Sleep and memory consolidation in temporal lobe epilepsy

Objectives. Work in our lab and others have found that individuals with temporal-lobe epilepsy show Accelerated Long-Term Forgetting, a situation characterized by normal memory at short intervals (15–30 mins, which is typical of clinical assessments) but a rapid rate of forgetting relative to controls over hours and days afterward. These findings suggest disruption to a process involving hippocampal-neocortical interaction enabling ‘consolidation’ of memory traces in the neocortex over time. Indeed, we have reported that functional connectivity between the affected hippocampus and temporal neocortex is positively associated with recall after 72 hours (Audrain & McAndrews, Cortex 2018). Here, we investigated how slow-wave sleep (SWS) might moderate these effects. Methods. We examined associative recognition memory in 19 patients with temporal lobe epilepsy on the Epilepsy Monitoring Unit at the Toronto Western Hospital. Critically, we focused on forgetting over a period of 6 hours to 16 hours as a function of duration of SWS overnight and frequency of frontotemporal interictal epileptic discharges (IEDs) during the same period. We also compared behavioral forgetting rates against age-matched healthy controls. Results. While there was no significant difference between patients and controls in forgetting rates over that interval, most controls showed a facilitative impact whereas there was considerable variability amongst the patient group. In patients, there was a significant relationship between duration of SWS and forgetting (r = 0.60), such that greater SWS was associated with greater forgetting, which is opposite to the typical facilitative effect of SWS on retention in controls. We hypothesized that IEDs might mediate this effect, as they might serve as disruptive signals to consolidation processes during sleep and are more abundant during SWS. However, while IEDs did correlate with forgetting (r = 0.42), this did not prove to be a significant independent mediator. Conclusions. The counter-intuitive effects of SWS on memory retention in individuals with temporal lobe epilepsy found here replicates one previous study. We consider the possibility that hippocampal IEDs, not captured adequately by scalp recordings, may play a significant role and we plan to investigate this in a subsequent study. Funding. Natural Sciences and Engineering Research Council of Canada.