Self-esteem, the value that people ascribe to the self, is represented in the ventromedial prefrontal cortex and dynamically updated when people learn how others value them.

Veterans with PTSD show increased attention to a history of unexpected outcomes during loss learning, both as measured by computational model-derived behavioral parameters and in increased neural signaling in amygdala and insula.

Neural representations are fast-evolving trajectories, and distinct components of these trajectories reappear during retrieval with distinct consequences for learning.

Map-like organisations of relational knowledge can be extracted from the hippocampal-entorhinal system in situations where relationships are non-spatial rather than spatial, discrete rather than continuous, and unavailable to conscious awareness.

While current theories assume that mood is dominated by most recent experiences, computational modeling shows that mood is shaped by first experiences that carry a long-lasting overarching influence on mood.

Decision-makers are able to intentionally control neural excitability to strategically bias sensory evidence accumulation towards the decision bound that maximizes reward within a given ecological context.

The anticipation of rewards turns out to have its own hedonic value, on top of that of the reward itself; a wide range of behavioral and neurophysiological data suggest that this anticipation is boosted by prediction errors.

Paranoia is underwritten by variation in prior beliefs about how the world will change and how to learn from those changes.

A dimensional approach to psychiatry demonstrates the specificity and generalizability of a neurocognitive marker of compulsive behavior and intrusive thought via large-scale online testing.

Chronic alcohol exposure results in a long-lasting, enhanced endocannabinoid signalling at orbitostriatal synapses that is expressed in a projection-, synapse-, and behaviorally computation-specific manner.