Patrick Verga

Extinction of conditioned fear is better learned and recalled in the morning than in the evening

Sleep helps emotional memories consolidate and may promote generalization of fear extinction memory. We examined whether extinction learning and memory might differ in the morning and evening due, potentially, to circadian and/or sleep-homeostatic factors. Healthy men (N = 109) in 6 groups completed a 2-session protocol. In Session 1, fear conditioning was followed by extinction learning. Partial reinforcement with mild electric shock produced conditioned skin conductance responses (SCRs) to 2 differently colored lamps (CS+), but not a third color (CS-), within the computer image of a room (conditioning context). One CS+ (CS + E) but not the other (CS + U) was immediately extinguished by un-reinforced presentations in a different room (extinction context). Delay durations of 3 h (within AM or PM), 12 h (morning-to-evening or evening-to-morning) or 24 h (morning-to-morning or evening-to-evening) followed. In Session 2, extinction recall and contextual fear renewal were tested. We observed no significant effects of the delay interval on extinction memory but did observe an effect of time-of-day. Fear extinction was significantly better if learned in the morning (p = .002). Collapsing across CS + type, there was smaller morning differential SCR at both extinction recall (p = .003) and fear renewal (p = .005). Morning extinction recall showed better generalization from the CS + E to CS + U with the response to the CS + U significantly larger than to the CS + E only in the evening (p = .028). Thus, extinction is learned faster and its memory is better generalized in the morning. Cortisol and testosterone showed the expected greater salivary levels in the morning when higher testosterone/cortisol ratio also predicted better extinction learning. Circadian factors may promote morning extinction. Alternatively, evening homeostatic sleep pressure may impede extinction and favor recall of conditioned fear.

Modeling reconsolidation in kernel associative memory.

Memory reconsolidation is a central process enabling adaptive memory and the perception of a constantly changing reality. It causes memories to be strengthened, weakened or changed following their recall. A computational model of memory reconsolidation is presented. Unlike Hopfield-type memory models, our model introduces an unbounded number of attractors that are updatable and can process real-valued, large, realistic stimuli. Our model replicates three characteristic effects of the reconsolidation process on human memory: increased association, extinction of fear memories, and the ability to track and follow gradually changing objects. In addition to this behavioral validation, a continuous time version of the reconsolidation model is introduced. This version extends average rate dynamic models of brain circuits exhibiting persistent activity to include adaptivity and an unbounded number of attractors.

Row-less Universal Schema.

Universal schema jointly embeds knowledge bases and textual patterns to reason about entities and relations for automatic knowledge base construction and information extraction. In the past, entity pairs and relations were represented as learned vectors with compatibility determined by a scoring function, limiting generalization to unseen text patterns and entities. Recently, ‘column-less’ versions of Universal Schema have used compositional pattern encoders to generalize to all text patterns. In this work we take the next step and propose a ‘row-less’ model of universal schema, removing explicit entity pair representations. Instead of learning vector representations for each entity pair in our training set, we treat an entity pair as a function of its relation types. In experimental results on the FB15k-237 benchmark we demonstrate that we can match the performance of a comparable model with explicit entity pair representations using a model of attention over relation types. We further demonstrate that the model performs with nearly the same accuracy on entity pairs never seen during training.