The mechanisms of autism are the molecular and cellular processes believed to cause or contribute to the symptoms of autism. Multiple processes are hypothesized to explain different autism spectrum features. These hypotheses include defects in synapse structure and function,^[1][2] reduced synaptic plasticity,^[3] disrupted neural circuit function, gut–brain axis dyshomeostasis,^[4][5][6] neuroinflammation,^[7] and altered brain structure or connectivity.^{[8][9][10][11]}

Pathophysiology

Unlike some brain disorders which have clear molecular hallmarks that can be observed in every affected individual, such as Alzheimer's disease or Parkinson's disease, autism does not have a unifying mechanism at the molecular, cellular, or systems level. The autism spectrum may comprise a small set of disorders that converge on a few common molecular pathways, or it may be a large set of disorders with diverse mechanisms.^[13] Autism appears to result from developmental factors that affect many or all functional brain systems.^[14] Some factors may disturb the timing of brain development rather than the final product.^[12]

Brain growth

Neuroanatomical studies and the association between autism and teratogens strongly suggest that autism affects brain development soon after conception.^[15] This anomaly appears to start a cascade of pathological events in the brain that are significantly influenced by environmental factors.^[16] Just after birth, the brains of children with autism tend to grow faster than usual, followed by normal or relatively slower growth in childhood.^[17] It is unknown whether early brain overgrowth occurs in all children with autism. It appears to be most prominent in the frontal and temporal lobes, which are associated with higher cognitive specializations such as social cognition, and language development.^[18] Hypotheses for the cellular and molecular bases of pathological early overgrowth include an excess of neurons that causes local overconnectivity in key brain regions,^[17] and disturbed neuronal migration during early gestation.^[19][20]

Synapse dysfunction

Synapse and dendritic spine growth may be disrupted in autism due to impaired neurexin–neuroligin cell-adhesion signaling^[21] or dysregulated synthesis of synaptic proteins.^[22][23] Disrupted synaptic development may also contribute to epilepsy, which may explain why the two conditions are associated.^[24]Studies have suggested that excitatory–inhibitory networks may be imbalanced in autism.^[20]

Neurotransmitters such as serotonin, dopamine, and glutamate have been implicated in autism.^[1] Fragile X, the most common genetic cause of autism, is linked to dysfunction of group I metabotropic glutamate receptors (mGluR), leading some to consider their potential role in autism.^[25]

Altered circuit connectivity

The underconnectivity theory of autism posits that autistic people tend to have fewer high-level neural connections and less global synchronization, along with an excess of low-level processes.^[27] Functional connectivity studies have found both hypo- and hyperconnectivity in brains of autistic people.^[28] Hypoconnectivity is commonly observed for interhemispheric and cortico-cortical functional connectivity.^[29] Some studies have found local overconnectivity in the cerebral cortex and weak functional connections between the frontal lobe and the rest of the cortex.^[30] Abnormal default mode network (task-negative) connectivity is often observed. Toggling between task-negative network activation and task-positive network activation (consisting of the dorsal attention network and salience network) may be less efficient, possibly reflecting a disturbance of self-referential thought.^[31] Such patterns of low function and aberrant activation in the brain may depend on whether the brain is performing social or nonsocial tasks.^[32]

Some studies have suggested that autism is a disorder of the association cortex.^[33] Event-related potentials with respect to attention, orientation to auditory and visual stimuli, novelty detection, language and face processing, and information storage are altered in autistic individuals; several studies have found a preference for nonsocial stimuli.^[34] Magnetoencephalography studies have observed delayed processing of auditory signals in autistic children.^[35]

The mirror neuron system (MNS) theory of autism hypothesizes that disrupted development of the MNS impairs autistic people's ability to imitate others, leading to core autistic features of social impairment and communication difficulties. In animals, the MNS activates when an animal performs an action or observes another animal perform the same action. The MNS may contribute to an individual's understanding of other people by enabling the modeling of their behavior via embodied simulation of their actions, intentions, and emotions.^{[36] [37]} Several studies have tested this hypothesis by demonstrating structural abnormalities in MNS regions of individuals with ASD, delay in the activation in the core circuit for imitation in individuals with ASD, and a correlation between reduced MNS activity and severity of the syndrome in children with ASD.^[38] However, individuals with autism also have abnormal brain activation in many circuits outside the MNS^[39] and the MNS theory does not explain the normal performance of children with autism on imitation tasks that involve a goal or object.^[40]

Common copy number variation associations have suggested similarities between the mechanisms of autism and schizophrenia. For loci such as 16p11.2, 16p13.1, 22p11, and 22q13, deletion is associated with autism whereas duplication is associated with schizophrenia. Conversely, 1q21.1 and 22p11.2 duplication is associated with autism and deletion with schizophrenia.^[41]

Inflammation

The immune system is thought to play an important role in autism. Children with autism have been found by researchers to have inflammation of both the peripheral and central immune systems as indicated by increased levels of pro-inflammatory cytokines and significant activation of microglia.^[42][43][7] Biomarkers of abnormal immune function have also been associated with increased impairments in behaviors that are characteristic of the core features of autism such as, deficits in social interactions and communication.^[43] Interactions between the immune system and the nervous system begin early during the embryonic stage of life, and successful neurodevelopment depends on a balanced immune response. It is thought that activation of a pregnant mother's immune system such as from environmental toxicants or infection can contribute to causing autism through causing a disruption of brain development.^[44][45][46] This is supported by recent studies that have found that infection during pregnancy is associated with an increased risk of autism.^[47][48]

Some evidence suggests that gut–brain axis abnormalities may be involved by means of impaired serotonin signaling and inflammation.^[6] A 2015 review proposed that immune dysregulation, gastrointestinal inflammation, autonomic nervous system malfunction, gut microbiota alterations, and food metabolites may cause brain neuroinflammation and dysfunction.^[4] A 2016 review concluded that enteric nervous system abnormalities might play a role in neurological disorders such as autism.^[5]

Metabolism

Some data suggests neuronal overgrowth observed in autism may be caused by an increase in several growth hormones^[49] or impaired regulation of growth factor receptors. Some inborn errors of metabolism are associated with autism, but probably account for less than 5% of cases.^[50]

Neuropsychology

Two major categories of cognitive theories have been proposed to explain links between autistic brains and behavior.

Social cognition

The first category focuses on deficits in social cognition. Simon Baron-Cohen's empathizing–systemizing theory postulates that autistic individuals can systemize—that is, they can develop internal rules of operation to handle events inside the brain—but are less effective at empathizing by handling events generated by other agents. An extension, the extreme male brain theory, hypothesizes that autism is an extreme case of the male brain, defined psychometrically as individuals in whom systemizing is better than empathizing.^[51] These theories are somewhat related to Baron-Cohen's earlier theory of mind approach, which hypothesizes that autistic behavior arises from an inability to ascribe mental states to oneself and others. The theory of mind hypothesis is supported by the atypical responses of children with autism to the Sally–Anne test for reasoning about others' motivations,^[51] and the mirror neuron system theory of autism described in Pathophysiology maps well to the hypothesis.^[38] However, most studies have found no evidence of impairment in autistic individuals' ability to understand other people's basic intentions or goals; instead, data suggests that impairments are found in understanding more complex social emotions or in considering others' viewpoints.^[52]

Nonsocial cognition

The second category focuses on nonsocial or general processing: the executive functions such as working memory, planning, inhibition. In his review, Kenworthy states that "the claim of executive dysfunction as a causal factor in autism is controversial", however, "it is clear that executive dysfunction plays a role in the social and cognitive deficits observed in individuals with autism".^[53] Tests of core executive processes such as eye movement tasks indicate improvement from late childhood to adolescence, but performance never reaches typical adult levels.^[54] A strength of the theory is predicting stereotyped behavior and narrow interests;^[55] two weaknesses are that executive function is hard to measure^[53] and that executive function deficits have not been found in young children with autism.^[56]

Weak central coherence theory

Weak central coherence theory hypothesizes that a limited ability to see the big picture underlies the central disturbance in autism. One strength of this theory is predicting special talents and peaks in performance in autistic people.^[57] A related theory—enhanced perceptual functioning—focuses more on the superiority of locally oriented and perceptual operations in autistic individuals.^[58] Yet another, monotropism, posits that autism stems from a different cognitive style, tending to focus attention (or processing resources) intensely, to the exclusion of other stimuli.^[59] These theories map well from the underconnectivity theory of autism.

Issues with categories

Neither category is satisfactory on its own; social cognition theories poorly address autism's rigid and repetitive behaviors, while most of the nonsocial theories have difficulty explaining social impairment and communication difficulties.^[60] A combined theory based on multiple deficits may prove to be more useful.^[61]