Back

Neural signatures for temporal-order memory in the medial posterior parietal cortex

Zuo, S.; Wang, C.; Wang, L.; Jin, Z.; Kusunoki, M.; Kwok, S. C.

2023-08-18 neuroscience
10.1101/2023.08.17.553665 bioRxiv
Show abstract

Episodic memory involves encoding and remembering the order of events experienced over time. Previous work examining the mechanisms of temporal-order memories has focused on the hippocampus and prefrontal cortices but has largely ignored the memory ensembles in the medial posterior parietal cortex (mPPC). Combining in vivo multi-unit electrophysiology and a temporal-order judgment task with cinematic material in the macaque, we find that mPPC neuronal activity reflects temporal context both during encoding and recall. During learning, mPPC neuronal ensembles encode the temporal information as well as contextual information embedded in the videos, whereas at retrieval these neurons fire in synchrony prior to memory-guided decisions. Moreover, the similarity between encoding and retrieval correlates with the animals performance. A control experiment further ruled out eye saccades, fixation, and scan path for their confounding roles in the neural results. Together, these data suggest that neurons in mPPC track the passage of time and contextual changes, thereby orchestrating for successful temporal-order memories at retrieval. SIGNIFICANCERemembering the temporal order of events is a vital function of the brain. This study provides the first evidence for the retrieval of temporal order of episodes at both single neuron and ensemble levels in the primate parietal cortex. The studys key breakthrough is the revelation that ensembles of time perceptive neurons, acting as "temporal context cells", can encode a spectrum of time constants and contexts for the experienced past and subsequently utilize this temporal record to discriminate the chronological order of past experiences. These results enhance theoretical frameworks for how encoded temporal past is maintained and reconstructed during memory retrieval.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.