Abstract Details

Title Open-ended, value-based decision making requires semantic memory: a computational, imaging, and clinical study

Topic Aging, Dementia, and Behavioral Neurology

Presentation(s) P1 - Poster Session 1 (8:00 AM-9:00 AM)

Poster/Presentation
Number 7-005

Objective To address the paucity of neurocomputational models describing differences in open-ended, value-based decision-making, here we develop and test a model incorporating semantic memory and preference to assess the hypothesis that memory deficits correlate with distinct decision making impairments across neurodegenerative disorders.

Background Results of neuropsychological evaluations can fail to correlate with the quality of less structured, real-world decisions, which can be impaired in divergent ways across disorders (Chiong, Neurocase, 2013). Importantly, difficulties with such decisions – e.g. choosing a financial advisor – are frequent targets of concern by caregivers. Here we develop a neurocomputational model to explain such category-based choices (Zhang, PNAS, 2021) and determine the contributions of distinct neurobehavioral substrates to decision-making deficits.

Design/Methods Individuals with Alzheimer’s Disease (AD; n=13), behavioral-variant frontotemporal dementia (bvFTD; n=13), and healthy controls (HC; n=16) were recruited at UCSF. Subjects completed three tasks: a category fluency task; a menu-based choice (MENU) task, in which a list of options was provided; and an open-ended choice (OPEN) task, in which subjects were required to generate the options themselves. Primary outcome variables included the number of choice switches from OPEN to MENU, and the length of the fluency list.

Results Patients with AD and bvFTD switched significantly more than HC (p < 0.05). HC demonstrated significantly greater category fluency than either dementia group (p < 0.05), and bvFTD patients showed significantly greater fluency than did AD patients (p < 0.05). Across individuals matched for dementia severity (target CDR<=1), category fluency was significantly and inversely associated with the number of choice switches (r = -0.51, p < 0.001). Preliminary findings show that both bvFTD and AD had lower choice consistency than controls (p < 0.05), but not compared to each other.

Conclusions The degree of semantic memory impairment correlates with decision-making deficits across neurodegenerative disorders, and partly explains choice instability.

Authors/Disclosures
Andrew Kayser, MD, PhD, FAAN (UCSF) PRESENTER	Dr. Kayser has received personal compensation in the range of $500-$4,999 for serving as a Consultant for Boehringer Ingelheim. The institution of Dr. Kayser has received research support from NIH. The institution of Dr. Kayser has received research support from VA RR&D.
	No disclosure on file
	No disclosure on file
	No disclosure on file
	No disclosure on file
Winston Chiong, MD (UCSF Memory and Aging Center)	The institution of Dr. Chiong has received research support from the National Institutes of Health. The institution of Dr. Chiong has received research support from the Dana Foundation. Dr. Chiong has received personal compensation in the range of $500-$4,999 for serving as a member, NeuroEthics Working Group with National Institutes of Health.
	No disclosure on file

��ɫ��

��ɫ��