-
Driven criticality links universal computation and optimal representations
Authors:
Adrián Roig,
Miguel A. Muñoz,
Guillermo B. Morales
Abstract:
Near-critical dynamics are often linked to enhanced computation, but the underlying mechanism remains unclear. We address this question in reservoir computing, where a fixed recurrent network maps input sequences into high-dimensional states and only a simple readout is trained. We extend fixed-reservoir universality results to discrete-time input-driven reservoirs and connect their key geometric…
▽ More
Near-critical dynamics are often linked to enhanced computation, but the underlying mechanism remains unclear. We address this question in reservoir computing, where a fixed recurrent network maps input sequences into high-dimensional states and only a simple readout is trained. We extend fixed-reservoir universality results to discrete-time input-driven reservoirs and connect their key geometric condition, neighborhood separation, to dynamics. To this end, we introduce a finite-resolution neighborhood separability index and an input-conditioned maximal Lyapunov exponent. We find that neighborhood separability, chaotic time-series prediction, and smooth high-dimensional representation geometry are optimized in the same narrow window of marginal driven stability. In this regime, the covariance spectrum approaches the power-law scaling expected for near-optimal smooth representations. Our results link edge-of-instability computation, universality, readout performance, and optimal representation geometry within a common dynamical framework.
△ Less
Submitted 27 July, 2026; v1 submitted 23 July, 2026;
originally announced July 2026.
-
Ergodicity Breaking and High-Dimensional Chaos in Random Recurrent Networks
Authors:
Carles Martorell,
Rubén Calvo,
Adrián Roig,
Alessia Annibale,
Miguel A. Muñoz
Abstract:
The neural model introduced by Sompolinsky, Crisanti, and Sommers (SCS) nearly four decades ago has become a paradigmatic framework for studying complex dynamics in random recurrent networks. In its original formulation, with balanced positive and negative couplings, the model exhibits two phases: a quiescent regime, where all activity ceases, and a regime of ongoing irregular collective activity,…
▽ More
The neural model introduced by Sompolinsky, Crisanti, and Sommers (SCS) nearly four decades ago has become a paradigmatic framework for studying complex dynamics in random recurrent networks. In its original formulation, with balanced positive and negative couplings, the model exhibits two phases: a quiescent regime, where all activity ceases, and a regime of ongoing irregular collective activity, termed asynchronous chaos (AC), in which state variables fluctuate strongly in time and across units but average to zero across the network. Building on recent work, we analyze an extension of the SCS model that breaks this coupling balance, yielding a richer phase diagram. In addition to the classical quiescent and AC phases, two additional regimes emerge, marked by spontaneous symmetry breaking. In the persistent-activity (PA) phase, each unit settles into a distinct, stable activation state. In the synchronous-chaotic (SC) phase, dynamics remain irregular and chaotic but fluctuate around a nonzero mean, generating sustained long-time autocorrelations. Using analytical techniques based on dynamical mean-field theory, complemented by extensive numerical simulations, we show how structural disorder gives rise to symmetry and ergodicity breaking. Remarkably, the resulting phase diagram closely parallels that of the Sherrington-Kirkpatrick spin-glass model, with the onset of the SC phase coinciding with the transition associated with replica-symmetry breaking. All key features of spin glasses, including ergodicity breaking, have clear counterparts in this recurrent network context, albeit with crucial idiosyncratic differences, highlighting a unified perspective on complexity in disordered systems.
△ Less
Submitted 9 October, 2025;
originally announced October 2025.
-
Robust Scaling in Human Brain Dynamics Despite Latent Variables and Limited Sampling Distortions
Authors:
Rubén Calvo,
Carles Martorell,
Adrián Roig,
Miguel A. Muñoz
Abstract:
The idea that information-processing systems operate near criticality to enhance computational performance is supported by scaling signatures in brain activity. However, external signals raise the question of whether this behavior is intrinsic or input-driven. We show that autocorrelated inputs and temporal resolution influence observed scaling exponents in simple neural models. We also demonstrat…
▽ More
The idea that information-processing systems operate near criticality to enhance computational performance is supported by scaling signatures in brain activity. However, external signals raise the question of whether this behavior is intrinsic or input-driven. We show that autocorrelated inputs and temporal resolution influence observed scaling exponents in simple neural models. We also demonstrate analytically that under subsampling, non-critical systems driven by independent autocorrelated signals can exhibit strong signatures of apparent criticality. To address these pitfalls, we develop a robust framework and apply it to pooled neural data, revealing resting-state brain activity at the population level is slightly sub-critical yet near-critical. Notably, the extracted critical exponents closely match predictions from a simple recurrent firing-rate model, supporting the emergence of near-critical dynamics from reverberant network activity, with potential implications for information processing and artificial intelligence.
△ Less
Submitted 4 June, 2025;
originally announced June 2025.
-
Nanorods based on mesoporous silica containing iron oxide nanoparticles as catalytic nanomotors: study of motion dynamics
Authors:
Rafael Mestre,
Núria Cadefau,
Ana C. Hortelão,
Jan Grzelak,
Martí Gich,
Anna Roig,
Samuel Sánchez
Abstract:
Self-propelled particles and, in particular, those based on mesoporous silica, have raised considerable interest due to their potential applications in the environmental and biomedical fields thanks to their biocompatibility, tunable surface chemistry and large porosity. Although spherical particles have been widely used to fabricate nano- and micromotors, not much attention has been paid to other…
▽ More
Self-propelled particles and, in particular, those based on mesoporous silica, have raised considerable interest due to their potential applications in the environmental and biomedical fields thanks to their biocompatibility, tunable surface chemistry and large porosity. Although spherical particles have been widely used to fabricate nano- and micromotors, not much attention has been paid to other geometries, such as nanorods. Here, we report the fabrication of self-propelled mesoporous silica nanorods (MSNRs) that move by the catalytic decomposition of hydrogen peroxide by a sputtered Pt layer, Fe2O3 nanoparticles grown within the mesopores, or the synergistic combination of both. We show that motion can occur in two distinct sub-populations characterized by two different motion dynamics, namely enhanced diffusion or directional propulsion, especially when both catalysts are used. These results open up the possibility of using MSNRs as chassis for the fabrication of self-propelled particles for the environmental or biomedical fields.
△ Less
Submitted 5 November, 2020;
originally announced November 2020.
-
Multiferroic Iron Oxide Thin Films at Room-Temperature
Authors:
Marti Gich,
Ignasi Fina,
Alessio Morelli,
Florencio Sanchez,
Marin Alexe,
Jaume Gazquez,
Josep Fontcuberta,
Anna Roig
Abstract:
In spite of being highly relevant for the development of a new generation of information storage devices, not many single-phase materials displaying magnetic and ferroelectric orders above room temperature are known. Moreover, these uncommon materials typically display insignificant values of the remanent moment in one of the ferroic orders or are complex multicomponent oxides which will be very c…
▽ More
In spite of being highly relevant for the development of a new generation of information storage devices, not many single-phase materials displaying magnetic and ferroelectric orders above room temperature are known. Moreover, these uncommon materials typically display insignificant values of the remanent moment in one of the ferroic orders or are complex multicomponent oxides which will be very challenging to integrate in devices. Here we report on the strategy to stabilize the metastable epsilon-Fe2O3 in thin film form, and we show that besides its already known ferrimagnetic nature, the films are also ferroelectric at 300 K with a remanent polarization of 1 microC/cm2. The film polarization shows long retention times and can be switched under small applied voltages. These characteristics make of epsilon-Fe2O3 the first single-ion transition-metal oxide which is ferro(ferri)magnetic and ferroelectric at room temperature. The simple composition of this new multiferroic oxide and the discovery of a robust path for its thin film growth may boost the exploitation of epsilon-Fe2O3 in novel devices.
△ Less
Submitted 19 May, 2014;
originally announced May 2014.
-
New ultrasmall iron-oxide nanoparticles with high magnetisation as potential T1-MRI contrast agents for Molecular Imaging
Authors:
Elena Taboada,
Elisenda Rodriguez,
Anna Roig,
Judit Oro,
Alain Roch,
Robert N. Muller
Abstract:
Here we report on the synthesis of very small gamma-Fe2O3 nanoparticles (5 nm) presenting very narrow particle size distribution and exceptionally high saturation magnetisation. The synthesis has been carried out in an organic medium with subsequent transfer to an aqueous solution at physiological pH. The structural and magnetic properties were kept unaltered after the solvent exchange. NMR rela…
▽ More
Here we report on the synthesis of very small gamma-Fe2O3 nanoparticles (5 nm) presenting very narrow particle size distribution and exceptionally high saturation magnetisation. The synthesis has been carried out in an organic medium with subsequent transfer to an aqueous solution at physiological pH. The structural and magnetic properties were kept unaltered after the solvent exchange. NMR relaxometric measurements show the potential of these particles as specific reporters for magnetic resonance molecular imaging.
△ Less
Submitted 9 November, 2006;
originally announced November 2006.
-
High and low-temperature crystal and magnetic structures of epsilon-Fe2O3 and their correlation to its magnetic properties
Authors:
M. Gich,
C. Frontera,
A. Roig,
E. Taboada,
E. Molins,
H. R. Rechenberg,
J. D. Ardisson,
W. A. A. Macedo,
C. Ritter,
V. Hardy,
J. Sort,
V. Skumryev,
J. Nogues
Abstract:
The crystal and magnetic structures of the orthorhombic e-Fe2O3 have been studied by simultaneous Rietveld refinement of X-ray and neutron powder diffraction data in combination with Mossbauer spectroscopy, as well as magnetisation and heat capacity measurements. It has been found that above 150 K the e-Fe2O3 polymorph is a collinear ferrimagnet with the magnetic moments directed along the a axi…
▽ More
The crystal and magnetic structures of the orthorhombic e-Fe2O3 have been studied by simultaneous Rietveld refinement of X-ray and neutron powder diffraction data in combination with Mossbauer spectroscopy, as well as magnetisation and heat capacity measurements. It has been found that above 150 K the e-Fe2O3 polymorph is a collinear ferrimagnet with the magnetic moments directed along the a axis, while the magnetic ordering below 80 K is characterised by a square-wave incommensurate structure. The transformation between these two states is a second order phase transition and involves subtle structural changes mostly affecting the coordination of the tetrahedral and one of the octahedral Fe sites. The temperature dependence of the e-Fe2O3 magnetic properties is discussed in the light of these results.
△ Less
Submitted 28 April, 2006;
originally announced April 2006.
-
Magnetoelectric Coupling in epsilon-Fe2O3
Authors:
M. Gich,
C. Frontera,
A. Roig,
E. Molins,
J. Fontcuberta,
N. Bellido,
Ch. Simon,
C. Fleta
Abstract:
Nanoparticles of the ferrimagnetic epsilon-Fe2O3 oxide have been synthesized by sol-gel method. Here, we report on the measurements of the dielectric permittivity as a function of temperature, frequency and magnetic field. It is found that, coinciding with the transition from collinear ferrimagnetic ordering to an incommensurate magnetic state occurring at about 100 K, there is an abrupt change…
▽ More
Nanoparticles of the ferrimagnetic epsilon-Fe2O3 oxide have been synthesized by sol-gel method. Here, we report on the measurements of the dielectric permittivity as a function of temperature, frequency and magnetic field. It is found that, coinciding with the transition from collinear ferrimagnetic ordering to an incommensurate magnetic state occurring at about 100 K, there is an abrupt change (about 30 %) of permittivity suggesting the existence of a magnetoelectric coupling in this material. Indeed, magnetic field dependent measurements at 100 K have revealed an increase of the permittivity by about 0.3 % in 6 T. Prospective advantages of epsilon-Fe2O3 as multiferroic material are discussed.
△ Less
Submitted 5 September, 2005;
originally announced September 2005.
-
Cationic ordering control of magnetization in Sr2FeMoO6 double perovskite
Authors:
Ll. Balcells,
J. Navarro,
M. Bibes,
A. Roig,
B. Martinez,
J. Fontcuberta
Abstract:
The role of the synthesis conditions on the cationic Fe/Mo ordering in Sr2FeMoO6 double perovskite is addressed. It is shown that this ordering can be controlled and varied systematically. The Fe/Mo ordering has a profound impact on the saturation magnetization of the material. Using the appropriate synthesis protocol a record value of 3.7muB/f.u. has been obtained. Mossbauer analysis reveals th…
▽ More
The role of the synthesis conditions on the cationic Fe/Mo ordering in Sr2FeMoO6 double perovskite is addressed. It is shown that this ordering can be controlled and varied systematically. The Fe/Mo ordering has a profound impact on the saturation magnetization of the material. Using the appropriate synthesis protocol a record value of 3.7muB/f.u. has been obtained. Mossbauer analysis reveals the existence of two distinguishable Fe sites in agreement with the P4/mmm symmetry and a charge density at the Fe(m+) ions significantly larger than (+3) suggesting a Fe contribution to the spin-down conduction band. The implications of these findings for the synthesis of Sr2FeMoO6 having optimal magnetoresistance response are discussed.
△ Less
Submitted 14 July, 2000;
originally announced July 2000.