CONTRIBUTION OF AREA MST TO THE SPATIAL FUNCTIONS OF THE PARIETAL LOBE 
Jean-Pierre Roy, M.D., International Research Scholar 
Dr. Roy studies the response of neurons in the 
medial superior temporal (MST) area of the cerebral 
cortex of the behaving Rhesus monkey. The goal of 
this research is to understand how low-level signals, 
such as those indicating the direction of motion or 
binocular disparity preferred by a neuron, are 
transformed into higher-level signals — for ex- 
ample, those defining the spatial structure of the 
environment. 
During a fellowship with Dr. Robert Wurtz at the 
National Institutes of Health, Dr. Roy found a sub- 
class of cells in area MST that responded for one 
disparity when the stimulus moved in one direction 
— for example, crossed disparities corresponding to 
motion in front of the point fixated — and re- 
sponded for the opposite disparity, here uncrossed 
disparities, when the stimulus moved in the oppo- 
site direction, corresponding to motion behind the 
point fixated. These cells, the disparity-dependent 
direction-selective (DDD) neurons, were proposed 
to be involved in transforming the visual informa- 
tion about the foreground and background motions 
into information about the direction of self-motion 
of the subject in his or her environment. 
This can be understood if one pictures what hap- 
pens to the images of the environment as one moves 
forward while looking at an object located roughly 
perpendicular to the direction of motion. The fore- 
ground, the environment in front of the object fix- 
ated, appears to move in one direction, and the 
background, the environment behind the object fix- 
ated, moves in the opposite direction. A DDD neu- 
ron responding to leftward motion of the fore- 
ground and rightward motion of the background 
will respond for rightward motion of a subject look- 
ing forward irrespective of where in depth the sub- 
ject looks. This neuron, by combining the two 
correct signals about disparity and direction, indi- 
cates the direction of self-motion relative to the ob- 
ject fixated. 
These neurons provide a way of asking an impor- 
tant question about the transformation of another 
type of low-level signal, relative speeds, into a sig- 
nal about the spatial structure of the environment. 
Because the condition under which the DDD neuron 
discharges is known (self-motion in one direction 
with perpendicular gaze) , it is possible to explore 
this second question. When moving in one direc- 
tion while looking perpendicularly, a parameter 
different from direction and disparity has to be 
examined — relative speed. Objects from the envi- 
ronment that are close to the point fixated, whether 
they are in the foreground or background, move rela- 
tively slowly. Objects that are far from the object, 
i.e., very far or very close, move relatively fast. Rela- 
tive speed contains information about the depths of 
objects with respect to the object fixated. It has to 
be understood that it is not absolute speed that car- 
ries the depth information, since the absolute 
speeds will vary with the speed of displacement of 
the subject. Rather, it is the speed ratios between 
objects at different depths, i.e., speed gradients. 
To explore the response of DDD neurons from 
MST to such speed gradients, Dr. Roy is comparing 
the response of one DDD neuron with stimuli that 
do and do not include a speed gradient. This re- 
sponse is also compared with stimuli that do and do 
not include a disparity gradient to evaluate the con- 
tribution of this parameter. The prediction is that 
speed gradient does contribute to the cell's re- 
sponse, and if so, this could be how a DDD neuron 
conveys information about the three-dimensional 
structure of the environment. 
Preliminary data obtained recently seem to con- 
firm this prediction, but more work remains to be 
done before any firm conclusion can be reached. 
Dr. Roy is Assistant Professor of Neurology and 
Neurosurgery at the Montreal Neurological Insti- 
tute, McGill University. 
532 
