Rapin, I & Katzman, R. (1998). Neurobiology of autism. Annals of Neurology, 43(1), 7-14.
Amygdala Abnormality
and Its Role in
Autistic Socio-emotional Impairment:
A Proposed Study of Somatic Intervention
Among Macaque Monkeys
Davina Rosen
Department of Psychology
Villanova University
Edited by Cynthia Kaschub
Autism is a neurodevelopmental
disorder marked by pervasive impairments in behavior that manifest by three
years of age, and sometimes earlier (American Psychological Association (APA),
1994). The characteristic behavioral impairments include hyperactivity, highly
repetitive behaviors, emotional indifference, and lack of social engagement and
reciprocity. Although affliction rates stand as high as 1 to 2 of every 1,000
individuals, and approximately $3 billion is expended yearly on research and
social care, no definitive cause of the disorder has been identified (National
Institute of Mental Health [NIMH], 2003). In recent years, however, neuroimaging and biochemical techniques have elucidated the
pathophysiologic underpinnings that relate to the
inability to exhibit typical socially- and emotionally-relevant behaviors (for
review see Sweeten, Posey, Shekhar, & McDougle 2002). Face processing, when facial features are
analyzed in order to extract social or emotional information, is one such
behavior that is paramount to social interaction, and is impaired in autistics.
Humans have evolved mechanisms that allow them to
perceptually analyze a face, particularly the expression it holds (Grelotti, Gaussier, &
Schultz, 2002). During facial analysis, emotions and social signals are
spontaneously extracted. Utilizing infrared corneal reflection techniques that
follow eye movement as processing occurs, research has
established that normal individuals use configural
processing in face analysis (Pelphrey, Sasson, Reznick, Paul, Goldman,
& Piven, 2002). Specifically, an individual
directs most attention to the eyes, nose, and mouth, with approximately 70% of
this attention focused on the eyes (Pelphrey et al.,
2002). In contrast, visual scanpath observations during
autistic face processing demonstrate a different analysis strategy. Little to
no attention is directed toward the eyes or the nose; instead, autistics scan
the mouth and lower, non-feature regions of the face. Consequently, when
pictures are presented depicting a facial expression of one of the six
universally-basic human emotions (happiness, sadness, fear, anger, surprise,
disgust) (Ekman, 1992), autistics engage in non-configural (or segmental) processing and misidentify the
depicted emotion 30% of the time (note: 30% is seemingly low for
misidentification; this issue will later be clarified) (Pelphrey
et al., 2002). The percentage of misidentification increases in proportion to
an increase in the demanding nature or emotional weight of the task (Grelotti et al., 2002). Further, autistics fixate on faces
for significantly shorter periods of time in comparison to healthy individuals
(Pelphrey et al., 2002). Interestingly, facial
recognition is the only form of facial perception that shows little to no impairment
among autistics.
The crucial factor that impairs the face-processing
strategy in autism, while leaving facial recognition abilities intact, is the
presence of emotional and social information (Grelotti
et al., 2001). In recent years, functional neuroimaging
and biochemical studies have elucidated the neuroanatomical
structures active during emotional and social perception (Davidson and Slagter, 2000; Haznedar, Buchsbaum, Wei, Hof, Cartwright, & Bienstock,
2000). Moreover, differences have been localized and qualified that distinguish
healthy from autistic participants (Narumoto, Okada, Sadato, Fukui, & Yonekura,
2001; Sweeten et al., 2002). The neural substrates of emotion have been
identified by presenting affective stimuli while neuroimaging
techniques record regional activity (Hall, Szechtman,
& Nahmias, 2003). Facial expressions are the most
widely utilized form of affective stimuli, for they arguably carry the greatest
social and emotional import (Davidson et al., 2000; Narumoto
et al., 2001; Hall et al., 2003; Fudge and Emiliano,
2003).
An added benefit of using facial presentation arises
from the fact that different facial expressions can impart strong, moderate, or
absent emotional information. Therefore, specific regions have been established
as the substrates of emotional processing by comparing two forms of
task-related neuroimaging scans: those that show
neural activity during the presentation of emotional expressions, and those
that show neural activity during the presentation of neutral faces (Narumoto et al., 2001). These scans have implicated an area
of the cerebral cortex, the right superior temporal sulcus
(STS), as the region active during emotional processing of faces (Narumoto et al., 2001; Hall et al., 2003). Activity is
enhanced in the STS in response to the selective attention of healthy
participants to facial emotion. In contrast, non-emotional attention elicits no
response from the STS.
Other lines of research converge on the STS as a
substrate for social perception activities such as face-processing. Intercranial recordings of single cells located in the rostral/dorsal portion of the right STS highlight clusters
of neurons that respond solely to expression (Narumoto
et al., 2001). Related research demonstrated that electrical stimulation of
these clustered neurons impairs the ability to accurately identify expressions.
The expression-responsive neurons are therefore thought to comprise a sensory
processing area of the temporal lobe (Iacoboni, Koski, Brass, Bekkering, Woods, Dubeau, et al., 2001). This area essentially receives
visual information, to which it then connects semantic meanings. The STS is
also active during related forms of interpersonal communication such as
following the eye gaze of others. It is therefore noteworthy that the activated
cluster of neurons lies near the V5/MT brain region (Visual Area 5), which is
responsible for the perception of moving visual stimuli (Iacoboni,
2001; Narumoto et al., 2001). Consequently, the
ability to perceive emotional expressions as they change occurs due to axonal
connections that allow for communication between the right STS and V5/MT brain
regions.
The above neuroimaging
studies on neural activity during facial presentation establish the conclusion
that the right STS functions during fine-grained spatial processing. This is
precisely the form of processing that is necessary for expert facial analysis.
Moreover, the STS responds specifically to facial stimuli as the face imparts
social and emotional information, usually through expressions. As discussed
previously, autistics engage in segmental face-processing that ignores the eyes
and nose, thereby extracting little socially- and emotionally-relevant cues
from expressions (Pelphrey et al., 2002). Consequently,
the finding is logical that autistics demonstrate impaired function in the
right STS (for review see Davidson and Slagter,
2000). Research examining the functional and structural differences of the STS
has continued to utilize facial presentations. Hall, Szechtman,
and Nahmias (2003) examined levels of glucose
metabolism using positron emission tomography (PET) in concordance with the
presentation of facial emotions. PET scans record the amount of activity in a
region based on glucose metabolism. A task involving the identification of sex
from emotionally-neutral faces was included to establish which neural regions
respond when visual stimuli provide no emotional information. As expected,
significantly less activation was witnessed in the autistic STS in comparison
to healthy participants. Instead, the thalamus within autistics functioned as
greatly in response to facial presentations as did the STS in the control
group.
The thalamus demonstrates increased activity due to
the segmental face-processing approach utilized in autism. The segmental
approach is neither effortless nor spontaneous (in comparison to the effortless
configural approach of healthy individuals) (Grelotti et al., 2001). Instead, it assembles sensory
information from numerous sources rather than from solely the eyes and nose. In
turn, greater functional strain is imposed on the thalamus, which modulates the
sensory processing system. As previously discussed, autistics presented with
facial expressions depicting one of the six basic human emotions misidentify
the emotion 30% of the time (Pelphrey et al., 2002).
This figure appears low based on the characteristic inability of an autistic to
retrieve facial emotional information. However, autism is not a degenerative
disorder. Therefore, most autistic individuals are capable of learning
compensatory strategies for their emotional and social deficits (NIMH, 2003).
Segmental face-processing, in concordance with greater thalamic activation, is
one such strategy. This strategy is obviously not perfect, for autistics
continue to demonstrate gross impairments in social and emotional interaction
(Hall et al., 2003). However, with practice many are capable of distinguishing
the most basic of emotions: happiness and sadness. Thus, a 30%
misidentification rate denotes the underlying difficulties that autistics
continue to have in deciphering more complex expressions such as surprise and
disgust.
Within autism, the compensatory use of the thalamus
suggests a malfunction in the emotion-linked sensory processing system on which
normal individuals rely. Structural damage of the STS would seem the most
logical conclusion. However, postmortem examinations and Magnetic Resonance
Imaging (MRI) scans have determined that the volume and structure of the STS
are not abnormal within autistic individuals (Rapin
and Katzman, 1998; Sweeten et al., 2002). On the
other hand, numerous studies utilizing neuroimaging
techniques have discovered that the autistic amygdala
and hippocampus are anatomically abnormal (Rapin and Katzman, 1998; Aylward, Minshew, Goldstein, Augustine, & Yates, 1999; Sweeten
et al., 2002; Hall et al., 2003). The amygdala and
hippocampus are two of several structures that comprise the limbic system,
which is implicated in memory, learning, emotion, and motivation. MRI scans of
the autistic amygdala displayed decreased neuronal
size, increased neuronal packing, decreased dendritic
extensions, and small cell bodies (Aylward et al.,
1999). Consequently, amygdala volumes, both before
and after correction for total brain volume, are greatly reduced in autism. The
hippocampus has shown similar patterns of deformity to the amygdala.
Interestingly, children with diseases that damage the limbic system, such as
viral encephalitis, tuberous sclerosis, and cancer, exhibit autistic-like
symptoms (Sweeten et al., 2002). Therefore, limbic system dysfunction is most
likely at the root of autism.
Convergent
research has further demonstrated that amygdala,
rather than hippocampal, abnormality is responsible
for emotionally-relevant face-processing deficits among autistic individuals.
The hippocampus is active during tasks of learning and memory. On the other
hand, motivation, emotional recognition, and displays of aggression elicit amygdala activity. However, PET scans tracing regional
cerebral blood flow (rCBF) among autistics have
failed to find increased amygdala activity in
response to the presentation of facial expressions (Davidson and Slagter, 2000). In particular, healthy individuals demonstrate
greatest activity in response to fear. However, in the study among autistic
individuals who misidentified emotions 30% of the time, an average of 70% of
those misidentifications were between fear and anger (Pelphrey
et al., 2002). Another behavioral characteristic of autism is the bizarre,
erratic display of aggression directed both inward and outward (APA, 1994).
Thus, both the inability to detect fear in facial expressions and the
characteristic random acts of aggression implicate amygdala
abnormality as an underlying factor.
Amygdala response to motivation is also significant because
the ability to expertly discern the emotion of another individual is not
innate. Face-processing, similar to any other task one seeks to master,
requires extensive perceptual and cognitive attention and energy, and is
motivated by the human drive for social interaction (Grelotti
et al., 2002). In healthy individuals, the amygdala
responds to motivating social interactions by releasing the feel-good neurotransmitter dopamine (DA) (Fudge and Emiliano, 2003). DA reinforces behaviors; consequently,
healthy individuals strive to become experts in face-processing. This
motivational drive is largely absent among autistics. One could argue that
segmental face-processing, and thus use of the thalamus, demonstrates an
autistic’s motivation for social interaction. However, as
previously described, segmental face-processing is significantly shorter in
duration than configural-processing. PET scans have
found little to no anatomical abnormalities apparent in autistic attentional systems (Haznedar et
al., 2000). This has led to the conclusion that lack of motivation, rather than
a physical inability to attend, results in shorter processing during facial
presentation (Pelphrey et al., 2002). One theory has
consequently linked autistic lack of motivation to functional impairments of
the DA system (Buitelaar, 2003). However,
investigations into DA levels in autism have demonstrated few consistent
results. Some studies examined DA levels in response to repetitive behaviors
that an autistic individual finds pleasurable. The levels were no different in
autism than in normal individuals, discrediting the possibility of autistic DA
dysfunction. Findings, therefore, are consistent with the idea that amygdala abnormality underlies impairments in the
connection of motivation to social interactions.
The theory that amygdala
deformity subsumes social deficits in autism is further evidenced by research
examining the effects of experimental manipulations, natural diseases, and
injuries (Sweeten et al., 2002; Fudge and Emiliano,
2003). Monkeys are commonly utilized in experimental research due to the
similarity of their neural system to that of humans. One study elicited
over-stimulation in the basolateral nucleus of the amygdala (BLA) using stress-associated peptide corticotropin-releasing factor (CRF) (Sweeten, et al.,
2002). In normal individuals, as CRF levels rise, the BLA is excited, and a
deficit in typical social behaviors is demonstrated. Likewise, in this study
intentional over-stimulation resulted in severe and chronic disruption of
social interaction behaviors. Another study examined the effects of hippocampal versus amygdala
lesions, with striking support for amygdala damage in
concordance with autistic symptoms (Sweeten et al., 2002). Lesions of the
monkey amygdala by two months of age produced
disturbances in social and emotional functioning similar to the patterns seen
in autism. Hippocampal lesions resulted in similar
disturbances; however, by six months of age any deficits had disappeared. This
occurrence conflicts with the stable pattern of deficits in behavior that are
observed among autistic individuals. Thus, the theory that hippocampal
abnormalities result in autistic social dysfunction has been discredited.
In light of the aforementioned evidence, it is clear
that structural abnormality within the amygdala is
the most likely source of social impairments in autism. It must be noted,
however, that not all research has found the presence of amygdala
deformities (Haznedar et al., 2000). MRI scans
compared limbic structure volumes within 17 autistic and 15 control
participants. No differences were noted. Also, a PET scan of rCBF during a verbal learning test found little to no
difference among structural functioning. In comparison to numerous other
studies documenting amygdala abnormality, the MRI
results cannot be explained. The counterfactual PET results, however, appear to
be due to the nature of the task. A verbal learning test would not ordinarily
activate the amygdala (unless, possibly, some
emotionally-relevant words are selected). If anything, previous findings
suggest that the hippocampus instead would be activated, yet in this study it
is not. The present research seeks to produce anatomical abnormalities in the amygdala, and to then record rCBF
during the viewing of facial expressions. By doing so, a direct connection can
be established between the sole presence of amygdala
abnormalities and the impairment in social activities such as facial analysis.
Anatomic organization between the socio-emotional and
the sensory system places the amygdala at the center
of emotional perception in both humans and primates (Lane and Nadel, 2000; Fudge and Emiliano,
2003). The basolateral nuclear group (BLNG), located
within the BLA, is the primary receiving portion of the amygdala.
It is connected to axonal projections from the temporal cortex, including the
STS, the orbital and medial prefrontal cortex (OMPFC), and the hippocampus.
Initial perception of higher-order stimuli, such as facial expression,
activates the STS. The OMPFC determines the relative reward value of the visual
stimuli, with the hippocampus recording this value for future reference. Any
emotionally-relevant information regarding the facial expression is then
projected by each structure to the BLNG. Finally, the BLNG engages in the
deepest socio-affective processing of the facial expression by integration of
the information received.
As discussed previously, MRI and postmortem
examinations have found moderate to severe structural abnormalities in autistic
amygdala (Aylward et al.,
1999). Due to the role that the amygdala is evidenced
to play in social and emotional engagement, the next crucial step is to
determine the connection between its structural abnormality and autistic
impairments. Deformity includes dense neuronal packing, which appears to
explain the other deformities of decreased dendritic
complexity, small cell bodies, and highly complicated axonal pathways.
Consequently, the prevailing theory for autistic amygdala
deformity proposes a deficit in neuronal pruning during early development (Rapin and Katzman, 1998; Aylward et al., 1999). At birth, approximately 20-80% of
neurons in each region are excessive. Neurons compete for synaptic targets,
through which they receive nourishing neurotrophic
factors. Neurons that either receive inadequate amounts of neurotrophic
factor, or are useless due to a lack of synaptic connections, engage in
programmed cell death (apoptosis). During apoptosis, the neuron activates death
genes, which allow caspases to dissolve the chromatin
composition of the neuron. The excessive 20-80% of neurons in each region
engage in programmed cell death. This process is referred to as neural pruning,
the majority of which occurs during the first 10 to 35 weeks of life (for
review see Jacobson, 1991).
The result of dysfunctional apoptosis mirrors the
structural abnormality displayed in the autistic amygdala.
In particular, neuronal packing produces stunted dendritic
arborization, where the dendrites projecting from the
end of axons are too crowded to properly expand. The theoretical consequence of
stunted dendritic arborization
is an interference in messages relayed from connected regions, including the
STS, hippocampus, and OMPFC. This is consistent with previously-described
research that the autistic BLNG fails to integrate emotionally-relevant
information received from other structures. Further, the chronological pattern
of deficit appearance in autism begins as early as 12 months, as evidenced by
impairments in gaze following (Grelotti et al.,
2002). This pattern coincides with the timeline of apoptosis that normally
allows proper, functional synapses to be established by 12 months.
It must be noted that the present hypothesis does not
attempt to explain all the socio-emotional deficits within autism as arising
from amygdala dysfunction. However, the neural system
is integrated in such a manner that the more diffusely abnormal a structure,
the greater the malfunction of connected systems (Rapin
and Katzman, 1998). Therefore, the dysfunction of
even one integral structure (in this case, the amygdala)
can have profound negative effects on activity in the rest of the brain. As
discussed, numerous lines of research support the theory that amygdala abnormality due to faulty neuronal pruning
subsumes autistic impairments in social interaction and emotion. However,
current neuroimaging studies that reveal autistic amygdala abnormality do so in hindsight of the appearance
of deficits (Haznedar et al., 2000). Thus, a crucial
piece of the puzzle is to determine whether structural changes in the amygdala elicit socio-emotional impairments. The
alternative theory (which receives little support) is that behavioral deficits
alter the amygdala structure. However, experimental
manipulations of this theory have yet to be conducted for ethical and practical
reasons. No genetic or environmental marker has been discovered that predicts
the onset of autism. Thus, it has been unethical to clinically interfere with
the development of the amygdala within human
subjects. However, neuronal packing in the amygdala
will only be definitively proven a substrate of autism once science is given
the opportunity to interfere. Interestingly, the application of neurotrophic factor has been demonstrated to significantly
decrease apoptosis (Jacobson, 1991). This crucial point provides a means
through which amygdala abnormality can be
experimentally induced among non-human subjects.
In light of the aforementioned practical and ethical
concerns, experimental manipulation of the macaque monkey (Macaca fuscata) is the next necessary step. The
use of non-human subjects removes numerous extraneous variables that normally
must be considered within theories of autism. Among humans, such variables as
IQ, drug treatment, and varied environments can affect the level of autistic
deficits or neural structures. Manipulations in the macaque are generalizeable to humans due to the fact that the primate
neural structure highly mirrors that of humans (Lane and Nadel,
2000). Specifically, the STS processes socially-relevant information, which is
then communicated through axonal projections to the lateral amygdala.
The amygdala in turn engages in deep emotional
processing of sensory stimuli. Possibly due to the parallel organization of
human and macaque structures, macaques demonstrate numerous socially-interactive
behaviors. They exhibit and act on fear (Lane and Nadel,
2000), respond to others, engage in gaze-following (Ferrari, Kohler, Fogassi, & Gallese, 2000),
and most importantly, discern facial expressions through configural-processing
(Sweeten et al.; 2002). Similar to humans, gaze following and configural-processing abilities increase with age as
socially-interactive behaviors are reinforced (Ferrari et al., 2000). Lesion
studies of monkeys, which examine intentional or incidental abnormalities in
structure, have consequently facilitated an understanding of the role of the amygdala in normal social and emotional processing (Lane
and Nadel, 2000; Sweeten et al., 2002). Amygdala lesions early in life elicited autistic-like
characteristics among macaques, including little eye contact, lack of interest
in face-processing, and decreased display of facial expressions.
These findings are consistent with the theory that amygdala abnormality among both humans and monkeys subsumes
socio-emotional disturbances in autism. Consequently, the present research
seeks to mimic autistic neuronal packing in the BLA portion of the amygdala through the application of neurotrophic
factor. As described previously, neurons compete early in life for nourishing neurotrophic factor (for review see Jacobson, 1991). Those
that receive it fail to engage in apoptosis. Therefore, application of neurotrophic factor greatly increases the number of
surviving neurons, producing dense neuronal packing. It is hypothesized that
later presentations of facial expressions will result in impaired
face-processing, and that the impaired amygdala will
display decreased activation in comparison to control monkeys. It is further
hypothesized that no other structure will display impaired activity because
neuronal packing will be manipulated solely in the amygdala.
Should face-processing be impaired in accordance with only amygdala
impairment, evidence will be established that a lack of neuronal pruning in the
amygdala is responsible for social and emotional
deficits in autism.
Methods
Approximately 8 macaque monkeys (7 weeks of age) will
be used in the proposed experiment. Of the 8, 4 will be experimental and 4 will
be control, with an even number of males and females in each group. From the
outset of the research, guidelines set forth by the Institutional Animal Use
and Care Committee (IAUCC) will be followed for promoting the psychological
well-being of the primates. They will be housed among non-experimental macaques
in a mock-natural environment with the presence of natural visual, auditory,
olfactory, and tactile stimuli. Further, decisions about combining or
separating monkeys will be based on reducing territorial anxiety. All handlers
will be selected based on prior training with primates, as established by the
IAUCC. This enriched mock-environment will ensure that all 8 subjects are
provided the opportunity to develop healthy social and emotional interactive
behaviors.
The 8 macaques will be examined prior to commencement
of the experiment using MRI to ensure normal structural development. Macaques
displaying abnormal development of any structure will be rejected from the
subject pool. At approximately 9.5 weeks of age, the BLA of 4 macaques will be
outlined by 2 separate researchers through examination of the MRI scans. The 4
will be anesthetized. Guided by the location of the outlined amygdala, the macaques are to be injected with neurotrophic factor in order to suppress the occurrence of
apoptosis. After a week has passed, an absolute and a relative amygdala volume will be calculated in cubic centimeters to
determine the extent to which apoptosis has been suppressed. Whether further neurotrophic factor must be later administered (at the end
of 15 weeks) will be determined by comparisons to the relative amygdala volume of controls. Also, other structural volumes
will be compared to relative volumes in controls to ensure no abnormal neuronal
changes, such as packing, have occurred. Subjects that demonstrate abnormal
structural change other than in the amygdala will be
removed from the subject pool.
The project intends to use two forms of data
collection regarding behavioral manifestations of each monkey. The first form
of data will be observational, collected for the subsequent 45 weeks (at the
end of which the monkeys will be 1 year of age). Handlers, blind to the
identity of manipulated versus control monkeys, will record levels of daily
social interaction behaviors using a 5-point scale. For each of 12 behavioral
questions, a score of 5 will represent the most active engagement in social
behaviors. A score of 1 will represent no social behaviors. Points between 1
and 5 will represent a continuum of level of social behavior demonstrated.
Recordings will be made for the dimensions of gaze following, empathy towards
others, and helping behavior. Any stability or changes in social interaction
behaviors will be identified by comparing weekly scores for each monkey to
their scores from the prior week. Interrater reliability
for each score will be assessed through use of the statistical program SPSS.
This phase of data collection is obviously prone to
human bias and error. Therefore, the resulting data will be used solely to
create a socio-emotional profile of behavior for each monkey. Two new raters,
blind to the hypothesis of the experiment as well as the identity of each group
member, will compile the scores of all 45 weeks. Monkeys who score 0-1080 at
the end of the 45 weeks will be classified displaying autistic characteristics.
Scores of 1081-1485 will be deemed borderline autistic characteristics.
Finally, those with a score of 1486-2700 will be labeled displaying normal
characteristics. The classification for each monkey will then be compared with
whether the monkey is a member of the experimental or the control group.
The second form of data collection will utilize neuroimaging and immunocytochemical
techniques. At one year of age, MRI scans will demonstrate whether increased amygdala volume is still present among those monkeys
treated with neurotrophic factor. Control and
experimental subjects will then be presented 12 high-resolution, monochromatic
pictures of monkey faces, each depicting 1 of the 6 different expressions. Each
picture will remain for 7 seconds. The head and body of each monkey will be
comfortably restrained so that only the eyes may move. Eye movement is
monitored by an infrared corneal reflection technique to determine any
differences in face-processing. For each picture, the direction of corneal
attention on facial features will be recorded. Monkeys who demonstrate more
than 50% fixation on non-configural features will be
deemed segmental face-processors. Also, duration of facial-fixation will be
recorded through the corneal reflection technique.
In concordance with this face-processing task,
functional Magnetic Resonance Imaging (fMRI) scans
will record which neural structures are hemodynamically
active (receiving higher levels of rCBF) in response
to the presented emotionally- and socially-relevant facial expressions. FMRI,
like PET, records the changing activity of neural regions. However, the
advantage of fMRI over PET is that no injection is
required in order to witness differences in activity. During a second
presentation of expressions to both groups, the anterograde
tracer biotinylated dextran
amine (BDA) will be injected into the STS, OMPFC, and hippocampus. Research has
demonstrated BDA to provide excellent and abundant labeling of axons and
terminals (Veenman, Reiner,
& Honig, 1992). In other words, BDA, as it is
transported along axons to other neurons, provides excellent contrast for
microscopically viewing axonal pathways. In the experimental group, the
structures to which the tracers are transported will be compared to the healthy
control group to determine if impairments exist in axonal connections
innervating the amygdala.
Findings and Significance
Social and
emotional impairment in interactive behavior is the most characteristic manifestation
of autism (APA, 1994). Numerous lines of research have suggested that
structural abnormalities found in the amygdala,
specifically among the BLA, subsume this characteristic. However, to date
studies have utilized powerful neuroimaging techniques
to examine neural structures in hindsight of observed autistic impairments. The
result of such techniques has demonstrated a strong correlation between
autistic deficits and the presence of neural deformities. The present
hypothesis proposes the crucial next step in determining whether amygdala abnormality due to faulty neuronal pruning
subsumes autistic impairments in social interaction and emotion.
This study will induce structural abnormality
therefore macaque monkeys will be used. Results from a sample of macaque
monkeys are generalizeable to humans because their
neural structure mirrors that of humans. Consequently, it is expected that the
application of neurotrophic factor at 9.5 weeks of
age will suppress apoptosis in macaque monkeys, as it is shown to do in human
neurons. The consequence will be a packing of neurons within the amygdala that parallels the presence of amygdala
deformity demonstrated among autistic individuals. This deformity is expected
to be observable in MRI scans of amygdala volumes
conducted a week after administration of the neurotrophic
factor. It is, however, possible that apoptosis in the amygdala
does not normally occur until the lapse of a specific period of weeks after
birth. Therefore, should MRI scans find relative amygdala
volumes to be equal to that of controls, more neurotrophic
factor will be administered. It is further expected that suppression of
apoptosis will only occur in the amygdala, for the proposed research seeks to eliminate
abnormalities in other neural structures as an explanation for autistic
characteristics.
The organization of the macaque emotional sensory
processing system also mimics that of humans. Therefore, it is hypothesized
that the induced structural abnormalities of macaques will elicit autistic-like
impairments in social and emotional behavior. Scores obtained during the
45-week observational data collection period should reflect these impairments.
Specifically, it is expected that experimental monkeys will demonstrate an
initial decrease in scores from pre-manipulation levels, as well as in
comparison to controls. Further, these monkeys should place into the displaying
autistic characteristics
classification. Each socio-emotional
profile is expected to supplement the concrete data provided by corneal
reflection techniques, and by fMRI scans of rCBF and microvascular
oxygenation during facial expression presentations. During the exposure to
faces, experimental macaques should show segmental facial analysis, as well as
a decreased duration of fixation time. The fMRI scans
should display relatively normal functioning among the affect-linked sensory
system structures, including the STS, OMPFC, and hippocampus. Finally, it is
expected that BDA tracers injected into the STS, OMPFC, and hippocampus will
fail to properly follow the BLA axonal pathways utilized by controls.
The finding that normal sensory processing occurs in
non-amygdala structures, in correspondence with the
existence of autistic-like behavioral impairments, will provide evidence for
the hypothesis that the pathophysiologic underpinning
of autism lies within the amygdala. An inability for
the BLA to axonally receive messages from
higher-order sensory systems, as evidenced by abnormal transportation of BDA
tracers, supports the subsequent hypothesis that neuronal packing impairs the
BLA in integrating information. Experimentally suppressing apoptosis in the
macaque amygdala is the next crucial step in
determining the pathologic underpinning of autism. Such a determination will
subsequently give rise to research regarding the exact origin of autistic
neuronal packing in the amygdala. Unfortunately, a
lack of advancements in neuroimaging and biochemical
techniques currently prevents research from determining whether a genetic
abnormality for apoptosis exists within the amygdala.
One alternative explanation is that an excessive level of neurotrophic
factor is released during early development, suppressing programmed neuronal
death. It is the hope of this proposal that progress in the understanding of
autism will motivate future researchers to develop these more advanced
techniques.
The present research will consequently play a
significant role in the future of diagnosis, treatment, and possible
preventative measures for autism. There is currently no gold standard with
which to diagnose the presence of this disorder (Rapin
and Katzman, 1998). A concrete understanding of its pathophysiology will allow such a standard to be created
based on the presence of specific abnormalities, such as in the amygdala. Due to the fact that a diagnosis of autism often
requires lifelong societal support, strenuous research has focused on curative
treatments. Most medications have sought to alleviate the social and emotional
impairments, such as segmental facial analysis, that hinder social reciprocity
(Werry, 2001). Alternatives have targeted numerous
sources such as neurotransmitter release, and the ability of the amygdala to integrate sensory information (Buitelaar, 2003). The latter treatment, a synthetic
acetylcholine analogue called Org 2766, was found to facilitate significant
improvements in eye-contact and social reciprocity. However, as with any other
drug, Org 2766 has emotionally-aggravating side-effects such as agitation and
irritability. It is clear that the future of research, including the present
proposal, must elucidate the physiological abnormalities within autism. Only
with such an understanding can preventive measures begin to fight such drastic
and heartbreaking impairments as an inability to draw social and emotional
clues from a facial expression.
References
American Psychiatric Association. (1994). Diagnostic and Statistical
Manual of MentalDisorders,
Fourth Edition. Washington, DC:
American Psychiatric Association.
Aylward, E.H., Minshew, N.J.,
Goldstein, G., Honeycutt, N.A., Augustine, A.M., & Yates, K.O.. (1999). MRI volumes of amygdala and hippocampus in non-mentally retarded autistic
adolescents and adults. Neurology, 53(9), 2145-2150.
Buitelaar, J.K. (2003). Why have drug treatments been so
disappointing? Novartis Foundation Symposium: Vol. 251. Autism:
neural basis and treatment possibilities (pp.235-249). Wiley: Chichester.
Davidson, R.J., & Slagter,
H.A. (2000). Probing emotion in the developing brain: functional neuroimaging in the assessment of the neural substrates of
emotion in normal and disordered children and adolescents. Mental Retardation and Developmental Disabilities Research Reviews, 6,
166-170.
Ekman, P. (1992). An argument for basic emotions. Cognition and Emotion, 6, 169-200.
Ferrari, P.F., Kohler,
E., Fogassi, L., & Gallese, V. (2000). The
ability to follow eye gaze and its emergence during development in macaque
monkeys. Proceedings of the National
Academy of Science of the United States, 97(25), 13997-14002.
Fudge, J.L., & Emiliano,
A.B. (2003). The extended amygdala and the dopamine
system: another piece of the dopamine puzzle. Journal Neuropsychiatry Clinical Neuroscience, 15(3), 306-316.
Grelotti, D.J., Gauthier, I., & Schultz,
R.T. (2002). Social interest and
the development of cortical face specialization: what autism teaches us about
face processing. Developmental
Psychobiology, 40, 213-225.
Hall, G.B.C., Szechtman, H.,
& Nahmias, C. (2003). Enhanced salience and
emotion recognition in autism: a PET study. American
Journal of Psychiatry, 160(8), 1439-1441.
Haznedar, M.M., Buchsbaum,
M.S., Wei, T.C, Hof, P.R.,
Cartwright, C., & Bienstock, C.A. (2000). Limbic circuitry in patients with autism spectrum
disorders studied with positron emission tomography and magnetic resonance
imaging. American Journal of
Psychiatry, 157(12),
1994-2001.
Iacoboni, M., Koski, L.M., Brass,
M., Bekkering, H., Woods, R.P., Dubeau,
M.C., et al. (2001). Reafferented copies of imitated
actions in the right superior temporal cortex.
Proceedings of
the National Academy of Science of the United States. 98(24),
13995-13999.
Jacobson, M. (1991). Developmental neurobiology: third edition.
New York City, NY: Plenum Press.
Lane, R.D, & Nadel, L.
(2000). The role of the amygdala in primate social
cognition. New York City, NY: Oxford University Press.
National Institutes of Mental Health. (2003). NIH awards grants for six new autism research centers. Retrieved October 22, 2003, from http://www.nimh.nih.gov/events/prautismcenters.cfm
.