10/26/2022 0 Comments Spot process separation studio 4 $389![]() During the encoding of conjunctive representations, hippocampal computations establish a minimal representational overlap between traces of events with partially shared features, in a process called pattern separation 4, 5. Theoretical and empirical work suggest the hippocampus rapidly forms conjunctive representations of arbitrary sets of co-occurring features 2, making the hippocampus critical for episodic memory 3. Learning in the brain operates over qualitatively distinct representations depending on brain system 1. In such problems, simple feature-response learning is insufficient and representations that include multiple features (e.g., leaf shape, color, and season) must be learned. Responses and predictions also depend on the status of other features or context. Such learning problems are challenging because similar conjunctions of features can require different responses or elicit different predictions about future events. The relationship between color and season distinguishes poison ivy from other plants like boxelder, which looks similar but is green in spring. Most North American hikers develop a reflexive aversion to poison ivy, which causes a painful rash, and learn to recognize its compound leaf with three leaflets that is green in summer and red in spring and autumn. Our results establish a novel role for hippocampal pattern separation and conjunctive representation in reinforcement learning. Pattern analyses on functional MRI data show the hippocampus forms conjunctive representations that are dissociable from feature components and that these representations, along with those of cortex, influence striatal prediction errors. Here, we test if the hippocampus forms separable conjunctive representations that enables the learning of response contingencies for stimuli of the form: AB+, B−, AC−, C+. One solution is to assign value to multi-featural conjunctive representations. Feature-based reinforcement learning fails when the values of individual features depend on the other features present. More often, outcomes are linked to multidimensional states (e.g., poison ivy is green in summer but red in spring). Associations can be based on relationships between object features (e.g., the three leaflets of poison ivy leaves) and outcomes (e.g., rash). Animals rely on learned associations to make decisions. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |