Conditioned olivocerebellar network activity elicits transient spike pauses and timed spike facilitation in the neurons of interposed nuclei that predicts and likely causes conditioned eyelid responses on a trial-by-trial basis.

Pronounced cerebellar activation during unexpected omission of a potentially harmful event suggests that the cerebellum has to be added to the neural network processing prediction errors underlying emotional associative learning.

CA1 and PFC bridge the temporal gap between cue and reward delivery during trace conditioning according to different underlying coding principles and task-related activity is reactivated during awake Sharp-Wave Ripples.

When exposed to prolonged stressful stimulation, rats start to emit long, frequency-unmodulated 44-kHz ultrasonic vocalizations, usually within the 35–50 kHz range, along with, previously described, long and unmodulated 22-kHz ultrasonic vocalizations.

Deconditioning is a safe and efficient new approach to updating traumatic memories, in which fear memory is rewritten to a very low level in a long-lasting way.

Although flies cannot discriminate shock-paired and -unpaired odor 24 hr after single-trial aversive conditioning, they choose to avoid both of them, which is derived from a merged long-term memory.

Stimulus-reward learning sharpens the local representation of the visual space while leaving the overall retinotopic map intact.

A novel high-throughput FPGA-based multi-animal tracking and training system was used to demonstrate trace and operant conditioning in Drosophila larvae.

Large-scale imaging analysis of CA1 reveals that distinct neural networks are involved in trace conditioning versus extinction learning, shedding light on how the hippocampus encodes different types of memory.

A slow and stable memory unit instructs fast and transient units by activating dopaminergic neurons via an excitatory hub interneuron connecting those units during second-order conditioning in Drosophila.