Biological Cybernetics

A central question in neuroscience is how neural processing generates or encodes behavior. Caenorhabditis elegans is well suited to addressing this question, given its compact nervous system and near-complete structural connectome. Despite this, findings from previous studies remain inconclusive. While some have shown that the connectome can robustly encode specific behaviors such as locomotion, others report that functional connectivity can be reconfigured across behaviors. We aim to understand the relationship between structural connectivity, functional connectivity and biological behavior in silico by using an experimentally motivated computational model leveraging the structural connectome. Stimulation of specific neurons in the model induces oscillatory neural responses, enabling us to infer neuronal functional connectivity. Functional connectivity is found to be stronger among some neurons, allowing us to identify functional communities. We find that electrical synapses play a critical role in determining functional communities, and the resulting mesoscale functional architecture is predominantly gap junctionally assortative. Furthermore, comparison with behavioral circuits shows that locomotion circuits are largely segregated into distinct functional communities while other circuits are more distributed across multiple functional communities. We also observe that stimulation of neurons belonging to these distributed circuits elicits a more synchronized neuronal response compared to stimulation of neurons within the more segregated circuits. This is consistent with the presence of behavioral patterns that originate in one circuit and terminate in another (e.g., chemosensation leading to locomotion), such that stimulation of one circuit can activate the other and eventually result in a synchronized response. We also find a large repertoire of chimera-like synchronization patterns upon stimulation of certain behavioral circuits (chemosensation, mechanosensation) indicating high dynamical flexibility. Overall, our results demonstrate that while certain behaviors are governed by functionally segregated circuits, others emerge from the synchronization of multiple functional communities, which are, to begin with, influenced by the underlying structural connectivity. Author summaryAnimals constantly transform sensory inputs into actions, but it is still unclear how this mapping from neural activity to behavior is implemented in a real nervous system. Caenorhabditis elegans offers a unique testbed for this question because its entire wiring diagram is nearly completely mapped. Yet, previous works have reached mixed conclusions about how well this anatomical circuit diagram predicts actual patterns of activity and behavior. Here, we use a biologically inspired computational model of the C. elegans nervous system to bridge this gap between structure, function, and behavior. By virtually stimulating individual neurons and observing the resulting network-wide oscillations, we infer how strongly different pairs and groups of neurons interact in functional terms. We then use network analysis tools to identify groups of neurons that tend to co-activate, and relate these functional communities to known behavioral circuits for locomotion and sensory processing. We find that gap junctions play a key role in shaping functional communities, and that locomotion-related neurons are more functionally segregated than neurons involved in other behaviors, which are more functionally distributed. Our results suggest that some behaviors rely on specialized, functionally isolated circuits, whereas others emerge from the coordinated activity of multiple functional communities.

Muscle mass significantly influences skeletal muscle behaviour, potentially explaining why traditional massless Hill-type models struggle to predict the forces generated by larger muscles during dynamic, submaximal contractions. However, the applicability of mass-enhanced Hill-type models in human locomotion remains unexplored. Here, we compared the predicted force from a 1D mass-enhanced Hill-type muscle model with a traditional 1D massless Hill-type muscle model across a range of experimentally measured human movements. Kinematic and electromyographic data were collected from twenty participants performing locomotor tasks and supplemented with existing cycling data. Muscle size was geometrically scaled by factors from 0.1 to 10, which causes lengths to be scaled proportionally, cross-sectional area and peak isometric force F0 with the square, and mass with the cube of the factor. Muscle tissue mass (inertia) and cadence increased the differences between mass-enhanced and massless predictions of force and power. At high cadence and the largest scale, the normalized root mean square difference between force traces reached 7% of F0, (averaged across muscles). However, differences between models were minimal (<1%) at human-sized scale 1. Real muscle additionally deforms in 3D, we still do not know the extent to which this extra dimensionality affects muscle forces for these human movements.

A key component of intraneuronal communication is the modulation of postsynaptic firing frequencies by stochastic transmitter release from presynaptic neurons. The time interval between successive postsynaptic firings is called the inter-spike interval (ISI), and understanding its statistics is integral to neural information processing. We start with a model of an excitatory chemical synapse with postsynaptic neuron firing governed as per a classical integrate-and-fire model. Using a first-passage time framework, we derive exact analytical results for the ISI statistical moments, revealing parameter regimes driving precision in postsynaptic action potential timing. Next, we extended this analysis to include both an excitatory and an inhibitory presynaptic connection onto the same postsynaptic neuron. We consider both a fixed postsynaptic-firing threshold and a threshold that adapts based on the postsynaptic membrane potential history. Our analysis shows that the latter adaptive threshold can result in scenarios where increasing the inhibitory input frequency increases the postsynaptic firing frequency. Moreover, we characterize parameter regimes where ISI noise is hypo-exponential or hyperexponential based on its coefficient of variation being less than or higher than one, respectively.

Neuronal activity alters extracellular ion concentrations and electric potentials. Ephaptic effects refer to the feedback influence that these extracellular changes can have on neuronal activity. While electric ephaptic effects occur on a fast timescale due to extracellular potential perturbations, ionic ephaptic effects are driven by slower, accumulative changes in ion concentrations. Among the previous computational studies of ephaptic effects, the vast majority have focused exclusively on electric effects, while ionic ephaptic effects have largely been neglected. In this work, we present an electrodiffusive computational framework consisting of two-compartment neurons that interact via a shared extracellular space. By accounting for both electric potentials and ion-concentration dynamics in a self-consistent manner, our framework enables us to explore the relative roles of electric and ionic ephaptic effects. Through numerical experiments, we demonstrate that ionic and electric ephaptic interactions play very different roles. While ionic ephaptic interactions increase population firing rates, electric ephaptic interactions primarily drive subtle shifts in spike timing. Furthermore, we show that these spike shifts cause the phase difference (the distance in spike times between a small collection of neurons) to converge to a stable, unique phase difference, which we coin the ephaptic intrinsic phase preference. Author summaryNeurons predominantly communicate through synapses: specialized contact points where a brief electrical signal, known as a spike or action potential, in one neuron influences another. Neurons generate these spikes by exchanging ions with the surrounding extracellular space. This way, spiking neurons alter extracellular ion concentrations and electric potentials. Since neurons are sensitive to such changes in their environment, they can also influence one another indirectly through the shared extracellular medium. This form of non-synaptic interaction is known as ephaptic coupling. Most computational models of neuronal activity neglect ephaptic interactions, and those that include them typically consider only electric effects while ignoring ionic contributions. As a result, the relative roles of electric and ionic ephaptic effects remain poorly understood. Here, we introduce a computational framework that accounts for both mechanisms in a self-consistent way. Our results show a functional distinction: ionic ephaptic effects act slowly, regulating population firing rates, whereas electric ephaptic effects act on millisecond timescales and subtly shift spike timing. These shifts cause spike-time differences between neurons to converge to a stable value, a phenomenon we call ephaptic intrinsic phase preference.

Voltage perturbations to a repetitively firing Hodgkin-Huxley (HH) model of neuronal spiking in the bistable regime with coexisting limit cycle and stable steady node can either lead to the spikes phase resetting or collapse to the stable steady state. The latter describes a non-firing hyperpolarized quiescent state of the neuron despite the presence of constant external current. Using asymptotic phase response curve (PRC), the impact of voltage perturbations on a repetitively firing HH model is studied here while it is diffusively coupled to another HH model under identical external stimulation. It is observed that the pre-perturbation state of synchronization and the coupling strength critically determine the PRC response of the perturbed HH dynamics. Higher coupling strengths of perfectly in-phase (anti-phase) synchronized HH models shrink (expand) the combinatorial space of perturbation strengths and the oscillation phases causing collapse to the quiescent state. This indicates reduced (enlarged) basin of attraction, viz. the null space, associated with the steady state in the HH phase space. The findings bear important implications to the spiking dynamics of diverse interneurons, as well as special cases of pyramidal neurons, coupled through electrical synapses via. gap junctions, and suggest the role of gap junction plasticity in tuning vulnerability to quiescent state in the presence of biological noise and spikelets.

Clavicle fractures often exhibit markedly different clinical outcomes: some patients recover acceptable function despite shortening or displacement, whereas others with apparently similar deformity develop persistent pain, functional loss, or poor healing. To explain this distinction, we propose a minimal nonlinear mechanical model for prognostic analysis of clavicle fractures. The model describes the interaction between fracture-related shortening and compensatory shoulder-girdle posture through a reduced equilibrium equation incorporating stiffness, geometric nonlinearity, and shortening-posture coupling. Within this framework, we analyze equilibrium branches, local stability, and the emergence of critical thresholds. We show that post-fracture destabilization can be interpreted as a fold bifurcation, while more complex parameter dependence gives rise to cusp-type structures and multistability. These bifurcation mechanisms provide a mathematical explanation for sudden deterioration after injury or treatment, as well as for strong inter-individual variability. We further introduce an optimization principle based on a utility functional to guide treatment planning. The analysis predicts that the optimal safe correction should lie strictly below the bifurcation threshold, thereby generating a natural safety margin. Although the model is simplified and has not yet been calibrated against patient data, it nevertheless provides a theoretical framework for understanding why fracture prognosis may deteriorate abruptly near critical mechanical conditions and offers a dynamical-systems interpretation of empirical treatment thresholds used in clinical practice.

Cerebral blood flow (CBF) control is essential for normal brain function and is disrupted in pathological conditions. Arterial diameters are tightly regulated to provide on demand increases in blood flow in regions of neuronal activity. Pericytes (PCs) exhibit robust myogenic tone and may also respond to neuronal activity to fine-tune local resistance and blood flow. Thus, mural control of microcirculatory resistance may extend beyond arteries and arterioles. Yet, PCs electrophysiology and contractility have not been thoroughly characterized, and this prohibits an integrated view of brain blood flow control. In this study, we develop a detailed mathematical model of mural cell electrophysiology, Ca2+ dynamics and biomechanics. The model is informed by electrophysiological data in smooth muscle cells (SMCs) or PCs and predictions are compared against pressure-induced responses in isolated arterioles and capillaries, respectively. Simulations recapitulate myogenic constrictions and examine differences in contractile dynamics as we move from arterioles to proximal and distal capillaries. In arteriole-to-capillary transitional (ACT) zone PCs, increased mechanosensitivity, more Ca2+ influx through non-selective cation (NSC) channels and/or a higher sensitivity of the contractile apparatus to Ca2+ can compensate for reduced L-type voltage-operated (VOCC) Ca2+ influx and allow for robust constrictions at the lower operating pressures of capillaries relative to the arterioles. A significant Ca2+ influx through NSC relative to VOCC, however, can decouple the PCs contractile apparatus from electrical signaling. Vasoactivity to chemomechanical stimuli along the arteriole to capillary axis is progressively driven by VOCC-independent Ca2+ influx and Ca2+ sensitization with slow kinetics. The proposed cell model can form the basis for detailed multiscale and multicellular models that will examine physiological function at a single vessel or vascular network levels and investigate CBF control in health and in disease. Key pointsO_LIA mural cell model of electrophysiology, calcium (Ca2+) dynamics and biomechanics is informed by data and adapted for modeling cerebral arteriole smooth muscle cells and capillary pericytes. C_LIO_LIIon channel activities are characterized by patch-clamp electrophysiology in isolated cerebral smooth muscle cell and pericytes, and capillary and arteriole electromechanical responses to transmural pressure changes are assessed using novel ex vivo preparations. C_LIO_LIMyogenic constrictions in arterioles can be reproduced by pressure-induced non-selective cation channel (NSC) activation that depolarizes the cell, opens L-type Ca2+ channels (VOCCs) and increases Ca2+ influx. C_LIO_LIRobust myogenic constrictions in arteriole-to-capillary transition (ACT) zone pericytes may reflect significant Ca2+ influx through NSC, increased mechanosensitivity, or higher sensitivity of the contractile apparatus to Ca2+, potentially compensating for reduced VOCC density relative to arteriolar smooth muscle. C_LIO_LIA significant contribution of NSC relative to VOCC in Ca2+ influx, can decouple the contractile apparatus from electrical signaling. C_LIO_LIThe model shows how gradients in ionic activities, mechanosensitivity and/or Ca2+ sensitivity can alter contractile phenotype and electromechanical coupling along the arteriole to capillary continuum. C_LIO_LIThe proposed model can form the basis for detailed multiscale and multicellular models that will investigate cerebral blood flow control in health and in disease. C_LI

In "Experiment-free exoskeleton assistance via learning in simulation", Luo et al. [1] present an ambitious framework for developing exoskeleton controllers through reinforcement learning exclusively in computer simulation. The authors report that a control policy trained on a small dataset from one subject was directly transferred to physical hardware, reducing human metabolic cost during walking, running, and stair climbing by more than any prior device. If confirmed, this would represent a major breakthrough for the field of wearable robotics and their clinical applications. However, a close examination of the published materials casts doubt on these claims. The reported experimental results violate physiological limits on the relationship between mechanical power and muscle energy use during gait2,3,4. The algorithmic claims are surprising and cannot be verified; in contrast with established replicability standards in machine learning5,6, executable code has not been made available. We conclude that the goals of this study have not yet been verifiably achieved and make recommendations for avoiding publication errors of this type in the future.

The environments inhabited by flying insects demand a balance between flight efficiency and flight manoeuvrability. In structural oscillators such as the insect indirect flight motor, efficiency (arising from resonance) and manoeuvrability (arising from kinematic modulation) are typically quid pro quo, with modulation incurring penalties to efficiency. Band-type resonance is a phenomenon that offers, in theory, a strategy to lessen these penalties via careful navigation through a band of efficient kinematic states. However, identifying this band is challenging: no methods exist to identify the complete band in realistic motor models, involving elasticity distributed across thorax and wing. Nor are the effects of elasticity distribution on the band known. In this work, we address both open topics. We present a suite of numerical methods for identifying the complete resonance band in general systems. Applying them to models of the insect flight motor with distributed elasticity--thoracic and wing flexion--reveals that distributed elasticity is moderate-risk but high-reward morphological feature. Well-tuned distributions expand the resonance band over fourfold whereas poorly-tuned distributions completely extinguish the resonance band. These results indicate that distributing elasticity across the insect flight motor can have adaptive value, and motivate broader work identifying distributions across species.

Despite its ubiquity in natural flows, the effects of turbulence on fish locomotion and behavior remain poorly understood. The prevailing hypothesis is that these effects depend on the spatial and temporal scales of the turbulence relative to the fishs size and swimming speed. But in conventional facilities, turbulence usually increases with mean flow, which forces higher swimming speeds and can leave these relative scales unchanged. We therefore present a novel experimental facility that leverages a jet array to decouple the turbulence from the mean flow and systematically control its scales. This approach allows the ratio of turbulent to fish inertial scales to be varied over an order of magnitude, providing a controlled framework for quantifying fish-turbulence interactions. The facility also supports experiments probing strategies fish may use to cope with turbulence, including collective behaviors. Insights from this work have broader implications for ecological studies and engineering applications, including the design of effective fishways and bio-inspired underwater vehicles.

Understanding how animals produce a versatile locomotor repertoire requires unraveling the interplay between higher centers, decentralized locomotor circuits, and sensory feedback. However, the principles governing their integration remain elusive. We investigated amphibious centipedes through stepwise neural lesions and neuromechanical modeling. Behavioral experiments revealed that while decentralized circuits autonomously generate coordination, the brain and subesophageal ganglion provide situational flexibility, such as modulating trunk undulation and initiating leg folding. Integrating these findings, our model demonstrated how higher centers selectively inhibit or release lower circuit dynamics. Simulations verified that varying only a few descending control parameters reproduces transitions between slow walking, fast walking, and swimming. This work may capture the essence of the locomotor circuitry that harnesses decentralized self-organization to coordinate the bodys large degrees of freedom.

Spaceflight-Associated Neuro-ocular Syndrome (SANS) involves complex interactions between intracranial pressure (ICP), intraocular pressure (IOP), and cerebrospinal fluid (CSF) dynamics within the optic nerve subarachnoid space (ONSAS). While existing computational models address specific aspects of these interactions, they lack a comprehensive, system-level representation. To bridge this gap, we present the HEAD (Hemodynamic Eye-brain Associated Dynamics) model. By consistently integrating several previously proposed physiological sub-models, HEAD provides a unified lumped-parameter framework that fully couples cerebrovascular autoregulation, multi-territory ocular hemodynamics, and compartmentalized craniospinal-ONSAS CSF circulation under gravitational loading. This formulation enables the simultaneous analysis of eye-brain-CSF dynamics within a single computational tool. Model predictions were validated against experimental data from supine (0{degrees}) to head-down tilt (HDT, -30{degrees}) postures, accurately reproducing posture-dependent IOP increases and achieving an excellent ICP match against clinical benchmarks at the -6{degrees} HDT standard bed-rest angle. The coupled system predicts bed-specific ocular hemodynamic responses, with retinal blood flow exhibiting the largest relative increase under HDT compared to the ciliary and choroidal circulations. Crucially, explicitly modeling the ONSAS as a distinct compartment reveals a posture-dependent pressure drop of 1.89-3.69 mmHg between the intracranial and perioptic spaces. This compartmentalization yields a translaminar pressure profile that remains positive (8.05-11.83 mmHg) across all simulated conditions but is chronically reduced under sustained HDT. Ultimately, the HEAD model elucidates the physiological mechanisms linking gravitational stress to translaminar mechanics, providing a robust computational foundation to investigate SANS and supply boundary conditions for structural models of the optic nerve head.

Sensorimotor learning and tool use involve synchronizing with external dynamics. Many everyday tools possess nonlinear hidden dynamics. Here we investigate how learning to synchronize with the complex dynamics of a tool depends on the degree of predictability and reciprocal coupling between user and tool. We introduce the concept of optimal coupling to measure adaptive user-tool coordination. Groups of participants practiced tracking an auditory stimulus in three conditions: 1) the tool was non-interactive and produced a periodic stimulus, 2) non-interactive and unstable stimulus, and 3) unstable but interactive stimulus which was coupled weakly to the participants movements and thus afforded control. Learning, retention, and transfer to visual modality were assessed using unpracticed test stimuli. Directional effective coupling was quantified using transfer entropy. Results showed that learning tended to be task-specific and there was no transfer to the visual modality. Interactive unstable practice exhibited some retention and generalization. We found a convergent reorganization of coupling during practice with the interactive unstable tool: stimulus-to-human coupling started high and decreased while human-to-stimulus coupling started low and increased. This suggests that embodiment of personalized rehabilitation technologies brings optimal reciprocal coupling in which sensorimotor-tool control is consistent with the minimal intervention principle postulated for within-body control.

Receptive fields provide a concise description of the stimulus selectivity of visual neurons. But this stimulus selectivity is neither static nor linear, and these nonlinear effects are not well captured by standard linear or pseudo-linear receptive field models. At the same time, receptive field models incorporating nonlinear effects are largely empirical, and are not easily interpreted in terms of underlying cellular and synaptic mechanisms. Here we show that two nonlinear mechanisms in the primate outer retina shape neural responses and that these contribute significantly to responses to natural stimuli and to the retinal output signals. Incorporating these outer retinal nonlinearities into models for visual function will improve our ability to identify the mechanistic origin of specific features of downstream visual responses.

Building a simple model that precisely and functionally characterizes a neuron is a challenging and important task to select the best concise and computationally efficient model. However, this type of work has only been done for subthreshold properties of neurons. Here, we take a different perspective and suggest a method to obtain point-neuron models from morphologically-detailed models with dendrites. To do this, we focus on the functional characterization of the neuron response under in vivo conditions, and compute the transfer function of the detailed model. The parameters of this transfer function, in terms of mean voltage, voltage standard deviation and correlation time, can be used to compute the "best" point-neuron model that generates a transfer function very close to that of the morphologically-detailed model. We illustrate this approach for two very different neuronal morphologies, one from Drosophila larvae and one from mammals. In conclusion, this approach provides a tool to generate point-neuron models from detailed models, based on a functional characterization of the neuron response. Significance StatementThis study provides a new computational method to reduce morphological models into point-neuron models. To do so, we calculate the transfer function parameters, ie the voltage standard deviation, the mean voltage and the correlation time, of the morphological model and fit a point neuron-model onto this data. Here, we successfully apply this approach for two very different neuron morphologies, a drosophila neuron and a rat motoneuron.

Local field potentials recorded from the subthalamic nucleus (STN) in Parkinsons disease (PD) exhibit a distinctive multiscale spectral signature: exaggerated beta-band oscillations (13-30 Hz) coupled to high-frequency oscillations (HFOs, 200-400 Hz), with HFO amplitude being phase-locked to the beta cycle. This phase-amplitude coupling (PAC) has been identified as a promising biomarker of the parkinsonian state, yet no biophysical model has explained how it emerges, what determines the HFO frequency, or how HFOs can exist without beta modulation in the medicated STN. Here we show that a heterogeneous population of excitatory Izhikevich neurons with recurrent coupling produces three dynamical regimes: (i) asynchronous tonic firing, (ii) asynchronous bursting, in which neurons burst individually producing broadband HFO power but without coherent population-level PAC, and (iii) synchronous bursting, which gives rise to beta-HFO PAC. The regimes are governed by two biophysically interpretable parameters that capture complementary effects of dopamine depletion: one reflecting changes in intrinsic neuronal excitability, the other reflecting changes in synaptic coupling strength. The transition from asynchronous to synchronous bursting in this model captures the emergence of pathological STN neuronal activity in the parkinsonian state. HFO peak frequency varies continuously across the two-parameter landscape, providing a mechanistic account of the clinically observed shift from slow (200-300 Hz) to fast (300-400 Hz) HFOs between medication states. The character of the synchronization transition depends on baseline excitability, ranging from a sharp co-emergence of bursting and synchrony at low excitability to a decoupled two-stage process at intermediate excitability where burst recruitment precedes synchronization. The model generates testable predictions for future clinical and experimental studies, provides a numerical dissection of how mesoscopic LFP features map onto microscopic neuronal dynamics, and serves as a computational building block for future circuit-level models that can guide brain stimulation strategies tailored to the patient-specific dynamical state of the STN. Author summaryIn Parkinsons disease, local field potentials (LFP) from the subthalamic nucleus (STN) contain two prominent rhythms: a slow beta rhythm (13-30 Hz) and fast oscillations (200-400 Hz). In the parkinsonian state, these rhythms become coupled, with fast oscillation amplitude varying systematically with beta phase, a relationship absent in the medicated state. We built a biophysical spiking neuron network model that captures two key effects of dopamine depletion on STN neuronal activity: changes in the intrinsic neuronal excitability and changes in synaptic coupling strength. The model produces fast oscillations from rapid intraburst firing, while the slow beta rhythm and its coupling to fast oscillations emerge with the onset of synchronized bursting across the population. Importantly, the frequency of the fast oscillations shifts continuously depending on both parameters, explaining a puzzling clinical observation that these oscillations change frequency between medication states. The model also reproduces the modulation pattern in the spike-triggered average of HFO envelope amplitude reported in patient recordings, confirming consistency with single-unit observations as well as LFP-level spectral features. By mapping how multi-timescale LFP spectral features relate to the dynamical regime of the underlying neuronal population, this work offers a framework for brain stimulation strategies informed by patient-specific dynamical states.

Spinal motoneuron (MN) pools behave as linear systems that transmit common synaptic input to muscles. However, MNs are biophysically heterogeneous and intrinsically nonlinear. How different MN subpopulations integrate and transmit high-frequency inputs remains poorly understood, partly because conventional analyses treat the MN pool as a single functional system rather than examining subpools with different firing rates. Here, we addressed this gap using a combination of computational simulations and human MN recordings. Simulations of MNs receiving a common synaptic input at varying frequencies showed that MNs firings become phase-locked to input oscillations when the input frequency approximates the neurons firing rate or its harmonics. We refer to this frequency-dependent synchronization as entrainment. Importantly, this subpool-specific effect was masked when MN activity was analysed at the whole-pool level. Because entrained MNs effectively sample the input at their firing instants, we developed a MN-firing locked method that uses individual MN firings as endogenous triggering events for peristimulus frequencygrams across the pool. In simulations, this method revealed entrainment-driven firing rate modulations across MN subpools. We then applied this MN-firing locked method to MNs decomposed from high-density surface electromyography recordings obtained during isometric contractions in healthy individuals. We found that faster-firing MNs exhibited larger transient firing rate increases, time-locked to slower MN activity. Furthermore, these modulations correlated with common input in the alpha and beta bands implicating high frequency common input as the driving source. Together, these findings demonstrate that MN nonlinearities generate heterogeneous, frequency-dependent dynamics that remain hidden in conventional pool-level analyses.

Spectral similarity is often judged with a single metric such as RMSE, yet this can be misleading: physically different errors can produce similar scores. This is a critical limitation for computational biomechanics, where spectral agreement underpins both model validation and machine-learning loss design. Here, we develop a multi-metric framework for objective spectral biofidelity and test whether it better captures meaningful disagreement across complex frequency-domain responses. We evaluated 12 complementary similarity metrics, including CORA and ISO/TS 18571, using controlled spectral perturbations that mimic common real-world deviations such as resonance shifts, localized spikes, and broadband tilts. We then applied the framework to an SBI-tuned finite-element middle-ear model to assess convergence with training dataset size and robustness to measurement noise across repeated stochastic runs. No single metric performed reliably across all distortion types. Shape-based metrics tracked resonance morphology but could miss vertical scaling, whereas MaxError remained important for narrowband anomalies that smoother metrics underweighted. CORA and ISO 18571 did not consistently outperform simpler metrics. Rank aggregation using Borda count provided a robust consensus across metrics, enabling objective identification of training-data saturation and noise thresholds beyond which similarity rankings became unstable. These results show that spectral biofidelity cannot be reduced to a single norm. A multi-metric consensus provides a clearer and more physically meaningful basis for comparing experimental and simulated spectra, and offers a more defensible foundation for data-fidelity terms in physics-informed and simulation-based machine learning.

A critical aspect of human cognition is the ability to use our knowledge about the laws of physics to make predictions about physical events. Whether this ability is based on abstract processes or is grounded in our body-environment interactions remains an open debate. We used physical reasoning under altered gravity as a model system to show that humans real-time embodied experience modifies their high-level physical reasoning. Specifically, we tested participants in computerised reasoning games, while disrupting their gravitational signalling using Galvanic Vestibular Stimulation (GVS). Participants failed more and had suboptimal strategies under the GVS condition compared to no-GVS in games requiring reasoning about terrestrial gravity. However, the effects of GVS were reduced when the games included reasoning about altered gravity. Our findings demonstrate how the physical experience of the body shifts high-level cognitive skill as reasoning, suggesting that humans mental representation of the world is grounded in adaptable physical mechanisms.

A widely used framework for studying the computational mechanisms of decision making is the Drift Diffusion Model (DDM). To account for the presence of both fast and slow errors in empirical data, the DDM incorporates across-trial variability in parameters such as the drift rate and the starting point. Although these variability parameters enable the model to reproduce both fast and slow errors, they rely on the assumption that over trials each parameter is independently sampled. As a result, the DDM effectively predicts that errors-- whether fast or slow--occur randomly over time. However, in empirical data this assumption is violated, as error responses are often temporally clustered. To address this limitation, we introduce the autocorrelated DDM, in which trial-to-trial fluctuations in drift rate, starting point, and boundary evolve according to first-order autoregressive (AR1) processes. Using simulations, we demonstrate that, unlike the across-trial variability DDM, the autocorrelated DDM naturally accounts for temporal clustering of errors. We further show that model parameters can be reliably recovered using Amortized Bayesian Inference, even with as few as 500 trials. Finally, fits to empirical data indicate that the autocorrelated DDM provides the best account of error clustering, highlighting that computational parameters fluctuate over time, despite typically being estimated as fixed across trials.