in
temporal cortical visual areas of infant
monkeys
H. R. Rodman, S. P. Scalaidhe and C. G.
Gross
Department of Psychology,
Princeton University
1. Inferior temporal cortex (IT) is a
"high-order" region of primate temporal visual
cortex implicated in visual pattern perception
and recognition. To gain some insight into the
development of this area, we compared the
properties of single neurons in IT in infant
monkeys ranging from 5 wk to 7 mo of age with
those of neurons in IT in adult animals. Both
anesthetized and awake behaving paradigms were
used.
2. In immobilized infant monkeys under
nitrous oxide anesthesia, the incidence of
visually responsive cells was markedly less than
in adult monkeys studied under similar
conditions. In infants 4-7 mo of age, only half
of IT neurons studied were visually responsive,
compared with > 80% in adult monkeys. In
monkeys < 4 mo old, even fewer (< 10%)
could be visually driven. "Habituation" of IT
cells to repeated stimulus presentation appeared
more pronounced in infant monkeys under nitrous
oxide anesthesia than in adult animals.
3. IT cells in the anesthetized infant
monkeys that did respond showed receptive field
properties similar to those of responsive adult
IT neurons studied under similar conditions. Two
thirds of the receptive fields plotted in the
anesthetized 4 to 7-mo-old group were bilateral,
and median field size did not differ between the
infants and comparable adult groups, being
approximately 20 degrees on a side in each case.
4. In contrast to the results obtained under
anesthesia, most IT cells in alert infant
monkeys 5 wk-7 mo of age (80%) were responsive
to visual stimuli, and this incidence of
visually responsive IT neurons did not differ
from that obtained in awake adult macaques.
However, response magnitude, measured as spikes
per second above baseline rate, was
significantly lower in the infant alert sample
than in the adult control (5.2 vs. 12.6
spikes/s, mean +/- SE, deviation from
spontaneous rate, respectively).
5. In addition to having lower magnitudes of
visual response, IT cells in the awake infants
also tended to have longer and more variable
latencies. The overall mean for the infant cells
was 196 ms, compared with an overall mean of 140
ms for IT neurons in the alert control adult.
6. Although the magnitude of response of
neurons in alert infant IT cortex was lower
overall, the incidence and features of stimulus
selectivity shown by alert infant IT neurons
were strikingly similar to those of IT cells of
both anesthetized and unanesthetized adult
monkeys. Within the 2nd mo of life (i.e., as
early as we could test), individual IT neurons
exhibited responses selective for shape
(boundary curvature), for faces, for arbitrary
geometrical patterns, and for color. Several
measures of form selectivity for the alert
infant and adult samples indicated that the
overall degree of stimulus selectivity did not
differ between the groups.
7. For both anesthetized and alert infant
groups, rates of spontaneous activity in IT
cortex were lower than those seen under
comparable conditions in adult monkeys. The low
spontaneous rates in infant temporal cortex in
both anesthetized and awake behaving paradigms
suggest that the low response magnitudes in
infant IT reflect general characteristics of
cellular function in high-order cortical areas
of infant monkeys.
8. We also recorded from neurons in the
superior temporal polysensory area (STP),
another high-order region of temporal cortex,
which appears to be involved in analysis of
complex visual motion and in orientation
functions. The appearance of visual responses in
STP paralleled that seen in IT cortex. In the
anesthetized infants, responses were virtually
absent in STP before 4 mo of age, but had
adultiike properties when they first appeared in
slightly older animals. All visually responsive
STP cells studied in 4 to 7-mo-old infants had
bilateral visual receptive fields, and about
half were multimodal. Virtually all STP cells
studied in alert infant and adult animals were
visually responsive.
9. As a control for the paucity of visually
responsive neurons in IT and STP of the youngest
infant monkeys under nitrous oxide anesthesia,
we also recorded from striate cortex and
extrastriate visual area MT in several sessions.
Eighty-five percent of cells in these areas were
responsive, arguing against a general
suppression of cortical function. Cells in MT
were selective for direction of stimulus motion,
and cells in striate cortex were selective for
orientation of a bar of light.
10. The delayed contribution of IT cortex to
visual behavior in monkeys and the slow
development of adult capacity are not due to a
nonspecificity of neuronal responses in IT
cortex in infancy. Rather, these phenomena may
reflect the weakness of signals deriving from IT
within the first half year of life and possibly
beyond. More generally, the development of
adultlike levels of neuronal excitability
appears to be more protracted in high-order
temporal areas subserving complex functions than
in cortical areas earlier in the monkey visual
pathway.
Responses of an IT neuron in an anesthetized
infant monkey that responded preferentially to
face stimuli. For each stimulus, the line
underneath the PSTH corresponds to the 2-s
period during which the stimulus was presented.
Stimuli 1-11 were colored slides; stimuli 12-18
were white on a black background. Face stimuli
consisted of images of different facial
expressions from 2 stimulus monkeys, a profile
view of a 3rd, and scrambles of the faces. The
stimuli were presented at the fovea and were
3-6° diam.
