SAFETYLIT WEEKLY UPDATE

We compile citations and summaries of about 400 new articles every week.
RSS Feed

HELP: Tutorials | FAQ
CONTACT US: Contact info

Search Results

Journal Article

Citation

Rorden C, Karnath HO. Neuropsychologia 2010; 48(9): 2758-2763.

Affiliation

Center for Advanced Brain Imaging, Georgia Institute of Technology, GA, USA.

Copyright

(Copyright © 2010, Elsevier Publishing)

DOI

10.1016/j.neuropsychologia.2010.04.018

PMID

20433859

PMCID

PMC3129646

Abstract

Cancellation tasks are popular clinical and scientific tools for identifying spatial neglect, with neglect patients tending to miss targets on the contralesional side of the test. However, methods for analysis are not well established. Indeed, these tests are often used as a binary classifier to simply identify the presence or absence of spatial neglect, even though it is clear that there is a spectrum of disability on these tasks. We suggest that the Center of Cancellation (CoC) provides an intuitive, continuous and robust measure of neglect severity. First employed by Binder and colleagues Arch Neurol. 1992; 49: 1187-1194., its use has not been replicated since. Our aim was to ease deployment of this measure through validation, development of software and focused exposition. To validate this index, we evaluated a group of 110 individuals with right hemisphere injury. For two different cancellation tasks (the Bells Test and the Letter Cancellation Task) we predicted spatial neglect (as defined by independent measures) using the new CoC index. Examining each individual's performance on a single cancellation task, we were able to correctly determine with better than 98% accuracy whether three tests with binary classifiers would define them as having spatial neglect. Specifically, an acute CoC score greater than 0.081 on the Bells Test or 0.083 on the Letter Cancellation Task turned out to indicate neglect behavior after a right hemisphere brain lesion. Finally, we provide free software allowing other groups not only to rapidly analyze new but also previously existing (paper-and-pencil based) datasets using this measure.


Language: en

NEW SEARCH


All SafetyLit records are available for automatic download to Zotero & Mendeley
Print