158 Building a Data Processing Pipeline for Tissue and Pathway Activation Modeling in Human Amygdala Stimulation for Memory Modulation

The amygdala is a brain region implicated in both memory (Bass et al., 2012; Inman et al., 2018; LaBar & Cabeza, 2006) and emotion (Satpute et al., 2019). Direct electrical stimulation of the amygdala has been shown to enhance declarative memory in humans in a stimulus-specific manner (Inman et al., 2018); however, response to stimulation is not consistent. Some participants experience memory enhancement, while others do not respond to stimulation or experience a memory impairment. While this variability in response is to be expected in any research into normally-distributed capacities like human memory, it is nonetheless in the best interests of researchers in this field to understand and control the variables affecting stimulation response in order to ensure a more reliable memory enhancement effect. This is especially important if amygdala-mediated memory enhancement is to realize its potential as a therapeutic intervention for memory disorders. Several potential participant variables have been examined for potential roles in memory modulation (Hollearn et al., in prep). Of these, three are of particular interest. First, baseline normative memory scores were found to influence response to stimulation across participants, with participants with higher baseline memory scores being less likely to experience any memory effect, enhancement or otherwise. Second, participant sex was found to be trending toward significance (p = 0.079) as a modulatory factor, with female participants less likely to experience a strong memory enhancement. Finally, the linear distance between the specific location of electrical stimulation and the participant’s hippocampus was not found to meaningfully influence stimulation response. This final finding is of particular interest because it seemingly contradicts previous literature showing that electrical stimulation of the hippocampus impairs memory performance (Suthana & Fried, 2014). One potential explanation for this null finding is that linear distance measurements do not account for enough variables to explain the spatial distribution of electrical stimulation within the brain. A myriad of stimulation-specific variables, such as its amplitude and pulse width (an element of the shape of the electrical wave), all influence the actual area impacted by stimulation. Thus, a more accurate and nuanced visualization tool is needed. Lead-DBS (Horn & Kühn, 2015; Butenko et al., 2020) is just such a tool. It is a MATLAB-based toolbox of visualization software intended to accurately consider all of the variables of an individual stimulation in order to precisely estimate the actual amount of cortical tissue activated for each participant. It allows for visualization of both the local neural tissue activated by stimulation, as well as activation of white matter tracts extending to downstream cortical areas. Both of these capabilities make it a powerful tool for future research in this field. The purpose of this project was to integrate Lead-DBS into the data processing pipeline used in previous studies, to pave the way for future examinations of the variables influencing stimulation response. Imaging data from previous studies were used within Lead-DBS to create visualizations of tissue and pathway activation, providing a qualitative proof of concept for future direct examinations into hypotheses concerning the spatial influences of stimulation on memory outcomes.

References