Mere exposure to scenes with statistical contingencies in the visual or haptic modality alone allows participants to decompose scenes into object-like multimodal representations.

Prior expectations, induced by statistical learning, affect sensory processing only for attended stimuli, thereby constraining neurocomputational theories that cast perception as an inferential process integrating prior knowledge and sensory evidence.

Learned expectations allow for a distinct anticipatory distractor representation at the feature but not at the spatial level.

Hippocampal learning in dentate gyrus follows a U-shaped function, with moderate, but not high or low, overlap between representations leading to differentiation of neural patterns.

Neural sensory representations impose an inductive bias over the space of learning tasks, allowing some tasks to be learned by a downstream neuron more sample-efficiently than others.

Electrophysiological activity indicates superior implicit crossmodal learning in infants and a change in learning mode from infancy to adulthood.

When a common error drives parallel implicit and explicit learning systems, increases in the explicit response will suppress implicit adaptation.

Autistic traits are associated with weaker influence of prior expectations in visual perception, which is due to more precise sensory representations.

Building on previous work (Vogt et al. 2014), separate visual pathways have been identified to convey information for brightness and color to the learning-related area of the Drosophila brain.

A complete public catalogue of brain recording data from a single laboratory has revealed new scientific findings, and has also shed light on aspects of the scientific process, such as statistical power and publication bias.