A computational model shows that it's possible to learn and replay extended temporal sequences in a network of spiking neurons with a modular architecture and a biologically realistic learning rule.

Teaching signals from "tutor" brain areas should be adapted to the plasticity mechanisms in "student" areas to achieve efficient learning in two-stage systems such as the vocal control circuit of the songbird.

Recurrent neuronal networks learn to predict movement in a self-supervised way using biologically plausible learning rules.

A biologically plausible learning rule enables recurrent neural networks to model the way in which neural circuits use supervised learning to perform time-dependent computations.

Combined pre- and postsynaptically expressed long-term plasticity of neuronal connections improves sensory discrimination, and enables rapid relearning of previously encountered information.

Discrete classes of cerebellar Purkinje neurons show distinct changes in synaptic and spiking activity during motor learning, with simple spikes playing a shifting role during acquisition, expression, and maintenance of learned responses.

Building on previous work (Aso et al., 2014a; Aso et al., 2014b), cell-type-specific drivers and optogenetics are used to control the local release of dopamine and reveal that distinct learning rules are implemented in parallel memory units.

A biologically plausible learning rule allows recurrent neural networks to learn nontrivial tasks, using only sparse, delayed rewards, and the neural dynamics of trained networks exhibit complex dynamics observed in animal frontal cortices.

Output neurons in the mushroom body of the fruit fly brain encode the positive or negative survival value of stimuli, enabling insects to choose adaptive approach and avoidance behaviors through associative learning.