tsallis, Introduction to Nonextensive Statistical Mechanics: Approaching a Complex World (Springer, 2009)

    2009

  7. [15]

    Bouchaud and M

    J.-P. Bouchaud and M. Potters, Theory of Financial Risk and Derivative Pricing: From Statistical Physics to Risk Management(Cambridge University Press, 2003)

    2003

  8. [16]

    N. G. Van Kampen, Stochastic Processes in Physics and Chemistry, Vol. 1 (Elsevier, 1992)

    1992

  9. [17]

    B´ enichou, C

    O. B´ enichou, C. Loverdo, M. Moreau, and R. Voituriez, Intermittent search strategies, Rev. Mod. Phys. 83, 81 (2011)

    2011

  10. [18]

    J. L. Deneubourg, S. Aron, S. Goss, and J. M. Pasteels, The self-organizing exploratory pattern of the argentine ant, J. Insect Behav. 3, 159 (1990)

    1990

  11. [19]

    D. J. Sumpter, Collective Animal Behavior(Princeton University Press, 2010)

    2010

  12. [20]

    Nakayama, H

    B. Nakayama, H. Nagase, H. Takahashi, Y. Saito, S. Hatayama, K. Makino, E. Yamamoto, and T. Saiki, Tunable pheromone interactions among microswimmers, Proc. Natl. Acad. Sci. U.S.A. 120, e2213713120 (2023)

    2023

  13. [21]

    d’Alessandro, A

    J. d’Alessandro, A. Barbier-Chebbah, V. Cellerin, O. Benichou, R. M. M` ege, R. Voituriez, and B. Ladoux, Cell migration guided by long-lived spatial memory, Nat. Commun. 12, 4118 (2021)

    2021

  14. [22]

    W. T. Kranz, A. Gelimson, K. Zhao, G. C. L. Wong, and R. Golestanian, Effective dynamics of microorganisms that interact with their wwn trail, Phys. Rev. Lett. 117, 038101 (2016)

    2016

  15. [23]

    W. T. Kranz and R. Golestanian, Trail-mediated self-interaction, J. Chem. Phys. 150, 214111 (2019)

    2019

  16. [24]

    Giuggioli, J

    L. Giuggioli, J. R. Potts, and S. Harris, Animal interactions and the emergence of territoriality, PLOS Comput. Biol. 7, e1002008 (2011)

    2011

  17. [25]

    J. R. Potts, K. Mokross, and M. A. Lewis, A unifying framework for quantifying the nature of animal interactions, J. R. Soc. Interface 11, 20140333 (2014)

    2014

  18. [26]

    D. J. Amit, G. Parisi, and L. Peliti, Asymptotic behavior of the “true” self-avoiding walk, Phys. Rev. B 27, 1635 (1983). 22

    1983

  19. [27]

    H. C. Ottinger, The generalised true self-avoiding walk-a model with continuously variable exponent ν, J. Phys. A 18, L363 (1985)

    1985

  20. [28]

    Peliti and L

    L. Peliti and L. Pietronero, Random walks with memory, Riv. Nuovo Cimento. 10, 1 (1987)

    1987

  21. [29]

    V. B. Sapozhnikov, Self-attracting walk with nu ν <1/2, J. Phys. A 27, L151 (1994)

    1994

  22. [30]

    Grassberger, How fast does a random walk cover a torus?, Phys

    P. Grassberger, How fast does a random walk cover a torus?, Phys. Rev. E 96, 012115 (2017)

    2017

  23. [31]

    Pemantle, A survey of random processes with reinforcement, Probab

    R. Pemantle, A survey of random processes with reinforcement, Probab. Surv. 4, 1 (2007)

    2007

  24. [32]

    J. G. Foster, P. Grassberger, and M. Paczuski, Reinforced walks in two and three dimensions, New J. Phys. 11, 023009 (2009)

    2009

  25. [33]

    Metzler, S

    R. Metzler, S. Redner, and G. Oshanin, First-Passage Phenomena and their Applications, Vol. 35 (World Scientific, 2014)

    2014

  26. [34]

    Gu´ erin, N

    T. Gu´ erin, N. Levernier, O. B´ enichou, and R. Voituriez, Mean first-passage times of non- markovian random walkers in confinement, Nature 534, 356 (2016)

    2016

  27. [35]

    Grassberger, Self-trapping self-repelling random walks, Phys

    P. Grassberger, Self-trapping self-repelling random walks, Phys. Rev. Lett.119, 140601 (2017)

    2017

  28. [36]

    Florey, Principles of Polymer Chemistry(Cornell University Press, New York, 1953)

    P. Florey, Principles of Polymer Chemistry(Cornell University Press, New York, 1953)

    1953

  29. [37]

    De Gennes, Scaling Concepts in Polymer Physics(Cornell University Press, 1979)

    P.-G. De Gennes, Scaling Concepts in Polymer Physics(Cornell University Press, 1979)

    1979

  30. [38]

    Pietronero, Critical dimensionality and exponent of the “true” self-avoiding walk, Phys

    L. Pietronero, Critical dimensionality and exponent of the “true” self-avoiding walk, Phys. Rev. B 27, 5887 (1983)

    1983

  31. [39]

    Obukhov and L

    S. Obukhov and L. Peliti, Renormalisation of the ‘true’ self-avoiding walk, J. Phys. A 16, L147 (1983)

    1983

  32. [40]

    Barbier-Chebbah, O

    A. Barbier-Chebbah, O. B´ enichou, and R. Voituriez, Self-interacting random walks: Aging, exploration, and first-passage times, Phys. Rev. X 12, 011052 (2022)

    2022

  33. [41]

    Dumaz and B

    L. Dumaz and B. T´ oth, Marginal densities of the “true” self-repelling motion, Stoch. Process. Their Appl. 123, 1454 (2013)

    2013

  34. [42]

    R´ egnier, M

    L. R´ egnier, M. Dolgushev, S. Redner, and O. B´ enichou, Universal exploration dynamics of random walks, Nat. Commun. 14, 618 (2023)

    2023

  35. [43]

    R´ egnier, M

    L. R´ egnier, M. Dolgushev, and O. B´ enichou, Record ages of non-Markovian scale-invariant random walks, Nat. Commun. 14, 6288 (2023)

    2023

  36. [44]

    Br´ emont, O

    J. Br´ emont, O. B´ enichou, and R. Voituriez, Exact propagators of one-dimensional self- interacting random walks, Phys. Rev. Lett. 133, 157101 (2024)

    2024

  37. [45]

    M. A. Prasad, D. P. Bhatia, and D. Arora, Diffusive behaviour of self-attractive walks, J. Phys. A 29, 3037 (1996). 23

    1996

  38. [46]

    Ordemann, E

    A. Ordemann, E. Tomer, G. Berkolaiko, S. Havlin, and A. Bunde, Structural properties of self-attracting walks, Phys. Rev. E 64, 046117 (2001)

    2001

  39. [47]

    Davis, Reinforced random walk, Probab

    B. Davis, Reinforced random walk, Probab. Theory Relat. Fields 84, 203 (1990)

    1990

  40. [48]

    Agliari, R

    E. Agliari, R. Burioni, and G. Uguzzoni, The true reinforced random walk with bias, New J. Phys. 14, 063027 (2012)

    2012

  41. [49]

    Alamgir and U

    M. Alamgir and U. von Luxburg, Multi-agent random walks for local clustering on graphs, IEEE Int. Conf. Data Mining , 18 (2010)

    2010

  42. [50]

    Dashti-N, M

    H. Dashti-N, M. N. Najafi, and H. Park, Self-repelling bipedal exploration process, Phys. Rev. E 104, 054135 (2021)

    2021

  43. [51]

    W. B. Grueber and A. Sagasti, Self-avoidance and tiling: Mechanisms of dendrite and axon spacing, Cold Spring Harb. Perspect. Biol. 2, a001750 (2010)

    2010

  44. [52]

    Cardy, Scaling and Renormalization in Statistical Physics, Vol

    J. Cardy, Scaling and Renormalization in Statistical Physics, Vol. 5 (Cambridge University Press, 1996)

    1996

  45. [53]

    Kardar, Statistical Physics of Fields(Cambridge University Press, 2007)

    M. Kardar, Statistical Physics of Fields(Cambridge University Press, 2007)

    2007

  46. [54]

    Anomalous Diffusion and Emergent Universality in Coupled Memory-Driven Systems

    M. Chupeau, O. B´ enichou, and R. Voituriez, Cover times of random searches, Nat. Phys. 11, 844 (2015). 24 Supplementary Material for “Anomalous Diffusion and Emergent Universality in Coupled Memory-Driven Systems” Nick Dashti, 1, 2,∗ M. N. Najafi, 3 and Debra J. Searles 1, 4, 5...

    2015

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.