Developing neurons form synapses at a high rate. Synaptic transmission is very energy-demanding and likely requires ATP production by mitochondria nearby. Mitochondria might be targeted to active synapses in young dendrites, but whether such motility regulation mechanisms exist is unclear. We investigated the relationship between mitochondrial motility and neuronal activity in the primary visual cortex of young mice in vivo and in slice cultures. During the first 2 postnatal weeks, mitochondrial motility decreases while the frequency of neuronal activity increases. Global calcium transients do not affect mitochondrial motility. However, individual synaptic transmission events precede local mitochondrial arrest. Pharmacological stimulation of synaptic vesicle release, but not focal glutamate application alone, stops mitochondria, suggesting that an unidentified factor co-released with glutamate is required for mitochondrial arrest. A computational model of synaptic transmission-mediated mitochondrial arrest shows that the developmental increase in synapse number and transmission frequency can contribute substantially to the age-dependent decrease of mitochondrial motility.

Forgetting is a normal process in healthy brains, and evidence suggests that the mammalian brain forgets more than is required based on limitations of mnemonic capacity. Episodic memories, in particular, are liable to be forgotten over time. Researchers have hypothesized that it may be beneficial for decision making to forget episodic memories over time. Reinforcement learning offers a normative framework in which to test such hypotheses. Here, we show that a reinforcement learning agent that uses an episodic memory cache to find rewards in maze environments can forget a large percentage of older memories without any performance impairments, if they utilize mnemonic representations that contain structural information about space. Moreover, we show that some forgetting can actually provide a benefit in performance compared to agents with unbounded memories. Our analyses of the agents show that forgetting reduces the influence of outdated information and states which are not frequently visited on the policies produced by the episodic control system. These results support the hypothesis that some degree of forgetting can be beneficial for decision making, which can help to explain why the brain forgets more than is required by capacity limitations.

Forgetting is a normal process in healthy brains, and evidence suggests that the mammalian brain forgets more than is required based on limitations of mnemonic capacity. Episodic memories, in particular, are liable to be forgotten over time. Researchers have hypothesized that it may be beneficial for decision making to forget episodic memories over time. Reinforcement learning offers a normative framework in which to test such hypotheses. Here, we show that a reinforcement learning agent that uses an episodic memory cache to find rewards in maze environments can forget a large percentage of older memories without any performance impairments, if they utilize mnemonic representations that contain structural information about space. Moreover, we show that some forgetting can actually provide a benefit in performance compared to agents with unbounded memories. Our analyses of the agents show that forgetting reduces the influence of outdated information and states which are not frequently visited on the policies produced by the episodic control system. These results support the hypothesis that some degree of forgetting can be beneficial for decision making, which can help to explain why the brain forgets more than is required by capacity limitations.