Open in TheBrain

Gotts, Steve Frontal Cortex Priming

<p>Mar 23, 2009 10:54 PM

&#160;

STEVE GOTTS &#160;now at NIMH.

&#160;

Gave a talk on March 25 in MBC &#160;with &#160;Jay McClelland on&#160;

&#160;

Tuining Curves for Object Concpts in Human Brain.

&#160;

Steve was a student of Dave Plaut &#160;at &#160;CMU ... did a lot on repetition priming.

&#160;

Now at Brain and Cognition Lab &#160;works with Bob Desemone &#160;&#160;collab with Alex Martin

&#160;

&#160;

they ID'd &#160;thousands of meaningful objects &#160;eg &#160;&#160;COW HORSE HAMMER

&#160;

use &#160;&#160;fMRI &#160;adaptation &#160;(ie &#160;decr response)

&#160;

present objects at one per second &#160;(for perhaps an hour or so).

&#160;

present &#160;&#160;five objects &#160;(the ANCHOR) &#160;&#160;a1 &#160;to a5 &#160;then &#160;a deviant &#160;&#160;(eg &#160;face of a COW instead of the body)

then a landscape then the &#160;&#160;response object &#160;&#160;(eg a hammer)

&#160;

used 7 subjects.... &#160;&#160;their instruction &#160;"how similar are these objects.

&#160;

similarity &#160;&#160;decreases exactly as you'd expect &#160;&#160;from &#160;cow &#160;&#160;thru horse &#160;to lobster eg.

&#160;

look at &#160;adaptation and recovery from adaptation

&#160;

get a parallelogram &#160;where &#160;image selection &#160;(ie similarity by raw image is similar on top) &#160;vs

category &#160;selective on bottom &#160;&#160;&#160;(similarity by concept)

&#160;

ie &#160;pereceptual &#160;is closer to the IMAGE &#160;&#160;whereas &#160;Conceptual is closer to category.

&#160;

raw beta weights; &#160;repetition supression

&#160;

conceptual broadening vs sharpening categories.

&#160;

&#160;

&#160;

&#160;

&#160;

&#160;

&#160;</p>

Attachments