Unformatted text preview: e performance responsethe coherent
the 0% plotted
function (Equation
(D) Response times
the optimal
motion of the motion strengths. The response line motion strength 250 is reasonable approximation to
at chance (not shown). The
a best time cumulative Weibull,
each stimulus
h aatcertainis probability. time isisdecisionfitting height.Equation 9)fromwhich ms.aonset until behavioral responsethe logistic factors could cau
probabiliB Response times usingThesebarrier (from The time plusstimulus Note that numerous
function (Equation 8). (D)
the optimal
is plotted
at each of the motion strengths. The response time is decision time (from Equation 9) plus 250 ms.
ated bythe average decision time also depends on B that there are three other intervals, each of a fixed durataking repeated measurements
of , including intrinsic factors like att
Note that
of and the particular list(or group of neurons) to nondecision time t onextrinsic factors like the light level
a neuron of motion strengths chosen by the tion: the a
and each presentation (e.g.,
fix probabilities. In this case, for a given weight
Review
ce (e.g., the stopping point in Banburismus):
305
g Experimental Evidence
Pr(h1m) Neurophysiology: Banburismus in the Brain
weight of evidence B
(3) Pr(anderror, in which the barrier is raised and lowered until
h0m) the maximum rate of reward is achieved. Interestingly, in
this case, the constant of proportionality that relates the
accumulated difference (x y) to the weight of evidence
1 is not needed to find the barrier height that leads to the
maximum rate of reward. That information is not lost,
however: once the barrier height is fixed, it corresponds 0
1
to a particular level of overall performance and thus can
be expressed in units of the weight of evidence, such
as natural bans. Note that this quantity is not the weight
of evidence that would be calculated based on knowledge of the stimulus motion strength and the associated
sensory response distributions because that information would 1
lead to perfect performance at all motion
B
strengths. Rather, it is the weight of evidence that corresponds to a fixed level of uncertainty across all stimulus
strengths in an experime...
View
Full
Document
This note was uploaded on 09/15/2011 for the course COGS 1 taught by Professor Lewis during the Spring '08 term at UCSD.
 Spring '08
 LEWIS

Click to edit the document details