Authors :
Presenting Author: Aastha Bansal, Student – Emory University School of Medicine
chen Erica, undergrad – Emory University School of Medicine; Rebecca Roth, BS – Emory University School of Medicine; Adam Dickey, MD – Emory University School of Medicine; Nigel Pedersen, MD – UC Davis; Noah Okada, BS – Emory University School of Medicine; Alena Stasenko, PhD – University of California San Diego; Erk Kaestner, PhD – University of California San Diego; shade Taylor, BS – Emory University School of Medicine; Evan Brady, BS – Emory University School of Medicine; kelsey Hewitt, PsyD – Emory University School of Medicine; Natalie Voets, PhD – Oxford University; Emma Acerbo, PhD – Emory University School of Medicine; Leonardo Bonilha, MD – Emory University School of Medicine; Christine Smith, PhD – University of California San Diego; Robert Gross, MD, PhD – Emory University School of Medicine; McDonald Carrie, PhD – University of California San Diego; Daniel Drane, PhD – Emory University School of Medicine
Rationale:
Memory involving the association between unfamiliar faces and semantic information represents a complex task involving multiple brain regions including the temporal lobe (TL). Although the Face-Name Associative Memory Exam (FNAME) was developed to detect preclinical Alzheimer’s disease, it may also be useful in examining visual-linguistic associative abilities in epilepsy patients, particularly those with TL deficits. As it includes both visual, verbal, and associative learning that crosses modalities it expands upon the typical cadre of memory measures use in neuropsychology, which usually focus on a single stimulus domain (e.g., visual, verbal). We present preliminary data from patients undergoing epilepsy surgery (ES) evaluation and healthy controls (HCs), hypothesizing that different performance patterns will emerge by laterality of seizure onset.
Methods:
The FNAME is a computerized neuropsychological (NP) test that asks individuals to associate names and occupations with 12 novel faces across two learning trials. We administered the FNAME to 27 ES patients undergoing evaluation at Emory University and 6 HC subjects. All patients underwent long-term video-EEG monitoring and are currently in some stage of surgery work-up. Seizure type and lateralization was based on EEG monitoring results, 3T MRI, PET scans, and NP testing. The current cohort has an approximately equal mix of patients with left, right, or bilateral seizure onset and two patients with generalized epilepsy, with no subgroup differences on relevant demographic variables. Performance on the FNAME was analyzed by using an ANOVA with post-hoc comparisons.
Results:
All presurgical subgroups performed worse than HC subjects across the various metrics of the FNAME, with most scales showing significant differences (see Table 1). In general, ES subgroup differences appear to be present, but do not reach statistical significance based on the modest sample size. Nevertheless, there was a clear trend of both left and right-sided ES candidates performing worse than HCs, with the left (language dominant) patients performing worse on the name related tasks (verbal). The bitemporal subgroup performed worse than unilateral patients on all measures. Of note, all subgroups performed much worse on the associational tasks than the HCs, and several of these presurgical patients showing deficits on the FNAME had performed normally on most clinical memory measures.
Conclusions:
Overall, it appears that combining visual and verbal modalities of learning in an associational task may be useful in the evaluation of ES candidates and will provide valuable cognitive assessment information that exceeds traditional NP measures. Preliminary data suggests that subgroups of epilepsy (e.g., unilateral vs. bilateral; left vs. right) may perform differently on this measure, which could be useful for diagnostic and surgical planning purposes. The FNAME may prove useful to filling a gap in the presurgical assessment of epilepsy, although larger sample sizes or patient and HC data are required.
Funding:
NIH/NINDS (R01 NS088748)