Investigations of Human Memory using Auditory Stimulation


Rhythmic fluctuations of electrical activity in the brain provide insights into the proposed mechanism by which we encode experiences and then maintain, forget, modify, and retrieve them. Yet there is still much to learn about how neural oscillations relate to memory function. The purpose of this research is to discover the behavioral consequences and related neural activity that arises related to long-term declarative memories in humans when we use sounds to systematically trigger brain activity. In the experiments that follow, I use sounds to first influence ongoing neural processing and then to influence the content of memory. In Chapter 2, I examine the importance of brain oscillations in memory formation. To do this, I used sounds modulated at the theta frequency to show that theta activity during encoding aids in memory formation and retention. Listening to 4-Hz (theta) amplitude modulated pink noise paired with object-location encoding increases theta power in electroencephalography (EEG) measured at the scalp. Further, the degree of theta power increase predicts memory strength at test. There was a significant interaction between theta power during 4-Hz sounds (median split of participants based on theta power: high entrainers and low entrainers) and memory accuracy (4-Hz sounds and control sounds) such that high entrainers had higher memory accuracy for objects paired with 4-Hz sounds than those paired with control sounds. In Experiments 2 and 3, I use a technique called Targeted Memory Reactivation (TMR) in which sounds associated with prior learning are presented during sleep to mimic endogenous reactivation that has been observed in the rodent hippocampus. In Experiment 2, I used TMR in patients with intracranial electrodes implanted in the hippocampus to show that sound cues during slow-wave sleep (SWS) evoke hippocampal activity. Patients had more accurate memory for object-locations associated with sounds that were cued during sleep. Moreover, during TMR in SWS, there was an increase in theta power and sigma power in the hippocampus during sounds associated with previously learned items compared with sounds that were not associated with previous learning. In Experiment 3, I show that TMR on a night with seizure activity disrupts memory processing for the cued items. Patients who experienced focal seizures during TMR experienced more forgetting for the cued items than uncued items. By using sounds to change ongoing oscillations during encoding to improve memory, I show that theta oscillations are causal in successful memory processes. By using sounds to bias the content of memory reactivation during sleep in patients with intracranial electrodes, I show, for the first time in humans, oscillatory changes that take place in the medial temporal lobe when memory is reactivated during sleep. Then I show how the disruption of those processes due to seizure activity impairs memory consolidation. These findings show that sounds can be used as a powerful tool to probe the mechanisms of memory and suggest these methods could be used in future experiments to reveal yet more about memory processes.

Alternate Identifier
Date created
Resource type
Rights statement