ExperienceDependent Brain Development

1
Experience-Dependent Brain Development
The processes of CNS development summarized in
previous lecture occur largely during prenatal
life, that is, before the developing organism has
a chance to interact with the highly complex
external environment, including other animals.
Thus, some aspects of the initial growth of the
CNS appear to be regulated mostly by intrinsic
processes and genetic programs. However, The
subsequent, further growth, maturation, and fine
tuning of the CNS depends critically in
experience in the form of exposure to sensory
stimulation, motor performance, and
social interactions. The ethologist Konrad
Lorenz was among the first to provide
experimental evidence for the important role of
sensory and social experience for
behavioral/social (and thus CNS) development in
postnatal life Imprinting exposure to moving
animal (mother) during critical postnatal period
required for formation of social bond
readiness of CNS to acquire and encode
specific set of sensory stimuli
2

• Rene Spitz (1940s) psychoanalyst
• compared two groups of children
• 1. Foundling home for abandoned children
• (1 nurse/7 children, covered cribs, social
  and sensory deprivation)
• 2. Nursing home attached to prison for women
• (mother can be with children, open cribs)
• - 4 months foundling gt nursing home kids (motor
  and intellectual performance)
• 1 year founding lt nursing home
• 2-3 years nursing home kids show normal
  development (motor, language)
• 2/26 foundling home kids able to walk and talk
• sensory and social deprivation results in severe
  developmental delays

3
Harry Margaret Harlow (1960s) reared
monkeys in isolation (first 6-12 months) -gt
physically healthy -gt severe behavioral
disturbances (socially withdrawn, lack of
playing, fighting, sex, stereotypes
behavior, similar to autistic children) -
effects occurred only during critical period -
partial reversal with close social contact later
in life Marius von Senden (1932) - review of
literature on long-term effects pf childhood
cataracts (opacity in lens) - usually removed in
infants - removal after age 10-20 years leads to
permanent deficits in form perception Austin
Riesen (1947) -two monkeys raised in darkness
for first 3-6 months of life -gt profound visual
deficits (e.g., object discrimination) extensive
training can normalize deficits Summary
several lines of independent work suggested that
interactions with the sensory and social
environment necessary to allow normal brain
development and maturation.
4
D. Hubel and T. Wiesel -single cell recordings
in primary visual cortex 1. Orientation-selecti
ve cells (line/orientation detectors)
5
2. Eye-selective cells (ocular dominance, OD)

• -Hubel and Wiesel showed that
• ocular dominance columns are
• present at birth, prior to
• visual experience
• -monkeys show OD before birth
• -in ferrets, OD is present before
• retina and cortex become light-
• responsive
• -eye removal (PD0 in ferrets) does
• not block OD formation
• intrinsic, genetic factors
• regulate the initial development
• of OD columns they do not
• require visual experience

6
However, OD columns formed are labile and depend
on visual inputs for their refinement and
stabilization. Hubel and Wiesel showed this by
using visual deprivation during early postnatal
life (binocular and monocular deprivation). As
shown below (A), in normally reared monkeys,
there is an orderly representation of both eyes
in V1 (white stripes indicate inputs from eye
injected with transneuronal marker, black eye
represent the non-injected eye). Closing one eye
soon after birth (2 weeks) results in a
significant loss of inputs from the closed eye
(B showing injections into the open eye, white C
showing injections into the closed
eye). (from Hubel,
Wiesel, and LeVay, 1977)
7
Electrophysiological recordings in V1 showed that
the majority of cells is excited by
stimulation of either left or right eye, but not
both, thus, they show ocular dominance. After
monocular deprivation, visual stimulation of the
previously closed eye no longer is effective in
driving V1 neurons, that is, cells no longer
respond to inputs from that eye.
(adapted from Hubel
and Wiesel, 1977)
8
The shift in cortical OD columns occurs due to a
loss of axonal inputs from the lateral geniculate
nucleus (LGN) carrying information from the
deprived eye. Conversely, axons from the open eye
expand/branch out to occupy increased areas of
cortical space. Functionally, animals are blind
for the deprived eye, despite normal transmission
in the retina and LGN. This phenomenon is
explained by the loss of synaptic input from
the deprived eye in V1.
(from Kandel et al., Principles of Neural
Science, Elsevier)
9
Competition between the inputs from the two eyes
appears to be the synaptic mechanism that allows
axons carrying input from the open eye to take
over part of the cortex previously devoted to the
closed eye. It is likely that competition for
growth factors determines which synapses in V1
are maintained, grow, or are eliminated.
10
Hubel and Wiesels work provided direct evidence
that -different levels of sensory stimulation
alter cortical structure -brain development
involves two distinct processes 1. Early
migration of axons (chemical signals) 2.
Refinement and stabilization of synaptic
connections by sensory experience and
competition 3. The effects of visual
deprivation are greatest when they occur
in early postnatal life equivalent effects are
not seen when deprivation is given to adult
animals (critical period)
Example of a critical period for monocular
deprivation effects in kittens (from Olson and
Freeman, 1980)
11
Clear ocular dominance columns are found in layer
4 of the visual cortex (layer 4 is the input
layer for axons from the LGN). After layer 4,
inputs carrying information from each eye
converge and cells become responsive to
binocular inputs (stereopsis, depth perception).
Binocular Convergence
Ocular Dominance
Binocular Convergence
LEFT RIGHT LGN
12
Monocular deprivation causes a loss of
binocularly responsive cells in layers 2-3.
(from Bear et al., Neuroscience, LWW)
13
Less invasive methods of disturbing visual input
also change the properties of neurons in
V1. E.g., experimentally-induced strabismus (eyes
are not properly aligned, cross-eyed,
wall-eyed). This causes inputs from the two
parts of the retina that normally converge in V1
to be off or misaligned, resulting in a loss of
binocularly responsive cells in V1. Clinically,
uncorrected strabismus leads to a loss of
stereoscopic depth perception, which depends on
comparing simultaneous input from both retinas. A
further, important clinical implications is that
visual deficits (cataracts, strabismus) should be
correct as early in life as possible, before
changes in V1 become hard-wired and
irreversible.
14
Interestingly, in postnatal life, classical
neurotransmitters/modulators act as
permissive factors that allow experience-dependent
reorganization of cortical synapses to occur.
Loss of acetylcholine and noradrenaline fibers to
V1 blocks the induction of OD plasticity in
kittens. Thus, transmitters can act somewhat like
growth/trophic factors that are necessary of
cortical synaptic plasticity.
(from Bear and Singer, 1986)
15
Summary Cortical synaptic development occurs in
distinct stages, (a) the initial,
experience- independent controlled growth of
axons and formation of ocular dominance columns,
followed by (b) the experience-dependent
stabilization and refinement of synapses.
(modified from Katz Crowley, Nat. Rev.
Neurosci. 3 2002)
16
What is the advantage of experience-dependent
development? -optimize match between features
of sensory environment and neuronal
representation -allow for adjustment for
changes in sensory input during early life
(e.g., head growth -gt need to adjust for
growth-related changes in binocular
vision) -optimal pruning of synapses based on
competition (select most functional synapses)
17
Other Examples of Experience-Dependent Brain
Development
The mature primary auditory cortex (A1) is
characterized by is tonotopic organization, that
is, different frequencies (pitches) of sound are
are mapped onto A1 in an orderly fashion.
Different columns of A1 have selective tuning to
a preferred frequency that produces maximal
excitation of cell in that column.
18
Chang and Merzenich (Nature 300, 2003) examined
whether the absence of patterned sound input
during early postnatal life alters the columnar
organization/tonotopic mapping of A1. in rats.
To deprive animals of patterned sound, white
noise was played continuously during the first
few weeks of postnatal life.
Tonotopic map of young rat, consisting mostly
of broadly tuned, high- frequency neurons
Over several weeks, a more organized frequency
map emerges
White noise-reared rats do not show this
develop- Mental maturation and refinement of A1
19
Importantly, the developmental abnormalities
produced by white noise rearing are not permanent
and can be reversed by exposure to normal sound
environments. Thus, white noise produces a
developmental arrest, not a permanent,
irreversible abnormality.
20
The results from Chang and Merzenichs study are
important in that they show that 1. Brain
development depends on sensory experience 2.
The brain requires patterned sensory input to
undergo normal development. non-patterned
input is not sufficient (white noise,
bright light) 3. Absence of patterned input
leaves the brain in an immature, plastic
state (i.e., critical developmental periods are
prolonged) gt the brain waits for
appropriate sensory stimulation before
committing to the development of a mature,
more hard-wired synaptic connectivity
and corresponding functional
specialization of neurons