Intracranial EEG for Neuronal Oscillatory Contingency during Cognitive Tasks

University Of Pennsylvania, Philadelphia PA

Investigators

Linked publications & trials

Abstract

DESCRIPTION (provided by applicant): Brain computer interfaces (BCI) involving motor and sensory systems have been successful in restoring function and in improving the lives of patients with neurological diseases. These devices, ranging from deep brain stimulators for Parkinson's disease to more involved devices capable of recapitulating motor function, are becoming increasingly available with recent technological advances and with a greater understanding of underlying neural circuitry. A major area not addressed by these devices is that of higher cognitive functions, such as memory and executive planning. Here, we propose to use intracranial electrocorticographic signals, captured from subdural and depth electrodes implanted in patients with pharmacologically intractable epilepsy, to extend the role of BCIs to the cognitive domain. The central hypothesis of this proposal is that the patterns of neuronal activity that underlie memory storage and recall can be used to adjust stimulus presentation in a cognitive task to augment learning. We propose to develop a BCI that measures and analyzes intracranial neural activity in real time as patients engage in a free recall task, a standard method of measuring one's ability to encode and retrieve episodic memories. Using machine-learning algorithms, we will identify the precise spatiotemporal patterns of electrophysiological brain activity that lead to optimal memory formation. We will compute these patterns separately for each patient. Our system will thus adapt to each individual's brain activity. To close the functional loop between the patient's brain and our system, we will take advantage of these naturally occurring optimal spatiotemporal patterns for memory formation and trigger stimulus presentation on their presence. Animal studies have shown that by presenting stimuli contingent on the oscillatory state of the hippocampus, learning rates can be improved. We hope to demonstrate that conditioning stimulus presentation on the presence of these optimal spatiotemporal patterns of neural activity will improve both memory storage and recall in our patients as well. Such a demonstration will establish a causative, rather than correlational, relationship between these patterns of activity and memory encoding. Furthermore, our research will extend the domain of BCIs to the realm of cognition and memory, and lay the groundwork for a range of BCI systems that can enhance a wide variety of human cognitive functions. PUBLIC HEALTH RELEVANCE: Brain computer interfaces (BCIs) are machines that directly interface with the brain by measuring and analyzing neural activity in real time. Recent BCIs have played an integral role in rehabilitating patients suffering from Parkinson's disease, severe depression, and epilepsy. One critical area not addressed by BCIs to date is that of higher cognitive functions, including memory -- one of the major deficits in a number of neural diseases, including Alzheimer's disease. Here we propose a novel BCI for improving memory performance both in memory-impaired and cognitively normal individuals.