Artificial neural networks can faithfully recapitulate realistic neuronal behavior with dramatically accelerated simulation runtimes, empowering accessible large-scale realistic network simulations.

Artificial neural networks could pave the way for efficiently simulating large-scale models of neuronal networks in the nervous system.

Single units in a deep convolutional neural network trained for image classification develop shape selectivity that is similar to that found in the primate visual cortex.

MotorNet is a Python toolbox for training artificial neural networks to control arbitrarily complex, differentiable, and biomechanically realistic musculo-skeletal effectors on user-defined sensorimotor tasks.

Six- to twelve-month old infants, who have little linguistic or object experience, classify objects by relying on a invariant representation of global shape known as the shape skeleton.

Computational modeling predicts that sleep replay plays a protective role against catastrophic forgetting by revealing synaptic mechanisms allowing overlapping populations of neurons to store multiple interfering memories.

The anterior cingulate cortex intermixes a precise spatial map with a cognitive map of the task, and spontaneously recalls multimodal information about unrealized choice outcomes during pauses in behavior after reinforcements.

Delivering specific patterns of electrical activity to the median nerve of the arm triggers reliable sensations of texture, suggesting that it may ultimately be possible to restore complex tactile information to users of prosthetic limbs.

For brain imaging to be useful despite its limitations in measuring neural activity, the neural code must be smooth both in a traditional sense and functionally.

A novel trans-humeral prosthesis control method based on artificial intelligence, and the proof of its effectiveness in virtual reality as well as on a robotic platform.