Evaluating Long-Term Memory: Comparative Validity of Emory Multimodal Learning Test (EMLT) and Standard Memory Tasks in Epilepsy Patients
Abstract number :
1.286
Submission category :
4. Clinical Epilepsy / 4A. Classification and Syndromes
Year :
2025
Submission ID :
256
Source :
www.aesnet.org
Presentation date :
12/6/2025 12:00:00 AM
Published date :
Authors :
Presenting Author: Faith Isibor, – Emory University
Noah Okada, BS – Cal Tech
Adam Dickey, MD, PhD – Baylor College of Medicine
Taylor Shade, BS – Emory University School of Medicine
Evan Brady, BS – Emory University School of Medicine
Nealen Laxpati, MD, PhD – Emory University School of Medicine
Mary Ann Manley, BS – Emory University School of Medicine
Keith Adams, BS – Spin VFX
Daniel Drane, PhD – Emory University School of Medicine
Rationale: We explored the validity of a novel memory measure, the Emory Multimodal Learning Test (EMLT), to overcome potential gaps in memory assessment in epilepsy surgery. The EMLT uses an immersive, “gamified” format requiring patients to recall narrative videos, allowing us to assess multimodal integration of information and features of classic simple domain encoding (i.e., unimodal learning of visual or auditory stimuli). Created in Unity game engine and employing AI Whisper for verbal transcription and automated scoring, the EMLT enables patient assessment over any duration of recall, enabling measurement of rapid long-term forgetting and consolidation/decay processes.
Methods: The EMLT was administered to epilepsy surgical patients (N=34) who had undergone clinical neuropsychological evaluation. To explore convergent and discriminant validity with standard test measures, we required a convergent finding to have a significant Pearson correlation coefficient (r) of at least 0.6 or greater magnitude, while we considered anything below r=0.4 to represent an unrelated task (i.e., discriminant validity). Although the EMLT produces multiple scores, we initially explored only the verbal memory and narrative (visual) content scores measured at immediate, 30-minute delayed recall, and 1-week recall.
Results: The EMLT immediate verbal memory score was most strongly related to a 30-minute delayed verbal gist learning task (r=.56, p< .01
Clinical Epilepsy