[Probcogsci] Fwd: next kutas lab meeting

[Vicente Malave]

This may be of interest to some of you. The punchline for Bayesians is
"don't make point estimates".

---------- Forwarded message ----------
From: Marta Kutas <mkutas at ucsd.edu>
Date: Fri, Nov 13, 2009 at 2:25 PM
Subject: next kutas lab meeting
To: department at cogsci.ucsd.edu, Esmeralda De Ochoa <edeochoa at cogsci.ucsd.edu>,
Paul Krewski <pkrewski at cogsci.ucsd.edu>, David Groppe <
dgroppe at cogsci.ucsd.edu>, Ben Cipollini <bcipolli at cogsci.ucsd.edu>, Vicente
Malave <vicente.malave at gmail.com>, Paul Hammon <paul.hammon at gmail.com>,
Nathan White <natew4 at yahoo.com>, Marta Kutas <kutas at cogsci.ucsd.edu>, Steve
Barrera <sjbarrera at ucsd.edu>, Irina Gorodnitsky <igorodni at cogsci.ucsd.edu>,
"Leslie J. Carver" <ljcarver at ucsd.edu>, "Steven A. Hillyard" <
sah at sdepl.ucsd.edu>, Seana Coulson <coulson at cogsci.ucsd.edu>, Aniruddh Patel
<apatel at nsi.edu>, Eric Halgren <ehalgren at ucsd.edu>, "Anders M. Dale" <
amdale at ucsd.edu>, Robert Kluender <kluender at ling.ucsd.edu>, Ying Wu <
ywu at cogsci.ucsd.edu>, David Brang <dbrang at ucsd.edu>, Jennifer Collins <
jcollins at cogsci.ucsd.edu>, Tristan Davenport <punninglinguist at gmail.com>,
Julia Evans <jevans at mail.sdsu.edu>, Marguerite Mcquire <
mmcquire at crl.ucsd.edu>, Matt Schalles <matthewdaves at yahoo.com>, Melanie
Tumlin <melanie.tumlin at gmail.com>, Laura Kertz <kertz at ling.ucsd.edu>, "Anna
E. M. Holt" <aeholt at cogsci.ucsd.edu>, Chris Barkley <cbarkley at ling.ucsd.edu>,
lrosenfelt <lrosenfelt at ling.ucsd.edu>, Klinton Bicknell <kbicknell at ucsd.edu>,
Wen-hsuan Chan <whchan at cogsci.ucsd.edu>, Philip Hofmeister <
phofmeister at ucsd.edu>, Lindsay Crissman <lcrissman at cogsci.ucsd.edu>,
Nathaniel Smith <njsmith at cogsci.ucsd.edu>, Marisa Sizemore <
msizemore at crl.ucsd.edu>, Ross Metusalem <rmetusal at cogsci.ucsd.edu>, Jamie
Alexandre <jdalexan at ucsd.edu>, Kim Sweeney <ksweeney at cogsci.ucsd.edu>



Monday, Nov 16, 3:30, CSB 280

Vicente Malave

Classification Caveats : Practical limits of Multi-Voxel Pattern Analysis in
fMRI.

Multivariate pattern classification techniques are attracting increasing
attention as an alternative data analysis technique in cognitive
neuroscience. The purpose of this talk is to provide a critical look at two
inferences using these methods. The first inference is simply that two
conditions of the experiment are different. We discuss the connection
between pattern classification methods and classical multivariate hypothesis
testing. We briefly discuss resampling-based statistics, statistical power,
and the impact on experimental design. The second inference is in
interpretation of the model learned by a linear classifier. In particular,
many studies compute "importance maps" that purport to show where
discriminating information between the two conditions is located in the
brain. We provide a critical look at this inference by noting that many very
different models can achieve similar classification accuracy, and many very
similar models can perform at chance. We show some techniques for exploring
the space of possible models, discuss proposed solutions from the
literature, and urge caution when making inferences about the importance of
particular feature in classification.


all welcome



-- 
Vicente Malave
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://pidgin.ucsd.edu/pipermail/probcogsci/attachments/20091113/8e3df481/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: mkutas.vcf
Type: text/x-vcard
Size: 360 bytes
Desc: not available
URL: <http://pidgin.ucsd.edu/pipermail/probcogsci/attachments/20091113/8e3df481/attachment.vcf>