Intended for healthcare professionals

Clinical Review

Diagnosis of autism

BMJ 2003; 327 doi: (Published 28 August 2003) Cite this as: BMJ 2003;327:488

Top down modulation and apical dendrites: possible links with the pathogenesis of autism

The case of SM, a women with a selective bilateral damage to the
amygdala and to a small portion of adjacent enthorinal cortex [1],
suggests that the amygdala has a role in guiding visual attention to
sensible cues in human faces. SM fails to direct her gaze toward the eyes
region of faces, when she explores facial expressions, but her recognition
of fear in faces is similar to controls when she is actively prompted to
explore that region.
It would be oversimplistic to directly associate the gazing behavior of SM
with the deficit in visual fixation shown by persons with autism, as it
would be simplistic to ascribe the whole of the autistic symptomatology to
a frontal dysfunction. SM shows an array of social impairments but she is
not autistic, and frontal patients are not autistic either, although they
may fail in executive tasks, by definition, or in mindreading tasks.
The point I would like to rise here is related to the guiding role of the
amygdala in selective attention.

Attention has been defined as “an enhancement of activity in the
attended set of pathways relative to the unattended set’’ [2]. Recent
studies have shown that higher neural centers can influence lower centers
when the latter are processing either external, or internal data. Bar [3]
proposed that feedback signals from the prefrontal cortex to lower visual
pathways have a role in the rapid recognition of percepts. Mechelli et al
[4] found that the prefrontal cortex activates category-selective zones of
the extrastriate cortex during visual imagery. Ishai et al [5] have
demonstrated that the brain areas that are activated during tasks of
visual imagery and rehearsal of famous faces are the same which are
activated during the real perception of the same targets, albeit in a less
extensive way.

It is plausible that both the limbic system and the prefrontal cortex
send signals to visual pathways at various levels in order to make
priorities among perceptive data, creating contrasts between contextually
significant and less relevant perceptive elements. Our subjective and
adaptive mirroring of the environment is probably actively shaped by top-
down processes, or feedbacks [6] realizing a parallel system of
interconnections between neural structures.

Top-down processes are seemingly involved in imagery, working memory
and attention. While visual pathways are the most studied system as far as
feedbacks are concerned, it is reasonable to assume that an analogous
modulation of brain functions takes place in other cortical (and possibly
subcortical) areas.

Our everyday experience provides useful hints, in this regard. While
we are able to perceive two or more different voices at the same time,
only one of these voices can be fully processed for comprehension.
Consider now the following situation: I am listening to an cognitively
interesting news bulletin when a second, familiar voice enters my
perceptive field. If the voice has a neutral content, most of the times I
will be able to automatically keep that voice in a unattended position in
my cognitive apparatus. But if the voice has an affective meaning for me,
it will probably displace the first attended track (the news) and gain its
way upward the hierarchy of neural computation. I will process that voice
in order to understand the messages it conveys and what is the most
appropriate action to take in response. I may still try to refocus on the
news, and be successful in that task, but only if the affective resonance
of the second voice is not particularly strong. I presume that in order to
gain prominence, the added voice has been recognized by my limbic system,
while my struggling to keep on listening to the news might be the
consequence of an activation of the prefrontal cortex. There is a
variability of responses, of course, and I suspect that, on the average,
men would be more prone to be captured by news (or football, according to
preferences) and women would be more ready to respond to the voice of a
close relative. Moreover, attention to football could be directed by
signals from the limbic system in men.

Frith [quoted by Hill and Frith, 7] has hypothesized that the pruning
of top-down synapses is delayed in autism, leading in some cases to an
enlarged brain, an executive dysfunction and a perceptual overload.

Brock et al [8], have recently proposed the “Temporal binding model
of autism”: they maintain that in autism there is an impairment in the
synchronization between distant brain organizations, that this could be at
the basis of a “weak central coherence” and that it could explain many of
the cognitive symptoms of autism.
A basic endophenotype of autism might be a reduction in the amplitude of
synchronously elicitable mental function with relative preservation of
discrete processing [9].

I would extend Brock et al’s line of reasoning, and speculate that
the whole system of top-down modulations might be a dysfunctional system
in at least some cases of autism. As a consequence, the underconnectivity
between brain areas which has been recently detected in autistic
individuals [10] might be a modulatory underconnectivity, a failure of
meta-connectivity. In physiological conditions, the information coming
from sensory organs and from the activation of processing areas in the
absence of external stimuli (id est imagery, speculative activities and
maybe dreaming) would be coordinated by a well functioning top-down
modulatory system. In severe autism, on the other hand, all sensory data
would have equal opportunity, or on the contrary a reduced opportunity to
be processed by distributed and potentially dedicated, but inconsistently
activated brain circuitries.

A failure in the synchronization of task-oriented neural webs might
easily explain the social deficits of persons with autism, as a reduced
power of computation of incoming complex stimuli would render other
peoples’ actions, intentions and ambiguous behaviors impossible to decode
with the necessary speed. Given the complexity of social relationships,
individuals with autism would prefer isolation to the frustrating
experience of interpersonal unfitness.

Spratling [11] has reviewed the role of apical dendrites in the
regulation of the response of pyramidal neurons to “feedforward” signals,
id est signals coming from the periphery of the nervous system and
ascending toward higher levels of neural computation. His model gives a
logical account of the more active role in perception, speculation and
action planning that human brain has been deemed as endowed with by many
researchers in recent years. I suggest that a role of apical dendrites and
related synapses should be considered when studying the basic dysfunction
of autism. An in-built, distributed system of modulation of brain
functions, with important nodes of activity in the limbic system, in the
prefrontal and parietal cortex and possibly in the cerebellum might be
dysfunctional in this disorder. Fatemi et al [12] have found a reduction
of Reelin, a glycoprotein responsible for the normal layering of brain in
superior frontal and cerebellar cortices of brains of persons with autism.
Mukaetova-Ladinska et al. [13] detected a reduction of MAP2 protein
associated with cytoarchitectonic changes in the dorsolateral prefrontal
cortex of individuals with autism. If a link between these
neurobiological, genetic, imaging and behavioral findings will be
confirmed, then new light may be shed on some pathogenic ways of autism
and useful elements might be added to treatment guidelines. If the deficit
in autism is one of undersynchronization and underorganization of
cognitive functions, then addressing to the globality of the child and her
hidden assets in a nearly spontaneous setting and in a period in which her
brain is very plastic should have more chance of success than
systematically drill on discrete areas of competence.


1. Adolphs R, Gosselin F, Buchanan TW, Tranel D, Schyns P, Damasio
AR. A mechanism for impaired fear recognition after amygdala damage.
Nature. 2005;433:68-72.

2. LaBerge, D. Attentional Processing. 1995; Harvard Univ. Press,
Cambridge, MA.

3. Bar M. A cortical mechanism for triggering top-down facilitation
in visual object recognition. J Cogn Neurosci. 2003;15:600-9. Review.

4. Mechelli A, Price CJ, Friston KJ, Ishai A. Where bottom-up meets
top-down: neuronal interactions during perception and imagery. Cereb
Cortex. 2004;14:1256-65.

5. Ishai A, Haxby JV, Ungerleider LG. Visual imagery of famous faces:
effects of memory and attention revealed by fMRI. Neuroimage. 2002;17:1729

6. Engel AK, Fries P, Singer W. Dynamic predictions: oscillations and
synchrony in top-down processing. Nat Rev Neurosci. 2001;2:704-16. Review.

7. Hill EL, Frith U. Understanding autism: insights from mind and
brain. Philos Trans R Soc Lond B Biol Sci. 2003;358:281-9. Review.

8. Brock J, Brown CC, Boucher J, Rippon G. The temporal binding
deficit hypothesis of autism. Dev Psychopathol. 2002;14:209-24.

9. Loddo, S.“ Autism: multiple genes acting on a distributed neural
target?” Canadian Journal of Psychiatry. 2004;49: 219-220. Letter.

10. Just MA, Cherkassky VL, Keller TA, Minshew NJ. Cortical
activation and synchronization during sentence comprehension in high-
functioning autism: evidence of underconnectivity. Brain 2004;127:1811-21.

11. Spratling MW. Cortical region interactions and the functional
role of apical dendrites. Behavioral and Cognitive Neuroscience Reviews.

12. Fatemi SH, Snow AV, Stary JM, Araghi-Niknam M, Reutiman TJ, Lee
S, Brooks AI, Pearce DA. Reelin signaling is impaired in autism. Biol
Psychiatry. 2005;57:777-87.

13. Mukaetova-Ladinska EB, Arnold H, Jaros E, Perry R, Perry E.
Depletion of MAP2 expression and laminar cytoarchitectonic changes in
dorsolateral prefrontal cortex in adult autistic individuals. Neuropathol
Appl Neurobiol. 2004;30:615-23.

Competing interests:
None declared

Competing interests: No competing interests

17 April 2005
Silvio Loddo
Child and Adolescent Psychiatrist
Neuropsichiatria dell'Età Evolutiva, AUSL n. 5 Oristano, Piazza S. Martino, 09170 Oristano