Understanding mechanisms of semantic memory through intracranial recording and stimulation

University Of California-San Francisco, San Francisco CA

Investigators

Abstract

The color of the sky, a list of fruits, the capital of France—the ability to access facts and other general information at astonishing speed is a remarkable and vital part of being human. Our vast stores of knowledge are known to be spread throughout multiple widespread regions in the brain. Yet, when a person hears a question and retrieves an answer, these disparate areas coordinate and sift through an ocean of details and associations, on-demand, to pull up specific facts. This ability is sometimes impaired with aging, and markedly deteriorates in a host of neurological conditions -- such as semantic dementia, Alzheimer’s disease, and epilepsy. Despite its clear importance, and multiple scientific studies, the actual mechanisms of semantic memory have been an elusive and enduring mystery. Understanding its neural basis could lead to transformational advances in education, ameliorate the effects of aging on semantic memory, and help treat neurological diseases that ravage memory. In addition to these broader impacts, this project will engage in community outreach through public science talks and create opportunities for the promotion of women and underrepresented minorities in STEM careers in neuroscience. This research involves a unique and precious source of data: recordings directly from the brain in awake, behaving humans, that are being performed for clinical (medical, surgical) purposes to treat intractable epilepsy. Using these intracranial recordings, researchers will seek to decipher how neural populations in key cortical hubs process semantic information while participants perform custom-designed tasks of both verbal and non-verbal semantic memory. The researchers will measure neural population responses from high-density electrodes (intracranial electroencephalography – iEEG) and use machine learning to understand and predict how information flows between key regions in the semantic system and medial temporal memory regions in the brain. The project will also directly test whether electrical stimulation of brain tissue during cognitive tasks results in predictable effects, revealing whether these same structures play causal roles in semantic memory processing. Together, this human iEEG research will yield a better understanding of how semantic knowledge retrieval is encoded and supported by local neuronal populations across distributed semantic and memory circuits and broader brain networks. This work may also lead to future therapies, including emerging brain stimulation approaches that may help rescue memory loss in neurological disease. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.