Aitor Morales-Gregorio

@aitormg.bsky.social

1388 Followers 1236 Following

Computational neuroscientist ๐Ÿง ๐Ÿ‡ช๐Ÿ‡บ @ Charles Uni (previously @fz-juelich.de) | Catching up on graphic design | Electrical engineer vibesโšก | he/him ๐Ÿณ๏ธโ€๐ŸŒˆ Personal website: https://www.aitormg.eu

Teaser of poster: PS05-09AM-679 Circuit supporting cortical ripples in the macaque visual cortex K. Studeniฤovรก, A.Morales-Gregorio, K. Korvasovรก For the 2026 FENS Forum in Barcelona. Showing a cortical ripple, different neuron types (waveforms) and a monkey
Poster banner from cosyne 2026
JND test result of 0.0013
Neurons around a recording ephys electrode
Sara Solla giving the Valentin Braitenberg award lecture at the Bernstein 2025 conference in Frankfurt
Brain networks from many animals have nonrandom structure. Including high clustering, short path lengths, heavy tailed weight distributions, and heavy-tailed degree distributions.
Comparing non-square matrices of brain activity with SAS, cosine similarity, and the Frobenius norm
Comparing nonsquare matrices of brain activity with SAS. (a) Schematic diagram of the neuroscience experiment. (b) Single-trial brain activity at identical recording electrodes (64 electrodes, vertical) for each type as example (trial types distinguished by color). Larger values of matrix entries indicate higher MUAe activity (shading in color bar, arbitrary units). (c) Histograms of SAS between all ( ๐‘›=120) individual trials. Filled distributions [color code as in (b)] indicate self-similarities, nonfilled ones indicate cross-similarities. Legends show effect sizes ๐œƒ for the comparison between self- and cross-similarities.
Self- and cross-similarity of different network models. [(a1)โ€“(f1)] Single instances of the different network models. Colored matrix elements indicate a connection between nodes. For the DCM model, connections with a multiplicity higher than one are shown with the same color intensity as single connections. [(a2)โ€“(f2)] Histograms of SAS between instances of the network models ( ๐‘›=100, all pairs compared). Filled distributions indicates self-similarities, nonfilled ones indicate cross-similarities. Legends show effect sizes ๐œƒ for the comparison between self- and cross-similarities.
Comparison of SAS with standard measures. [(a1)โ€“(e1)] Single instances of the different random matrix classes. For visibility, negative values are shown as white. [(a2)โ€“(e2)] Histograms of SAS, cosine similarity, and the Frobenius norm between instances of the random matrix classes ( ๐‘›=100, all pairs compared). Filled distributions indicate self-similarities (self-distances), nonfilled ones indicate cross-similarities (cross-distances). Legends show effect sizes ๐œƒ for the comparison between distributions.
Singular value decomposition and singular angles. Schematic representation of the transformations applied on the basis vectors e (yellow) by the components of SVD for two 2ร—2 matrices Ma (red), and Mb (blue). Small graphs next to the green arrows illustrate the isolated action of the corresponding transformations VT, ฮฃ, and U on e. Below the singular angles of the first [ฮฑj,+(V) and ฮฑj,+(U)] and second [ฮฑj,โˆ’(V) and ฮฑj,โˆ’(U)] kind are shown as the angles between the column vectors of Va and Vb, and Ua and Ub, respectively.
Geometric interpretation of Singular Angle Similarity (SAS). Matrices Ma (red) and Mb (blue). Their eigenvectors scaled by the square root of their eigenvalues span the main axes of ellipsoids. These square matrices capture the correlation structure of along the horizontal and vertical axis, respectively (double-headed colored arrows). SAS compares the angles between the corresponding ellipsoids (dashed colored arrows).
Aligning slices of a banana (or a brain image) makes the reconstructed 3D volume appear straight instead of curved. A reference volume or some other nonlinear metrics are needed to avoid this. Source: https://doi.org/10.1016/j.media.2017.03.004
Fig3 from Georgopoulos 1986.
Figure 7 from Ercsey-Ravasz et al 2013 Global and Local Communication Efficiency (A and B) Effects of graph density via sequentially deleting weak (blue, green) and strong (black, red) links. Data comparison with (A) EDR (Exponential Distance Rule) model, dashed lines and (B) CDR (Constant Distance Rule) model, dashed lines.
Excerpt from Figure 2 from Michaels et al 2024, showing a very detailed and high quality picture of a macaque using the kinarm setup.
Figure 2 from Kalle Kossio et al 2021 Drifting assemblies in spiking neural networks. (A) Schematics emphasizing strong synaptic coupling. While an assembly drifts freely (blue-colored assembly neurons) within the interior neurons, its input and output neurons (green and orange) follow it by adapting their synaptic weights. (B) Weights between interior neurons (blue weight matrix), from input and output neurons to the interior neurons (green and orange vertical weight matrices) and from the interior to the input and output neurons (green and orange horizontal weight matrices). Input (output) weights of neuron i are displayed as the ith row (column). Only weights of the four periphery neurons initially (and thus for all times) attached to assembly 1 are shown for clarity. Left column: Network initialization with three assemblies. Center column, after 27 min: Noisy autonomous spiking activity has already driven several interior neurons to attach to a new assembly (blue weight matrix, horizontal and vertical โ€œlinesโ€ indicating the changed input and output preference). Right column, after 30 h: The assemblies have drifted away, and the weight matrix is completely remodeled. (C) Like B but with neurons reordered according to assemblies that they belong to, using a clustering algorithm. The assemblies remain intact and the periphery neurons stay strongly coupled to assembly 1. (D, Upper) Spike trains of the input (green) and output (orange) neurons of assembly 1, of 12 neurons from each of the ensembles that initially form assembly 1 (5 to 16), 2 (17 to 28) and 3 (29 to 40) and of 4 inhibitory neurons (black). (D, Lower) Membrane potential of the first interior neuron fluctuates irregularly. Spikes are marked by vertical lines above threshold ; reset is to V0.
Figure 4 from Trรคgenap et al 2023 (A) Schematic illustration of the feedforward-recurrent alignment hypothesis. (B) Top: Trial-to-trial correlation for random input (green) and input maximally aligned to the recurrent network (blue). Bottom: Trial-to-trial correlation as a function of feedforward-recurrent alignment. (C) As (B), but for intra-trial stability. Top shows activity of a network unit. (D) As (B), but for dimensionality. Top shows variance explained by leading PCs of evoked activity (see Fig. 2B,C for comparison to experiment). Input is a 10-dimensional Gaussian distribution. (E) Model of spontaneous activity (schematic, bottom): A broad, high-dimensional distribution of inputs aligned with the recurrent network (see Methods), producing a lower-dimensional output (top). (F) Alignment of evoked to spontaneous activity. (G) Model prediction: trials for which the steady-state evoked response is more aligned to spontaneous activity also show higher intra-trial stability. N=300 samples drawn for the input distribution used in (D) with optimal alignment and temporally varying noise (H) Outline of grating evoked activity 1 s (magenta) and 3 s (cyan) after stimulus onset for a response pattern more weakly aligned (left) and for one more strongly aligned (right) to spontaneous activity 0.5 s after stimulus onset. (I) As (G) but for experimental data. (J) Correlation for the data shown in (I) (large blue circle) and for all other animals in the oldest age group (grey circles; 13 experiments from N=11 animals at EO+4 to +7). Error bars show Mean ยฑ SEM. respectively). (K) Contours show outlines of grating evoked patterns from different trials for a stimulus for which the average pattern is less (left) or more (right) aligned to spontaneous activity. (L) As in (J), but quantifying the relationship illustrated in (K). (M) Evoked dimensionality co-varies with the alignment of the first evoked PC to spontaneous activity. N=11 animals at EO+4 to +7
Figure 5 from Jordan et al 2021 Cartesian genetic programming evolves diverse correlation-driven learning rules. (A) Network sketch. Multiple inputs project to a single output neuron. The current synaptic weight wj and the eligibility trace Ecj are provided to the plasticity rule that determines the weight update. (B) Membrane potential u of the output neuron over the course of learning using Equation 17. Gray boxes indicate presentation of the frozen-noise pattern. (C) Fitness (Equation 13) of the best individual per generation as a function of the generation index for multiple runs of the evolutionary algorithm with different initial conditions. Blue and red curves correspond to the two representative plasticity rules selected for detailed analysis. Blue and red markers represent fitness of the two representative rules and the orange marker the fitness of the homeostatic STDP rule (Equation 17; Masquelier, 2018), respectively, on 20 validation tasks not used during the evolutionary search. Error bars indicate one standard deviation over tasks. (D, E): Learning rules evolved by two runs of CGP (D: LR1, Equation 19; E: LR2, Equation 20). (F): Homeostatic STDP rule Equation 17 suggested by Masquelier, 2018. Top panels: STDP kernels ฮ”wj as a function of spike timing differences ฮ”tj for three different weights wj. Bottom panels: homeostatic mechanisms for those weights. The colors are specific to the respective learning rules (blue for LR1, red for LR2), with different shades representing the different weights wj. The learning rate is ฮท=0.01.
Figure 7 from Zhang et al 2023 Schematic illustration of how increased hyperbolic radius affects neural representation of space. As radius grows over time, exponential distribution of place field sizes shifts toward smaller place fields or larger depth in a discrete tree structure, to improve spatial localization.
Part of Figure 2 from Sebastian et al 2023 c, Ripples recorded at the SP are represented in a high-dimensional waveform space (127D) determined by the sampling rate (2,500โ€‰Hz) and the window size (ยฑ25โ€‰ms). In this representation, each SWR is a point in the cloud made by all detected events. d, Topology of the SWR cloud was examined with persistent homology, which identifies holes and cavities in the data by using simplicial complexes. These are made of simplices (for example, dots, lines, triangles and tetrahedrons) that connect data points (SWRs in this case) in a given radius. Persistent homology looks for their persistence as the radius around each point varies in the high-dimensional space. The different homology groups (H) are defined from the number of cuts that separate simplices in pieces. e, Barcodes for the three homology groups (Ho, H1 and H2) show persistence of only one continuous component in the experimental SWR data cloud (data from a random subset of nโ€‰=โ€‰3,500 events).
Figure 1 from Schaub et al Network dynamics and eigenvalue spectra of two LIF networks. One network with uniform synaptic connections (left), and one with 20 groups of clustered excitatory connections (right). A Visualization of the network topologies (top) and exemplars of raster plots (bottom). The dynamics of the clustered network exhibit the banded structure associated with slow-switching group activity. The magnitude of this switching can be characterized statistically a posteriori from the data through the spike-rate variability metric Sฬ‚, defined in Materials and Methods Eq (18), as discussed in the text. In this case, the unclustered network has Sฬ‚=0.035 while the clustered network has a much larger value Sฬ‚=8.23. B Eigenvalue spectra of the network weight matrices W. The weighted connectivity matrix of the clustered network exhibits a clear eigengap ฮ”ฮป separating the 19 eigenvalues with largest real parts from the cloud of eigenvalues in the bulk. There is no such eigengap for the unclustered network. As indicated by the two arrows, both matrices have a pair of complex conjugate eigenvalues associated with the (damped) global activation modes of the networks characteristic of balanced networks.
The fronto-parietal network of neurons exhibits strong beta and low frequency synchronisation during planned and executed grasping movements. Summary figure kindly provided by Benjamin Dann :)
Figure 1 from Gast et al Heterogeneity linearizes neural population dynamics. (Aโ€“D) Two-dimensional (2D) bifurcation diagrams are depicted for different cell types. Regions colored in gray and green depict bistable and oscillatory regimes, respectively. The black and orange crosses depict approximate bifurcation points, estimated from the dynamics of simulated SNNs with a Lorentzian and a Gaussian distribution of the spiking thresholds, respectively. Spiking neural network dynamics were obtained from simulating networks of = 1,000 neurons connected by sparse, random couplings (coupling probability of 20%) The y-axis on the Left (Right) depicts the width of the Lorentzian (Gaussian) distribution used to generate the SNNs that result on the bifurcation points shown by the black (orange) crosses. (A) Excitatory regular-spiking neurons with low spike-frequency adaptation ( pA). (B) Excitatory regular-spiking neurons with high spike-frequency adaptation ( pA). (C) Inhibitory fast-spiking neurons. (D) Inhibitory low-threshold-spiking neurons.
Figure 4 from Froudist-Walsh et al An inverse relationship between cortical myelin and receptor density. a, Cortical T1w/T2w ratio, a proposed marker for myelin content, is strongly negatively correlated with the principal receptor gradient across 109 cortical areas (r(107)โ€‰=โ€‰โˆ’0.72 (range: โˆ’0.80 to โˆ’0.62), Pโ€‰=โ€‰0.02, corrected for spatial autocorrelation and multiple comparisons (Bonferroni)). b, Receptor-, cyto-architecture and myelo-architecture of the macaque primary visual cortex (V1). Changes in Gray Level Index, which represents a measure of the volume fraction of cell bodies, in myelin density and in receptor concentration (in fmol mgโˆ’1 of protein) throughout the cortical depth are provided by the profile curve overlaid onto each section. Note that the scale has been optimized for each profile to provide the best visualization of changes in receptor densities throughout the cortical ribbon. Roman and Arabic numerals indicate cyto-architectonic and myelo-architectonic layers, respectively. Positions of cyto-architectonic layers were transferred to the neighboring receptor images.
Figure 2 from Joglekar et al Signal Propagation in a Large-Scale Network with Recurrent Excitatory Loops Leads to Either Strong Attenuation or Instability (A) Network model of the 29 cortical areas with the strongest inter-areal projection strengths (FLN values >0.005). Input is injected to the excitatory population in area V1 at the lowest position in the cortical hierarchy, and the peak firing rate in area 24c at the top of the hierarchy is recorded. (B) The response of the excitatory populations in areas V1 and 24c to a current pulse of 250 ms to V1, using two close values of the excitatory global coupling parameter ๐œ‡๐ธโข๐ธ. Response (left) in 24c shows an attenuation of four orders of magnitude when ๐œ‡๐ธโข๐ธ=34. With ๐œ‡๐ธโข๐ธ=36 (right), excitatory firing rates in V1 and 24c exponentially grow, and the system becomes unstable.
Figure 1 from Kurth et al Strong scaling of a cortical microcircuit model on a conventional compute node. (a) Sketch of the microcircuit model with about 80โ€‰000 neurons and 300 million synapses organized into four layers of excitatory (blue) and inhibitory (red) populations of neurons. (b) Strong scaling for two placing schemes. Top graph shows realtime factor (RTF) over total number of threads; dashed horizontal line indicates realtime; black solid line indicates linear scaling. The sequential scheme (blue) minimizes distance of threads on hardware, the distant scheme (brown) maximizes it both using 1 MPI process per node; pale coloured line segments connect data points, dashed vertical lines indicate number of cores per processor (64) and node (128). On node hybrid parallelization by 2 MPI processes (1 per socket) for full nodes for comparison (green). Bottom graphs show fractions of wall-clock time consumed by different stages of the simulation cycle; update: integrates state of neurons, deliver: distributes spike events to target neurons, communicate: transfers spikes between MPI processes (for shared and distributed memory setups), other: not accounted for by timers. (c) Top three graphs: power measurements of a compute node during 100 s of model time in three configurations. The measurements are aligned to the start of the simulation phase starting at t = 0. Bottom graph: cumulative energy consumption of the simulations.
Figure 2 from Deistler et al Bayesian inference reveals wide range of energy consumption. (A) Inferring the posterior distribution by combining a rejection classifier and a deep neural density estimator. First, a classifier (trained on an initial set of simulations) predicts which circuit parameters sampled from the prior produce valid simulation outputs. We then proceed by sampling from the part of the parameter space that is accepted by the classifier, i.e., the restricted prior. All valid data (green) are used to train a deep neural density estimator, and all invalid data are discarded (red). Once this estimator is trained, it can be evaluated on experimental data to return the posterior distribution over model parameters. (B) Experimental data recorded from the pyloric network. Arrows indicate a subset of the physiologically relevant features, namely, the cycle period (arrow 1), phase delays (arrow 2), phase gaps (arrow 3), and burst durations (arrow 4). (C) Simulation output from a parameter set sampled from the prior distribution. The traces are AB/PD (Top), LP (Middle), and PY (Bottom). (D) Subset of the marginals and pairwise marginals of the 31-dimensional restricted prior, i.e., the subspace of parameters for which the model produces bursting activity. All maximal conductances are given in mS/. (E) Subset of the marginals and pairwise marginals of the posterior distribution, i.e., the subspace of parameters for which the model matches experimental data shown in B. (F) Sample from the restricted prior producing bursting activity but not matching experimental data. (G) Sample from the posterior distribution closely matching features of the experimental data. (H) Histograms over energy consumed by each neuron (blue, orange, and green) as well as by the entire circuit (black), divided by the duration of the simulation. Trace with lowest energy consumes nine times less energy than trace with highest energy. (I) Same as in H but for energy per spike.
Figure 1 from Zajzon et al Schematic overview of the sequential setup and input stimuli. Networks are composed of four modules with identical internal structure, with random (A) or topographically structured (B) feed-forward projections. Structured stimuli drive specific, randomly selected sub-populations in M0. For stimulus S1, the topographic projections (B, orange arrows) between the modules are represented explicitly in addition to the corresponding stimulus-specific sub-populations (orange ellipses), whereas for S2 only the sub-populations are depicted (blue ellipses). The black feed-forward arrows depict the remaining sparse random connections from neurons that are not part of any stimulus-specific cluster. (C) Illustrative example of the input encoding scheme: a symbolic input sequence of length T (3 in this example), containing |S| different, randomly ordered stimuli (S = {S1, S2}), is encoded into a binary matrix of dimensions |S| ร— T. Each stimulus is then converted into a set of 800 Poissonian spike trains of fixed duration (200 ms) and rate ฮฝstim and delivered to a subset of ฯตNE excitatory and ฯตNI inhibitory neurons.
Figure 1 from Lehr et al. Operations on a circular manifold. See the preprint for a detailed description.
Figure 1A from Barabasi et al Experimental setup; freely swimming fish are monitored with a camera while left- or rightward moving gratings are presented. Heading angle (ฮ”ฮฝ) and fish position (ฮ”x, ฮ”y) are extracted to lock visual stimuli to the animalโ€™s body axis.
Figure 1A from Wybo et al For any set of locations on a given morphology (left, here an L2/3 pyramidal cell [Branco and Hรคusser, 2010]), a reduced compartmental model can be derived (middle), with an associated schematic representation (right).
Figure 1 from van Meegen and van Albada 2021: Illustration of the theory. (a) We consider populations of randomly connected neurons ( ๐›ผ, ๐›ฝ) that communicate via spike trains ๐‘ฅ๐›ผ๐‘–โก(๐‘ก). The neurons of population ๐›ฝ are connected to those of population ๐›ผ with connection probability ๐‘๐›ผโข๐›ฝ. (b) The theory reduces a population to a single neuron driven by an effective stochastic input ๐œ‚๐›ผ. The first- and second-order statistics ๐œ‡๐›ผ๐œ‚ and ๐ถ๐›ผ๐œ‚ of ๐œ‚๐›ผ depend self-consistently on the output statistics, ๐œˆ๐›ผ and ๐ถ๐›ผ๐‘ฅ. (c) From the stationary spike train autocorrelation function ๐ถ๐‘ฅโก(๐œ)=๐œˆโข๐›ฟโก(๐œ)+ห†๐ถ๐‘ฅโก(๐œ), we obtain the correlation time ๐œ๐‘, the asymptotic decay ๐œโˆž, and the variability of the rate across neurons, ๐œŽ2๐œˆ. (d) Instead of the stationary autocorrelation function we sometimes consider the power spectrum ๐‘†๐‘ฅโก(๐‘“), which saturates at the firing rate, ๐‘†๐‘ฅโก(๐‘“)๐‘“โ†’โˆžโˆ’โ†’๐œˆ, and, for a renewal process, has the zero-frequency limit ๐‘†๐‘ฅโก(๐‘“)๐‘“โ†’0โˆ’โ†’๐œˆโขCV2. Throughout, we consider the populationโ€“averaged single-unit statistics (black curve) instead of the statistics of the population-averaged activity (gray curve).
Figure 1 from Kuล›mierz et al: (Top) Visualizations of neural networks with Gaussian (cyan) and Cauchy (orange) distribution of weights. Thickness and color saturation of edges correspond to the (nonlinearly transformed) strengths of the connections. (Middle) Probability density functions (left) and cumulative distribution functions (right) of Cauchy and Gaussian random variables. The Cauchy distribution features much thicker tails than the Gaussian distribution. (Bottom) Sample realizations of the Poisson critical branching process with the duration ๐‘‡ =11, size ๐‘† =18 (left), and ๐‘‡ =35, ๐‘† =367 (right). The initial seeds are marked with the green color. In this Letter, we show that activity of a fully connected Cauchy (but not Gaussian) network around the critical point can be mapped to the critical branching process.
Graphical abstract from Gutzen et al. For two different datasets the different steps of the wave detection process are shown, starting from raw data and ending with spatiotemporal waves.
Part of Figure 2 from Cusinato et al. C) Broadband timescales projected onto the inflated surface template, with brighter colors indicating higher values. The values in sleep change drastically from wake. Note the different scales of the colormaps. D) Correlation between broadband timescales and z-scored T1w/T2w, for each sleep stage. The T1w/T2w is a measure of anatomical cortical hierarchy, and higher values indicate more sensory regions. The correlation is computed with Spearmanโ€™s correlation coefficient and p-values are corrected for spatial autocorrelation via permutations with the โ€œvasaโ€ method. Shaded areas represent the 95% prediction interval of the regression. Note the different scales of the y-axes. s.e.m. = standard error of the mean. The striking result is that while there is a strong correlation of timescales and hierarchy in Awake and NREM sleep, it is no longer there in REM sleep.
Figure 3 from Dahmen et al: Dispersion of correlations measures stability of network dynamics. Connection strength in a sparse random network increases from top to bottom (vertical axis). (A) Connections (arrows) between a pair of observed neurons (black dots); indirect connections contribute to correlations via intermediate neurons (gray dots). (B) Bulk eigenvalues (colored dots) of the connectivity matrix in the complex plane [critical line at 1]. (C) Distributions of covariances. Enlargements are shown in Insets. (D) SD of distribution of covariances (black curve; colored symbols correspond to distributions shown in C).
Figure 8 from Naumann et al: Schematic of the dynamic blind source separation task, context space, and the modulated feedforward network. Information flow is indicated by black arrows, and the flow of the error during training with backpropagation through time (BPTT) is shown in yellow.
Macaque monkey with one eye open and the other eye closed. On top there is a string of dots representing the neural acivity in the visual cortex and how it changes between the eyes open/closed conditions.
Neural manifolds change with eye closure. Macaques were sitting in a dark room (shown on the top left) and their brain activity (MUAe) was reocorded from ~900 electrodes in V1 (bottom left for time changing activity and bottom right for the schematic positions of the arrays in the cortex). The activity showed to main patterns (visible as different clusters or manifolds in the PCA in the top right), one for the eyes-open and the other for the eyes-closed.
Plot of single neuron waveforms extracted from high-pass filtered extracellular ephys recordings from one electrode. X axis shows time (centered around the spike times) in sampling steps (40 points at 30kHz). Y axis shows the high-passed filtered signal, i.e. the waveforms.
Simulation of cell density changes over time. The number of cells in each iteration is multiplied by a random positive number around 1, representing the net cell divisions (>1) and deaths (<1) in each simulated patch. The distribution is initialized with a normal distribution of cell densities, which quickly evolves into a heavy-tailed lognormal distribution.
Schematic animation of the activity of three neurons, shown as changing colors for those neurons on the left side, and as single neuron activity plots on the right. The activity of single neurons seems rather nonsensical (some sine waves, and combinations thereof). When the activity is plotted together in the state-space (neurons against each other) they trace out a trefoil knot, showing that neurons can work together to encode complex things.
Electrophysiology electrode surrounded by pyramidal neurons. The neuron waveforms are shown on top of each neuron, nearby neurons have larger waveforms and faraway neurons have very tiny waveforms. Pyramidal neurons were modified from scidraw.io, Claudi, F. (2020). pyramidal neuron. Zenodo. https://doi.org/10.5281/zenodo.3925905)