Neuroscience 2005 Abstract
| Presentation Number: | 443.10 |
|---|---|
| Abstract Title: | Multi-site characterization of a fMRI working memory paradigm: reliability of performance. |
| Authors: |
Gollub, R. L.*1,2
; Vangel, M.2
; Wallace, S.2
; O'Leary, D.
; Bockholt, J.3
; Mueller, B.4
; Greve, D.2
; Manoach, D. S.1,2
1Psychiatry, Massachusetts General Hospital, Charlestown, MA 2Martinos Center, Massachusetts General Hospital, Charlestown, MA 3NM, Bldg Rm 2660, 02129, 4USA, Bldg Rm 2660, 02129, |
| Primary Theme and Topics |
Disorders of the Nervous System - Cognitive, Emotional and Behavioral State Disorders -- Schizophrenia: Functional deficits |
| Session: |
443. Schizophrenia: Functional Deficits I Poster |
| Presentation Time: | Monday, November 14, 2005 9:00 AM-10:00 AM |
| Location: | Washington Convention Center - Hall A-C, Board # TT45 |
| Keywords: | schizophrenia, neuroimaging |
We investigated the reliability of a working memory (WM) task, an adaptation of the Sternberg Item Recognition Paradigm (SIRP), for use as a biomarker for The MIND Institute (www.theMINDinstitute.org) multi-site, longitudinal study of schizophrenia.
Ten healthy subjects (5 ♀) traveled to the four sites (UNM, UMN, UI, MGH) and scanned while performing 4 SIRP runs on each of two days (test-retest). Subjects were presented with memory sets of 1, 3 or 5 digits and then responded to each probe digit by indicating whether it was a target (member of the memorized set) or a foil. Outcome measures were response accuracy and slope and intercept of reaction time (RT) plotted by WM load.
All subjects performed at or near ceiling accuracy (94.4 ± 3.2% correct). A straight line described the overall average RT by WM load data, 514.2 + 32.4*Load, R2=0.99. The RT data were best fit using a mixed model ANOVA with Load (1, 3, 5), Visit (test-retest), ProbeType (target, foil) and Order (sequence of sites) as fixed factors and Subject, Site, and Run as random effects. Subject contributed 74% of the variance in the model, Site 5.4%, their interaction 13.3% and Run 7.1%. The fixed effects of WM load, ProbeType, and Visit were all highly significant (p<0.00001). On average, RTs were faster at retest and slower on foils than targets. The test-retest change in RT intercept was significant but represented only 2% of the mean. The Order effect was marginally significant; further analyses will determine if this is specific to the SIRP.
Subjects showed reliable SIRP performance, both in terms of accuracy and the slope of RT, within subjects, across visits, and across sites. This validates the use of the SIRP as a stable and reliable probe of WM performance for multi-site longitudinal studies.
Supported by DE-FG02-99ER62764; fBIRN (www.nbirn.net); MGH GCRC MO1-RR001066
Sample Citation:
[Authors]. [Abstract Title]. Program No. XXX.XX. 2005 Neuroscience Meeting Planner. Washington, DC: Society for Neuroscience, 2005. Online.
Copyright © 2005-2025 Society for Neuroscience; all rights reserved. Permission to republish any abstract or part of any abstract in any form must be obtained in writing by SfN office prior to publication.
