<html>
<head>
<meta http-equiv="content-type" content="text/html;
charset=ISO-8859-1">
</head>
<body bgcolor="#FFFFFF" text="#000000">
Ethan Meyers, Postdoctoral Associate in the Center for Brains, Minds
and Machines at MIT<br>
"Using population decoding to understand how learning a new task
changes neural processing"<br>
<br>
<u>Abstract</u>: New machine-learning-based statistical methods are
revolutionizing the way data is analyzed in a broad range of fields.
In this talk I will discuss new methods that I developed which can
'decode' what types of information are contained in the activity of
populations of neurons. To illustrate the power of this technique, I
will describe a study where we examined how the information in the
prefrontal cortex (PFC) changes after macaque monkeys learned to
perform a new task. Given that primates are continually learning new
tasks, understanding how new information is integrated into neural
systems is fundamental for understanding how the brain enables
complex behaviors.<br>
<br>
Questions we addressed in this study include: 1) Does learning a new
task change the amount of information about basic visual features of
a stimulus or does it only change the amount of information about
more complex task-related variables? 2) Does the new information
arise due to the emergence of a few highly selective neurons, or is
information evenly distributed across the population? 3) Do neurons
become specialized to process only one type of information or can
the same neuron carry multiple types of information? 4) Is the new
information contained in a dynamic population code, or is there one
stationary pattern of neural activity that contains the new
information? and 5) Are there differences in the information content
between dorsal and ventral PFC, and does learning affect these two
brain regions equally? Future directions for how these methods can
be expanded to give additional insight into neural processing, and
how the methods can be applied to other areas outside of
neuroscience, will also be discussed.<br>
<br>
<u>Biography</u>: Ethan Meyers is a postdoctoral associate in the
Center for Brains, Minds and Machines at MIT. His research focuses
on creating machine-learning-based statistical data analysis methods
that are useful for analyzing high dimensional neural signals.
Through collaborations with experimental neuroscientists, his work
has given new insight into how information is stored in working
memory, how attention influences neural coding, and how new
information is incorporated into neural activity. Ethan received his
Ph.D. in computational neuroscience from MIT where he was a NDSEG
Fellow, and his BA in computer science from Oberlin College where he
graduated with high honors.<br>
<br>
<div class="moz-signature">-- <br>
<meta http-equiv="content-type" content="text/html;
charset=ISO-8859-1">
Paula Harmon, Administrative Assistant <br>
<div class="moz-signature"><small> School of Cognitive Science <br>
Hampshire College<br>
893 West Street Amherst, MA 01002 <br>
phone: 413.559.5502 <br>
fax: 413.559.5438 <br>
<a href="http://cs.hampshire.edu">http://cs.hampshire.edu</a></small>
</div>
</div>
</body>
</html>