Biology subjects

Mimica, B.

Publications and source records attributed to Mimica, B..

4 recordsLinked to original sources

bioRxiv2024

Vocal Call Locator Benchmark (VCL) for localizing rodent vocalizations from multi-channel audio

Understanding the behavioral and neural dynamics of social interactions is a goal of contemporary neuroscience. Many machine learning methods have emerged in recent years to make sense of complex video and neurophysiological data that result from these experiments. Less focus has been placed on understanding how animals process acoustic information, including social vocalizations. A critical step to bridge this gap is determining the senders and receivers of acoustic information in social interactions. While sound source localization (SSL) is a classic problem in signal processing, existing approaches are limited in their ability to localize animal-generated sounds in standard laboratory environments. Advances in deep learning methods for SSL are likely to help address these limitations, however there are currently no publicly available models, datasets, or benchmarks to systematically evaluate SSL algorithms in the domain of bioacoustics. Here, we present the VCL Benchmark: the first large-scale dataset for benchmarking SSL algorithms in rodents. We acquired synchronized video and multi-channel audio recordings of 767,295 sounds with annotated ground truth sources across 9 conditions. The dataset provides benchmarks which evaluate SSL performance on real data, simulated acoustic data, and a mixture of real and simulated data. We intend for this benchmark to facilitate knowledge transfer between the neuroscience and acoustic machine learning communities, which have had limited overlap.

neuroscience↗

bioRxiv2022

Behavioral decomposition reveals rich encoding structure employed across neocortex

AO_SCPLOWBSTRACTC_SCPLOWThe Cortical population code is pervaded by activity patterns evoked by movement, but how such signals relate to the natural actions of unrestrained animals remains largely unknown, particularly in sensory areas. To address this we compared high-density neural recordings across four cortical regions (visual, auditory, somatosensory, motor) in relation to sensory modulation, posture, movement, and ethograms of freely foraging rats. Momentary actions, such as rearing or turning, were represented ubiquitously and could be decoded from all sampled structures. However, more elementary and continuous features, such as pose and movement, followed region-specific organization, with neurons in visual and auditory cortices preferentially encoding mutually distinct head-orienting features in world-referenced coordinates, and somatosensory and motor cortices principally encoding the trunk and head in egocentric coordinates. The tuning properties of synaptically coupled cells also exhibited connection patterns suggestive of different uses of behavioral information within and between regions. Together, our results speak for detailed, multi-level encoding of animal behavior subserving its dynamic employment in local cortical computations.

neuroscience↗

bioRxiv2019

Action representation in the mouse parieto-frontal network

The posterior parietal cortex (PPC), along with anatomically linked frontal areas, form a cortical network which mediates several functions that support goal-directed behavior, including sensorimotor transformations and decision making. In primates, this network also links performed and observed actions via mirror neurons, which fire both when an individual performs an action and when they observe the same action performed by a conspecific. Mirror neurons are thought to be important for social learning and imitation, but it is not known whether mirror-like neurons occur in similar networks in other species that can learn socially, such as rodents. We therefore imaged Ca2+ responses in large neural ensembles in PPC and secondary motor cortex (M2) while mice performed and observed several actions in pellet reaching and wheel running tasks. In all animals, we found spatially overlapping neural ensembles in PPC and M2 that robustly encoded a variety of naturalistic behaviors, and that subsets of cells could stably encode multiple actions. However, neural responses to the same set of observed actions were absent in both brain areas, and across animals. Statistical modeling analyses also showed that performed actions, especially those that were task-specific, outperformed observed actions in predicting neural responses. Overall, these findings show that performed and observed actions do not drive the same cells in the parieto-frontal network in mice, and suggest that sensorimotor mirroring in the mammalian cortex may have evolved more recently, and only in certain species.

neuroscience↗

bioRxiv2018

Efficient cortical coding of 3D posture in freely behaving rats

In order to meet physical and behavioral demands of their environments animals constantly update their body posture, but little is known about the neural signals on which this ability depends. To better understand the role of cortex in coordinating natural pose and movement, we tracked the heads and backs of freely foraging rats in 3D while recording simultaneously from posterior parietal cortex (PPC) and frontal motor cortex (M2), areas critical for spatial movement planning and navigation. Single units in both regions were tuned mainly to postural features of the head, back and neck, and much less so to their movement. Representations of the head and back were organized topographically across PPC and M2, and the tuning peaks of the cells were distributed in an efficient manner, where substantially fewer cells encoded postures that occurred more often. Postural signals in both areas were sufficiently robust to allow reconstruction of ongoing behavior with 90% accuracy. Together, these findings demonstrate that both parietal and frontal motor cortices maintain an efficient, organized representation of 3D posture during unrestrained behavior.

neuroscience↗