The prevalence of autism spectrum disorders (ASD) is increasing and it is alarming. In fact, 1 in 88 newborns is currently affected by ASD in the USA, with a specific male dominance (5:1 ratio of males to females). The presence of a child with ASD in the family causes an overall impact on parents and siblings manifested in a significant increase in stress.

Only about 10 % of patients with a diagnosis of ASD have a defined aetiology (so-called syndromic autism, secondary to fragile X syndrome, neurofibromatosis, exposure to thalidomide), while 90 % of ASD cases are considered idiopathic (i.e. without a definite aetiological agent). In these cases, genetic factors are known to be relevant. However, genetics only cannot explain the rapid increase of the prevalence observed in recent years. For this reason, most authors consider that the aetiology of ASD lies in an interaction between genetic and environmental factors. The genes that have so far been identified seem to have an important role to play in brain development, particularly in synaptic development. The results of the screening of the human genome have shown the presence of different susceptibility loci on different chromosomes, in particular the 2, 16 and 17. The data obtained suggest the interaction of at least ten genes in the aetiology of autism . In practice, the strongest evidence is that there is no ‘one gene’ of autism, but quite a number of genes that contribute to a vulnerability to the onset of the disorder. There are many challenging environmental factors that appear to play a role in autism. It is unclear how these environmental factors (including vaccines) may interact with the individual genetic risk for developing autism . Research has not yet identified a direct causal link between any environmental factor and autism.

The term autism was coined in 1908 by the Swiss psychiatrist Bleuler to refer to a symptom of schizophrenia, pointing to it with the extreme narrowing of relations with the outside world that excludes anything except your own self: it derives from the Greek αυτός, or “if the same”.

The first scientific description of autism—a psychopathological syndrome in itself—is due to the German-American physician Leo Kanner in 1943, in the article “Autistic Disturbances of Affective Contact” published in The Nervous Child. He described eleven children prone to isolation, “self-sufficient”, “happy if left alone”, “as in a shell”, not very reactive in the relational field (Paucheri and Pfanner 1999) . Some appeared functionally silent or had echolalic speech, while others showed a characteristic inversion pronoun (“you” to refer to themselves and “I” to refer to another). Many had an obsessive fear of some change happening in the surrounding environment, while some had very specific skills, developed and isolated, but close to a general delay. Observing these traits and behaviours as typical but hitherto unknown, Kanner described this syndrome as “Early Infantile Autism”. Although experts in autism have been the first to recognize, under the same name, a clinical condition characterized by precise impaired social interaction, impaired communication and the constant presence of repetitive, stereotyped and restricted interests, and have changed the diagnostic and therapeutic approach over time, there are still large areas of uncertainty, doubt and problem.

In fact, numerous efforts to understand the causes of this syndrome and various assumptions have been made; the multifactorial origin, which needs the interaction of genetic and environmental factors, seems however to be the most reliable.

Autism is a behavioural syndrome caused by a developmental disorder that is biologically determined, with onset in the first 3 years of life. The areas relating to social interaction, the capacity to communicate ideas and feelings and the ability to establish relationships with others are mainly affected; therefore, it appears as a “permanent” disability that accompanies the subject through his lifetime, although the characteristics of the social deficit assume variable expressivity in time.

According to the definition of the American Psychiatric Association, significantly abnormal or deficient social interaction, impaired communication and language abilities and a considerably narrow pattern of activities and interests are the mainstays of the ASD diagnosis .

ASDs, in the DSM-V, are defined within two categories: “persistent impairment in reciprocal social communication and social interactions” and “restricted and repetitive patterns of behaviour,” both present since early childhood. In the DSM-V diagnosis must meet the following criteria A, B, C and D:

The diagnosis also requires the specification of the presence or absence of related intellectual disability, alterations of language as well as medical conditions or associated genetics.

The prevalence of ASD is increasing and the current prevalence rate is alarming. In fact one in 88 newborns is currently affected by ASD in the USA, with a specific male dominance (5:1 ratio) between males and females (Baio 2012). However, a recent study performed in the UK suggests lower figures (annual incidence rate of about 1.2/1000 boys and 0.2/1000 girls). Only about 10 % of patients with a diagnosis of ASD have a defined aetiology (the so-called syndromic autism, previously defined; Engel and Daniels 2011; Stagno and Whitley 1985) , while 90 % of ASD cases are considered idiopathic (i.e. without a definite aetiological agent) (Gentile et al. 2013) . In these cases, genetic factors are known to be relevant. However, genetics cannot explain the rapid increase of the prevalence observed in recent years. For this reason, most authors consider that the aetiology of ASD lies in an interaction between genetic and environmental factors (Garbett et al. 2012; Shi et al. 2003) . The absence of a definite causative agent makes the set-up of preventive measures for ASD impossible.

In the 1940s, Kanner described how the parents of his patients with autism were able to identify a characteristic common to all of them: that of being “cold and intellectual”. Afterwards, in his theory of psychoanalysis, Bettelheim affirmed that “the refusal on the part of the parents is a key element in the genesis of each observed case of autism” (Bettelheim 1967) . In practice, he opened the way to what has been, for decades, the cornerstone of the interpretation of the onset of autistic disorder, namely the theory of the “refrigerator mother”. Emphasizing the absence of an organic background and the particular type of parenting, autism was interpreted as a defense against anxiety, arising from a failure of the first object relations. This model, however, was not broadly accepted everywhere, and in the Anglo-Saxon countries psychoanalytic hypotheses never found wide scientific credit. Bernard Rimland, an American psychologist father of an autistic child, revolutionizes the approach to this disease, arguing that the problem was not to be found in the “parents” of the child, but rather in an organic cause: it was 1960. Since then different theories have started being processed; among them the socio-affective theory of mind and the executive function theory; the input to the search of a matrix within the neurobiological disorder began to materialize .

The neurobiological research in the area of ASD starts with the “Theory of Socio-Affective” (Hobson 1993) . This research area was aimed at defining the characteristics of mental functioning autistics, from which the behaviours that characterize the clinical picture should be derived. According to this theory, the autistic child has an innate inability, biologically determined (the existence of brain structures responsible for processing of social stimuli are assumed), to interact with other people emotionally. The absence of this empathy is not referential (Hobson and Lee 1989) or primary inter-subjective ( Trevarthen and Aitken 2001) : it is due to a sort of cascade reaction leading to an inability to learn and/or recognize the mental states of others, to impaired symbolization, to the deficit of language and of social interaction .

Other assumptions in this research direction limit their attention to the so-called central coherence and executive function. Central coherence refers to the ability to synthesize into a coherent whole the various experiences that affect senses in a fragmented way (Frith and Happé 1994) . A weakness of this aspect would therefore be to remain at parceled experiential data, not allowed to grasp the ultimate meaning of each stimulus, just as in autism. The concept of executive function is also very similar to the previous (Bennetto et al. 1996) . Indeed, there are skills that are crucial in the organization and planning of behaviours for the resolution of problems. This refers to the ability to activate an area of mental work, mentally formulate a plan of action, not to be rigidly anchored in the perceptual data, to inhibit impulsive responses and to be attentive to the feedback information in order to move flexible attention to the various aspects of the context. It is clear that a deficiency of these aspects leads precisely to events like impulsiveness, inability to inhibit appropriate responses, hyper-selectivity, remaining anchored to detail and perseverance: all these aspects are peculiarly close to autism .

The theory of mind (Baron-Cohen et al. 2000) was formulated in the eighties; it assumes that the child begins to acquire the ability to reflect on emotions, desires and beliefs about self and others at 4 years of age. If this does not happen, the child does not gain the ability to predict the behaviour of others, and will not be able to accomplish the level of “meta-representations” (does not know how to think that other people think and what they think). This translates into hyper-selectivity, repetitiveness, rigidity and perseveration and will possibly affect the area of social interaction and communication. The recent discovery of the mirror system (Cattaneo et al. 2007) suggests a “physiological” explanation to this theory. The mirror system is a neural system present in humans and non-human primates useful for understanding imitation and the intentions of others. The organization of intentional motor chains in children with typical development and with autism, by means of electromyography (EMG) was investigated; it was demonstrated that the intentional chains are altered in autistic children (Cattaneo et al. 2007) . The high-functioning autistic children understand the intentions of others cognitively, but lack a mechanism allowing direct and immediate experiential understanding of the action. High functioning autism is, in fact, characterized by preserved intelligence and general cognitive functioning accompanied by deficits in social interaction and communication, as well as repetitive and restrictive behaviour .

As a matter of fact, the causes of ASD are still unknown. Other hypotheses include genetics, obstetric complications, infection and toxic exposures (Larsson et al. 2005; Lawler et al. 2004; Constantino et al. 2010) . None of these, however, has been established as a definite aetiological agent.

Familial factors influence the risk for autism spectrum disorder . The rate of ASD in children born into families that already have an affected child is as high as 18.7 %, and the risk is twice as high in children born into families with two or more affected children (Ozonoff et al. 2011) . Girls born into a family that has a sick child with an ASD will have 2.8 more times the risk of having such a disorder (Ozonoff et al. 2011) . Twin studies have demonstrated a moderate degree of genetic heritability for ASD (Abrahams and Geschwind 2008, 2010; Hallmayer et al. 2011) , with environment making a substantial contribution to the development of these conditions in the study subjects (Hallmayer et al. 2011) . It was noticed that identical twins have a higher probability (concordance rate higher than 60 %) of both being autistic compared to non-identical twins, and that heterozygous parents of an affected child have a recurrence rate of having a second child with an ASD that varies between 2 and 8 %, in contrast to the prevalence in the general population that is around 3 %. Family members of patients who have similar though faded behavioural characteristics were also identified. Finally, the presence of a secondary or syndromic autism associated with genetically determined diseases such as Tuberous Sclerosis, Fragile X Syndrome, and Neurofibromatosis is significant, thus confirming the influence of the genetic component in the onset of the disorder (Muhle et al. 2004) .