AUTHOR=Kempermann Gerd TITLE=What Is Adult Hippocampal Neurogenesis Good for? JOURNAL=Frontiers in Neuroscience VOLUME=16 YEAR=2022 URL=https://www.frontiersin.org/journals/neuroscience/articles/10.3389/fnins.2022.852680 DOI=10.3389/fnins.2022.852680 ISSN=1662-453X ABSTRACT=

Adult hippocampal neurogenesis is a unique and exceptional process in the mammalian brain that in a lifelong and activity-dependent way generates new excitatory principal neurons. A comprehensive view on their function in greater contexts has now emerged, revealing to which extent the hippocampus (and hence brain and mind) depend on these neurons. Due to a postmitotic period of heightened synaptic plasticity they bias incoming excitation to the dentate gyrus to non-overlapping subnetworks, resulting in pattern separation and the avoidance of catastrophic interference. Temporally, this promotes the flexible integration of novel information into familiar contexts and contributes to episodic memory, which in humans would be critical for autobiographic memory. Together these local effects represent a unique strategy to solve the plasticity-stability dilemma that all learning neuronal networks are facing. Neurogenesis-dependent plasticity also improves memory consolidation. This relates to the surprising involvement of adult neurogenesis in forgetting, which is also hypothesized to be critically relevant for negative plasticity, for example in post-traumatic stress disorder. In addition, adult-born neurons also directly mediate stress-resilience and take part in affective behaviors. Finally, the activity- and experience-dependent plasticity that is contributed by adult neurogenesis is associated with an individualization of the hippocampal circuitry. While a solid and largely consensual understanding of how new neurons contribute to hippocampal function has been reached, an overarching unifying theory that embeds neurogenesis-dependent functionality and effects on connectomics is still missing. More sophisticated multi-electrode electrophysiology, advanced ethologically relevant behavioral tests, and next-generation computational modeling will let us take the next steps.