The hippocampus has long been hailed as a key player in statistical learning, a process that enables us to recognize patterns in our environment without explicit instruction or reward. However, the precise role of the hippocampus in this process has been a subject of debate. A new preprint study has shed light on this debate, suggesting that the hippocampus may indeed be essential for statistical learning, but not in the way we previously thought. The study, led by Athena Akrami, group leader at the Sainsbury Wellcome Centre at University College London, has revealed that the hippocampus can track the frequency of auditory events, acting as a 'general-purpose statistical learning machine'. This finding has significant implications for our understanding of how the hippocampus supports both episodic memory and statistical learning, and raises a deeper question about the nature of statistical learning itself. The study involved both human and mouse subjects, who were trained to respond to a particular sound by pressing a key or licking a waterspout. As they performed this task, they also heard an irrelevant four-note sequence at random times. After repeating this task 100 times, both humans and rodents showed strong pupil dilation, a sign of surprise, whenever the sequence of notes changed slightly. This indicated that they had passively learned the original musical motif and abstract rules about its structure. The study found that neuronal populations in the hippocampus encoded not only the original and altered tone sequences but also how frequently each occurred. However, pharmacologically or optogenetically shutting down hippocampal neurons in the mice prevented them from passively learning the auditory pattern and making generalizations about how often it played. This suggests that the hippocampus is indeed involved in statistical learning, but not in the way we previously thought. The study also revealed that the mice were able to learn abstract rules about tone sequences and generalize them to new sequences. If a mouse first learned a sequence of tones that consistently increased in pitch, such as the notes A, B, C, D, then other rising sequences (C, D, E, F, for example) were less surprising than non-rising ones (A, B, B, A). This finding has significant implications for our understanding of how the hippocampus supports both episodic memory and statistical learning. The study raises a deeper question about the nature of statistical learning itself. It suggests that the hippocampus may be involved in both sequence learning and statistical learning, but the precise nature of this involvement is still unclear. One explanation is that two distinct pathways involving the hippocampus support these different computations. The monosynaptic pathway, which encodes and updates general knowledge by averaging over time, goes from the entorhinal cortex directly to CA1. The trisynaptic pathway, which encodes novelty and specific information, goes from the entorhinal cortex to the dentate gyrus, to CA3 and then finally to CA1. The study also highlights the importance of the dorsal CA1 region of the hippocampus in statistical learning. When dCA1 activity was suppressed, mice appeared to be unable to passively learn new abstract rules or update those rules once the statistics of the environment changed. This finding has significant implications for our understanding of how the hippocampus supports both episodic memory and statistical learning. In conclusion, the study has shed light on the precise role of the hippocampus in statistical learning, but it has also raised a deeper question about the nature of statistical learning itself. The findings have significant implications for our understanding of how the hippocampus supports both episodic memory and statistical learning, and suggest that the hippocampus may be involved in both sequence learning and statistical learning. However, the precise nature of this involvement is still unclear and requires further investigation.
