Research

In this section you can find a brief summary of the most relevant scientific projects to which I have contributed

Local and Global nature of cortical slow waves

Slow Wave Activity is a robust low complexity spontaneous brain regime on which hyperpolarized Down states alternate with depolarized, firing, Up states. These are originated in the layer V of the cortex and propagate along the cortical mantle together with to subcortical regions.

In this work, we created computational models of travelling Ups along cortical populations to explore how they emerged as a balance of local components and long-range inputs. We performed an exhasutive sensitivity analysis of different cellular and synaptic currents with the aim to shed light on the intriguing coexistance of this low-dimensional homogeneous wave along most of the brain with particular motifs in certain cortical regions.

Nonlinear and nonstationary analyisis of neuronal oscillations

While oscillations are traditionally transformed into the Time-Frequency domain using fixed templates, either sinusoidal or with a different shape, most frequently neuronal waves have shape differences between cycles as well as a nonstationary pattern.

During my PhD I used the Noise-Assisted Multivariate Empirical Mode Decomposition (NA-MEMD) algorithm to surpass these limitations and obtain a T-F picture of neuronal oscillations with instantaneous frequency and amplitude. In brief EMD algorithms decompose the signal in a series of oscillatory functions in a data-driven manner, without the use of any template or stationarity assumption. This modes convey a single oscillator at any given time, allowing to use the Hilbert Transform to compute their phase and amplitude.

In this paper Time–frequency analysis of neuronal populations with instantaneous resolution based on noise-assisted multivariate empirical mode decomposition I show how NA-MEMD is an improvement compared to traditional T-F techniques, both in local field and spike oscillations.

Then, in this other paper Toward an Improvement of the Analysis of Neural Coding , I show how using this technique we obtain an enriched picture of the oscillatory dynamics that allows for a single trial classification in a texture discrimination task, as well as deciphering the stimulated electretrode from the evoked activity in a neuronal culture.

Representation of time interval at seconds scale in V1

In this project, I explored how the interval time between visual stimuli impacted the response in neuronal population of V1 of anesthetized rats. To do so, I characterized the temporal structure of the average spiking response of the population using the NA-MEMD algorithm in this paper Multiscale dynamics of interstimulus interval integration in visual cortex. It shows the response to the visual stimulation is higher for certain intervals (3-5s), when compared to shorter of longer times. Moreover, different interval times lead to a unique temporal structure of the response, which created a specific trajectory in a parameter space designed for this porpoise.

Discrete segregation of dorsal striatum

The dorsal striatum of rodents is considered to be the homologous of primate caudate and putamen. While multiple studies support that its lateral (DLS) and medial (DMS) regions are equivalent to these areas, they lack of anatomical borders and have a strong similarity in terms of their cellular components. To understand striatal dynamics and basal ganglia functions, it is essential to clarify the circuitry that supports this behavioral-based segregation.

Here Medium spiny neurons activity reveals the discrete segregation of mouse dorsal striatum, we show that the mouse DS is made of two non-overlapping functional circuits divided by a boundary. To do so we performed in vivo intracellular recordings of Slow Wave Activity (SWA) and compared this low-dimensional brain regime between DLS and DMS, as well as their interaction with multiple cortical regions. This allowed us to demonstrate that they are differently coupled to the rostral and caudal cortical axis during SWA. We did also described different parameters of the active «Up states» that allowed to discriminate between Medium Spiny Neurons belonging to DLS or DMS and discussed their possible physiological substrate. Last, we demonstrated that there is a sharp transition from one patter of SWA to the other, in line with two well differentiated circuits

Other selected scientific projects

Callosal inputs generate side-invariant receptive fields in the barrel cortex In this work we demonstrate the existence of a invariant representation of the middle line in the Row A of the barrel cortex. These neurons had equivalent subthreshold responses to contralateral or ipsilateral stimulation of the whiskers, suggesting a role in the perception of continuity between sides of the body.

Dopamine enhances multisensory responses in the dorsomedial striatum
In this work we explored how dopamine facilitates multisensory integration in the striatum. We show how in a context of responses with different intrinsic delays, it selectively accelerates the visual response in the dorsomedial striatum, allowing a more efficient integration of multisensory stimuli

A complete list of publications can be found here