Video Experimental Relacionado
Updated: May 27, 2025

09:45
New Variations for Strategy Set-shifting in the Rat
Published on: January 23, 2017
8.1K
Un circuito striatal oponente para el aprendizaje de refuerzo distributivo
Adam S Lowet1,2,3, Qiao Zheng1,4, Melissa Meng1,2
1Center for Brain Science, Harvard University, Cambridge, MA, USA.
Nature
|February 19, 2025
Resumen
Los modelos de aprendizaje por refuerzo distributivo (RL) recompensan las distribuciones, no solo los medios. Este estudio revela cómo la dopamina y las neuronas estriadas codifican la varianza de recompensa, avanzando en nuestra comprensión de la RL basada en el cerebro.
Área de la Ciencia:
- La neurociencia
- Neurociencia computacional
- Aprendizaje automático
Sus antecedentes:
- Los avances en el aprendizaje automático utilizan el aprendizaje por refuerzo distributivo (RL) al considerar distribuciones de recompensas completas.
- El sistema de dopamina mesolímbico del cerebro de los mamíferos está implicado en la RL, principalmente a través de actualizaciones de valores medios en el estriado.
- Existe un conocimiento limitado sobre cómo las neuronas estriatales codifican los momentos de orden superior de las distribuciones de recompensa.
Objetivo del estudio:
- Investigar la codificación neuronal de los momentos de distribución de la recompensa en el cerebro de los mamíferos.
- Determinar el papel del estriado y la dopamina en la RL distributiva.
- Elucidar la contribución de los diferentes tipos de neuronas estriadas a la codificación de la varianza de recompensa.
Principales métodos:
- Neuropixels de alta densidad graban la actividad striatal en ratones durante una tarea de acondicionamiento clásica.
- Manipulación independiente de la media de recompensa, la varianza y la identidad del estímulo.
- Ablación crónica de entrada de dopamina, imágenes de calcio de dos fotones y optogenética.
Principales resultados:
- Las neuronas estriatales exhiben una codificación abstracta de la varianza de recompensa, desafiando los modelos tradicionales de RL.
- La ablación de la entrada de dopamina interrumpió las representaciones distributivas pero no la codificación del valor medio.
- Las neuronas espinosas medianas D1 y D2 codifican diferencialmente las colas derecha e izquierda de la distribución de la recompensa, respectivamente.
Conclusiones:
- El estriado y el sistema de dopamina mesolímbico apoyan la RL distributiva a través de la codificación de la varianza abstracta.
- La dopamina juega un papel crucial en el mantenimiento de estas representaciones distributivas.
- Un nuevo modelo de función estriatal aprovecha la oposición de las neuronas D1 / D2 para obtener beneficios distributivos de RL.
Videos de Conceptos Relacionados
Neural Circuits
996
Neural circuits and neuronal pools are two of the main structures found in the nervous system. Neural circuits are networks of neurons that work together to carry out a specific task or process. They consist of interconnected neurons and glial cells, which provide structural and metabolic support.
Neuronal pools are collections of nerve cells with similar functions and interact through chemical and electrical signals. These pools include both interneurons (the central neural circuit nodes that...
Neuronal pools are collections of nerve cells with similar functions and interact through chemical and electrical signals. These pools include both interneurons (the central neural circuit nodes that...
996
Diencephalon: Thalamus and Information Relay
1.4K
The thalamus, often called “the gateway to the cerebral cortex,” is vital in processing and directing sensory and motor signals throughout the brain. Almost all inputs destined for the cerebral cortex, except for olfactory signals, are relayed through the thalamus. The thalamus is a sophisticated relay station, channeling information from various brain regions to the cerebral cortex, as well as a filter, prioritizing certain signals over others based on current physiological...
1.4K
Instinctive Drift
177
Instinctive drift refers to the tendency of animals to revert to their innate behaviors despite repeated reinforcement. Breland and Breland demonstrated this concept in an experiment with a raccoon. The raccoon was trained to pick up two coins and place them in a container in exchange for food. Initially, the raccoon learned to associate the coins with food, making them a conditioned stimulus or a substitute for food. However, over time, the raccoon became less willing to put the coins into the...
177
Associative Learning
283
Associative learning is a fundamental concept in behavioral psychology, wherein a connection is established between two stimuli or events, leading to a learned response. This process is critical in understanding how behaviors are acquired and modified. Conditioning, the mechanism through which associations are formed, can be divided into two main types: classical conditioning and operant conditioning, each elucidating different aspects of associative learning.
Classical conditioning, also known...
Classical conditioning, also known...
283
Indirect Motor Pathways
1.4K
The indirect motor or extrapyramidal pathways originate in the brainstem, the lower portion of the brain that connects it to the spinal cord. They consist of several distinct tracts, each with specialized functions. The four main tracts of the indirect motor pathways are the vestibulospinal tract, the reticulospinal tract, the tectospinal tract, and the rubrospinal tract.
The vestibulospinal tract originates in the vestibular nuclei of the brainstem. The vestibular system detects changes in...
The vestibulospinal tract originates in the vestibular nuclei of the brainstem. The vestibular system detects changes in...
1.4K

