WO2011027060A2 - Oxytocin treatment for behavioral characteristics associated with autism and shyness disorder - Google Patents

Abstract

The present invention relates to the use of oxytocin or an analog thereof for preventing or treating one or more behavioral characteristics associated with disorders of the autism spectrum (or autism) and shyness disorder.

Description

 OCYTOCINE TREATMENT OF BEHAVIORAL CHARACTERISTICS ASSOCIATED WITH AUTISM AND TIMIDITY

 PATHOLOGICAL DESCRIPTION

Technical area

 The present invention relates to the prevention or treatment of behavioral characteristics associated with autism (or autism spectrum disorder or ASD) and pathological shyness via the administration of oxytocin or an analogue thereof, preferably intranasally.

 In the description below, references in brackets ([]) refer to the list of references at the end of the text.

State of the art

 The human brain has the ability to detect and respond to simple social cues such as eye contact, as well as infer from more complex behaviors, the inherently social qualities of others such as equity or cooperation.

Autism refers to behavioral disorders that manifest themselves in three main areas (the "autistic triad"): verbal and / or nonverbal communication disorders, social interaction disorders, restricted interests and / or behaviors repetitive. Autism is a big handicap and has serious consequences on the family, social and professional levels. Autism under the general label of "Autism Spectrum Disorder" (ASD) usually includes (i) typical "Kanner" autism or infantile autism, (ii) Asperger syndrome, and (iii) "high-level" autism (HFA), also defined as autism without intellectual disability, applied to autists deemed to have "higher functioning" than other autistics, by one or more indicators. Thus HFA individuals misunderstand social signals and also respond poorly. However, HFA individuals generally have normal general intellectual abilities or normal language, although they avoid visual contact in daily life and do not interact spontaneously with others [Klin et al., Arch. Gen. Psychiatry, 59 (9): 809-816, 2002] [1]. On formal tests of social cognitive capacity, they have specific deficits in the understanding of the intentions of others [Baron-Cohen et al., J. Child Psychol. Psychiatry, 42: 241-251, 2001; Frith and Miller, J. Child Psychol. Psychiatry, 45: 672-686, 2004] [2,3], and are not able to make quick and intuitive decisions about social contexts [Klin et al., 2002, supra] [1].

The pathogenesis of autism is unclear although mutations in genes involved in synaptogenesis have been identified [Durand et al., Nat. Genet., 39: 25-27, 2007; Jamain et al., Nat. Genet., 34: 27-29, 2003] [4,5] and that various neurochemical, neurophysiological, and neuropathological abnormalities have been demonstrated in these patients [Waterhouse et al., Psychol. Rev., 103: 457-489, 1996] [6]. Some researchers have found that the brains of parents, brothers and sisters of at least one child with autism share some abnormalities with autism, namely that they had on average the motor areas and basal ganglia the brain is larger than normal and the cerebellum and frontal lobe smaller; a pattern generally associated with autism [Peterson et al., NeuReport, 17 (12): 1289-1292, 2006] [7]. These observations seem to support the observations of other researchers who have noted that the brothers and sisters of young autistic, although not presenting the least symptom of autism, however, had difficulty in fixing their interlocutor eyes, showing timidity almost pathological [Dalton et al., Biol. Psychiatry, 61 (4): 512-520, 2007] [8]. Pathological shyness affects healthy individuals with a low score for extraversion and a high score for neuroticism, which may contribute to susceptibility to disorders, according to the Revised Personality Inventory (NEO Personality Inventory Revised or NEO-PI-R). mood and anxiety. The international NEO-PI-R questionnaire is based on the idea that personality traits are organized hierarchically. It is the most influential model of the human personality, evaluating the five main dimensions of personality or domains (neuroticism, extroversion, openness, agreeableness and consciousness) and the most important traits of these dimensions also called "facets". Extraversion captures the social aspect of personality and includes six facets (warmth, gregariousness, assertiveness, activity, search for sensations, and positive emotions). Individuals with a high degree of extraversion have a preference for social interaction research and engagement, whereas introverts prefer to avoid social situations and tend to be reserved. Neuroticism (or neuroticism) is characterized by a great sensitivity to all that is negative [Costa et al., J. Pers. Assess., 57 (3): 393-398, 1991] [9] and comprises six facets (anxiety, anger-hostility, depression, social timidity, impulsivity, and vulnerability). Individuals with a high degree of neuroticism are more likely to rate situations as threatening than those with low levels. In our modern society, there are more and more individuals with pronounced introversion and shyness in interactions with others, whether in broad or limited social contexts. These individuals generally suffer from being unable to overcome their shyness to interact socially with others, and most of the time require medical help (ie the administration of drugs such as antidepressants).

Currently, there is no treatment per se of autism or disorders with similar behavioral characteristics. So there are no drugs treating autism itself, but drugs that can have a positive effect on his behavior. For example, risperidone reduces aggression, hyperactivity and repetitive behaviors, clozapine and quetiapine have produced positive effects but also significant side effects that require very rigorous monitoring of treatment. Studies on olanzepine and amisulpride are encouraging in terms of improving social behavior, language, decreased aggression, and reducing withdrawal and inhibitions, respectively, but still require further research. The means of treatment of autism are therefore mainly of a palliative nature, namely the management of autism through personalized behavioral therapies, generally from the age of 18 months and for 2 years, in order to improve the symptoms. .

Thus, it appears that social interactions affect every aspect of our lives, from taking care of our children to determining our success in life. Abnormal manifestations of social behavior, such as in autism, depression, individuals expressing negative affects and a high degree of neuroticism, profoundly affect the lives of those who suffer from these deficits. Recently, advances in social cognitive neuroscience, through animal models and new technologies in human research, have shown that social behavior is largely regulated by molecular mechanisms. More specifically, a central role of neuropeptides has emerged in the regulation of cognition and social behavior. Neuropeptides are distinguished by the evolutionary flexibility of their actions, in that they can act as neurotransmitters or neurohormones, activating receptors distant from their release site. Recently, neuroscientists have been studying the role of oxytocin in modulating human social behavior. Oxytocin is a 9-amino acid peptide comprising two disulfide-bonded cysteine residues, synthesized in the paraventricular and supraoptic nuclei of the hypothalamus, and secreted into the general circulation via the posterior pituitary gland where it acts as a hormone-stimulating hormone. childbirth or breastfeeding by stimulating contractions of the uterus and rising milk. Oxytocin is also released into the brain where it acts on the hippocampus, striatum, hypothalamus, amygdala, nucleus accumbens, and midbrain and hindbrain nuclei. In recent decades, oxytocin has been shown to exert a wide range of effects within the central nervous system. In particular, it has been reported that oxytocin is involved in the regulation of emotions, and has receptors in various regions of the brain, including the limbic system and amygdala [Huber et al., Science, 308: 245-248, 2005 ; Landgraf and Neumann, Front. Neuroendocrinol., 25: 150-176, 2004] [10, 11]. In mammals, oxytocin has been specifically associated with the development of prosocial behaviors such as mother-child attachment, grooming, approach behavior, sexual activity, and stress regulation [Carter, Psychoneuroendocrinol. , 23: 779, 1998; Ferguson et al., Front. Neuroendocrinol., 23: 200-224, 2002; Insel and Young, Nat. Rev. Neurosci., 2: 129-136, 2001; Lim and Young, Horm. Behav., 50: 506-517, 2006; Uvnàs-Moberg, Psychoneuroendocrinol., 23 (8): 819-836, 1998] [12-16]. Recently it has been shown that oxytocin facilitates the recognition of certain memorized faces and enhances the coding of social stimuli [Savaskan et al., Psychoneuroendocrinol., 33 (3): 368-374, 2008] [17].

An interesting hypothesis has recently implicated oxytocin in the etiology of autism [Insel et al., Biol. Psychiatry, 45: 145-157, 1999] [18], and in particular in the social disorders that are the hallmark of H FA [Waterhouse et al., 1996, supra; Bartz and Hollander, Horm. Behav., 50: 518-528, 2006; Insel, Am. J. Psychiatry, 154: 726-735, 1997 ; Lim et al., Int. J. Dev. Neurosci., 23: 235-243, 2005] [6, 19-21]. This theory was actually evaluated by measuring plasma oxytocin levels in children with autism. But whereas lower plasma oxytocin levels are present in children with autism compared to normal individuals, among children with autism, and contrary to normal individuals, higher rates of oxytocin correlated with weak everyday life and social interaction skills, as well as a larger overall deficit in social consciousness [Modahl et al., Biol. Psychiatry, 43: 270-277, 1998] [22].

 There is an increasing body of literature on the effects of oxytocin administration on modulation of social behavior and underlying functional brain activation. However, while this literature suggests efficacy of oxytocin in the treatment of social deficits in autism patients, it does not demonstrate this across all autistic behavioral characteristics. Indeed, only a reduction in repetitive behaviors has been demonstrated in patients with autism [International Application WO 2004/030524] [23], following parenteral administration of oxytocin. However this encouraging result on this particular behavioral characteristic associated with autism can not be generalized to the therapeutic effect of this hormone on other autistic behavioral characteristics. In fact, autists present a great inter-individual variability concerning the symptomatology itself; which requires a precise analysis of each of the behavioral characteristics that condition the deficient social behavior of an autist. This results in particular from the fact that the literature does not make it possible to establish a positive correlation between the rate of plasma oxytocin and the social capacities of individuals with autism, until very recently [International Application

WO 2008/042452] [24]. Moreover, if the literature describes an efficacy of oxytocin in the treatment of affective pain syndromes such as fibromyalgia and recurrent abdominal pain of non-organic origin [International Application WO 98/43660] [25], the schizophrenia, anxiety, and disorders associated therewith [International Application WO 2007/050353] [26] or to create eustasia [International Application WO 02/102832] [27], there is no evidence of efficacy of 'oxytocin in the treatment of behavioral characteristics associated with pathological shyness, which is a pathological condition distinct from the two previous ones as evidenced by research using the international NEO-PI-R questionnaire [Costa and McCrae, Assessment, 7 (4 ): 325-327, 2000] [28].

 Unfortunately, there are few treatment options for individuals with autism or disorders with similar behavioral characteristics. There is therefore a real need for neuropsychiatry to help the growing number of people living with these serious social disorders. In particular, it would be advantageous to have a relatively simple treatment to implement that would slow down or even control the development of behavioral characteristics associated with autism or pathological shyness, especially from childhood, and so reduce costs and improve the management of such patients.

Description of the invention

 The inventors have now clearly demonstrated the direct involvement of a low plasma concentration of oxytocin in the major deficits of individuals suffering from autism or pathological shyness, such as gaze avoidance, and the absence of social interaction with others. They have thus quite unexpectedly demonstrated that the intranasal administration of oxytocin, even in a low dosage, made it possible to improve said major deficits.

The subject of the present invention is therefore the use of oxytocin or an oxytocin analogue for the manufacture of a medicament intended to preventing or treating one or more behavioral characteristics associated with autism and pathological shyness, said behavioral characteristics being selected from the group consisting of social deficits, and gaze avoidance.

 For the purposes of the present invention, the term "oxytocin analogue" means oxytocin compounds or fragments having an activity at least similar to that of oxytocin and / or a longer shelf life. For example, it is 4-threonine-1-hydroxy-deaminoocytocin, 9-deamidocytocin, carbetocin, injectable pabal, methylergometrine (methergin), vasopressin, serotonin, etc.

 For the purpose of the present invention, the term "social deficits" is understood to mean eye-to-eye avoidance, withdrawal and isolation behaviors, anxiety traits, difficulties in interacting with others and an inability to have shared attention. Said deficits can be demonstrated by techniques well known in the art, for example the Eye tracking Tobii system, 1750 which records the trajectory of the eyes, the electrodermal responses, etc.

 For the purposes of the present invention, "gaze avoidance" means an inability to direct the gaze towards another person, for example during a simple conversation. Said avoidance can be demonstrated by techniques well known in the art, for example eye tracking which is a camera that records the trajectory of the eyes of the participants while they perform specific tasks on the computer.

According to one particular embodiment of the invention, oxytocin is administered parenterally (intravenously, intramuscularly, subcutaneously, intraperitoneally), intranasally, as with nasal solutions or sprays, aerosols or inhalants, by intravenous route. oral as with solutions, suspensions, tablets, capsules, pills, rectally as with suppositories, pessaries. Oxytocin can be administered intranasally, for example at approximately 24 μl. corresponding to three sprays per nostril. According to a particular embodiment of the invention, the treated autism is high-level autism or HFA.

 Other advantages may still appear to those skilled in the art on reading the examples below, illustrated by the appended figures, given for illustrative purposes.

Brief description of the figures

Figure 1 depicts the Cyberball video game and the ball cast distributions vis-à-vis each of the three partners. (A) Schematic representation of the modified Cyberball video game. On successive trials, the role of the participant alternates between rounds as (i) the observer of ball exchanges between two of the other players, (ii) the recipient of the balls, and (iii) the sender of balls. The behavior of partners A, B and C is computer-generated to define three different profiles from the point of view of the participant: an inclusion profile or "good" (A), a "neutral" profile (B), and an exclusion profile or "bad" (C). The length of the arrows in gray is proportional to the number of balls sent by a given player to each of the other players. The profiles represent the average behavior over the entire game, but rather than being fixed from the beginning, they were gradually adjusted using an algorithm described below. The black arrows represent the behavior of a representative healthy participant. (B) Bale distributions for healthy participants and HFA under placebo or oxytocin. Under oxytocin, there is a significant tendency for the HFA participant to ship the bullets to the "right" partner rather than the "neutral" partner (significant trend, z = 1.82, p = 0.06, bilateral) (average and standard deviation of mean, ^* indicates significant difference at p <0.05 or better on post-hoc pairwise comparisons).

- Figure 2 schematically shows the tasks of perception of the face. Participants examine images of faces while the movements of their eyes are recorded with the Tobii eye tracking system (A) Gender identification (B) Identification gaze direction.

 Figure 3 shows the destination of throwing balls during the game Cyberball (A, B, C). Cumulative number of balls sent by a healthy participant (A), HFA under oxytocin (B) and HFA under placebo (C), partners A ("good") and C ("bad")

Figure 4 represents the subjective post-experimental evaluation under placebo and oxytocin. A score (1 -7) reflects the subjective emotional states of confidence (A) and preference (B) for the three partners in placebo and oxytocin HFA participants (mean and standard deviation). mean, ^* indicates a significant difference at p <0.05 or better on post-hoc pairwise comparisons).

 Figure 5 represents the average duration of gaze fixation of HFA participants under placebo and oxytocin passed on the different regions of interest. (A) Duration of the look on the face and outside of the facial area when the participants have to identify the sex (male / female face) (B) Detection of the direction of gaze on the face and outside of the facial area (direct / indirect) (C, D) Duration of the gaze spent on the main areas of interest: eyes, mouth / nose, and other areas such as forehead / cheeks during identification sex (B) and detection of gaze direction (C).

Figure 6 shows plasma oxytocin concentrations in HFA and healthy participants. In HFA participants, plasma oxytocin concentration is measured in the blood before and 10 minutes after oxytocin administration (mean and standard deviation of the mean, ^* indicates a significant difference p <0.05 or better on post-hoc pairwise comparisons).

Fig. 7 shows positive correlations between plasma oxytocin concentration and extraversion domain (A), and its facets: sensation seeking (B), gregariousness (C), positive emotions (D), and vulnerability (E).

 Figure 8 shows the negative correlation between brain volume and (A) plasma oxytocin concentration and (B) sensation seeking facet.

 EXAMPLES

EXAMPLE 1

Materials and methods

participants

 Thirteen adults (1 1 men and 2 women, mean age = 26 years, age range = 17-39 years) with a clinical diagnosis of Asperger's syndrome (AS) or high-level autism (HFA) according to the DSM-IV R classification (American Pshychiatric Association, 2000) and ASDI [Asperger Syndrome (and high-functioning autism) Diagnostic Interview] [Gillberg et al., Autism, 5 (1): 57-66, 2001] [29 ] were recruited from the expert centers (Fondation FondaMental) of Mondor-Chenevier Hospital, Créteil, France.

 Eye trajectory analysis and plasma oxytocin were performed on 12 autistic patients instead of 13 due to technical problems. The study also involved a control group of 13 healthy individuals corresponding to autistics for age and sex (mean age = 26 years, age range = 18-40 years).

 All participants were free of medication for at least 2 weeks before the study and stayed for the entire study. The exclusion criteria were: pregnancy, medication, significant medical illness, drug or alcohol abuse, and consumption of more than 15 cigarettes per day. Participants refrained from eating and drinking (except water) 2 hours before the experiment; to smoke and consume caffeine 24 hours before the experiment.

All participants gave their informed consent and all procedures were approved by the Local Ethics Committee (Center Léon Berard, Lyon IV) and by the French Agency for Sanitary Safety of Health Products (AFSSAPS) competent for clinical trials on a medicinal product for human use.

 All diagnoses were performed by experienced clinicians, were based on clinical observations of participants and interviews with parents or caregivers, and were confirmed using ADI-R (Autism Diagnostic Interview) [Lord et al. ., J. Autism Dev. Dis., 24: 659-685, 1994] [30] (Table 1).

 Table 1

mutual social interaction 10, communication 8, and stereotyped behaviors 3, respectively. All participants were above the limits in the test algorithms. As part of the verification procedure, the French translation of the A-TAC interview (Autism, Tics, AD-HD (Attention Deficit Hyperactivity Disorder) and other Comorbidities) [Hansson et al., Br. J Psych., 187: 262, 2005] [31] was completed by the parents. This screening questionnaire focuses on a number of skills, behaviors, and behaviors in the functioning of children compared to peers. Parents were asked to report any problems or specific characteristics during any period of life, even when they are no longer present. None has been clinically delayed in language development; participants were excluded when the clinician or parents were uncertain about the early language development. Overall, study participants presented verbal and non-verbal intelligence estimates of medium to above average (Table 1).

Experimental protocol

 The study used an experimental double-blind, placebo-controlled, randomized project.

 Participants received oxytocin or placebo during two visits to the laboratory, separated by 7 days as shown below:

 13 HFA participants assessed for eligibility

Randomized study of 1s 13 HFA participants

 - 6 participants received first oxytocin placebo

 - after one week, the 6 participants have - after one week, the 7 participants received the placebo received oxytocin

 1 - 1 data lost on the analysis

- 0 lost data

 Plasma and the pursuit of the eye

- 0 withdrawal of consent

 - 0 withdrawal of consent

13 participants analyzed for:

 - effects of oxytocin on behavior in a social game

 - 1 participant excluded from the study for facial perception tasks and plasma oxytocin concentration measurements

 12 HFA participants tested in fine

 The participants were tested on a social game of "throwing balls" and on facial perception tasks, and completed a number of scoring scales following the game of balls at each visit.

The ball game was a variation of the Cyberball video game [Williams et al., J. Pers. Soc. Psychol., 7: 748-762, 2000] [32] in which the participant used an LCD touch screen (Elo 1725L) to participate in a multi-turn animation game of throwing balls on a computer network with three virtual partners A, B and C represented by cartoon characters (corresponding photo selected among the Nimstim face database under neutrality and similarity control) [Tottenham et al., J. Cognitive Neurosci., 14: S74, 2002] [33] (Figure 1A, insert). Participant "P" was represented by an additional drawing representing a pair of animated hands assuming a projection of the person. In order to make social interaction as plausible as possible, by entering the test room, the participant could see the backs of three other people sitting in separate booths, and was informed that they were the other players but that he could not see them directly to preserve privacy. Each try consisted of a single ball exchange represented by a short animation showing a player handling the ball, and a few seconds later, another player catching the ball. If a test ended with the participant as a receiver, he started the next test as a sender and had to send the ball to partner A, B or C by touching the corresponding photograph. The game included a monetary incentive to reinforce the participant's cognitive commitment to the task. Any player receiving the ball earned € 2. Therefore, by throwing the ball, the participant was making altruistic money for the other players, but he had no direct control over their behavior and could only hope that they would act reciprocally. The participant had to take into consideration the two possible outcomes: either "a good chance" that the chosen partner plays with him again or "the risk" that the selected player decides to raise the ball to someone else. The participant was informed that the game would end after a total of 80 throws, and that each player would receive a fixed percentage (10%) of his accumulated winnings. The behavior of partners A, B and C was orchestrated by the rules of the The ball-throwing choices of the participant constituted the dependent variable.

 The amount of ball exchange presented by the three partners was manipulated. Unbeknownst to the participant, the probability distribution associated with the throwing choices of each of the three partners was defined so that, over time, everyone was endowed with different levels of reciprocity. Reciprocity is a kind of social behavior that requires cooperation while the individual gradually builds expectations from the actions of others. At the beginning of the game, the probabilities were homogeneous for all the partners, that is to say that the participant had a probability p = 1/3 of receiving the ball from any one of the three partners. After a predetermined number of laps, the partners' profiles diverged so that the partner A (the "good" profile) sent, on average, 70% of his balls in play to the participant (P), the partner B (the "neutral" profile) sent 30% of his balls into play to P, partner C (the "bad" profile) sent 10% of his balls into play to P. The probability distributions of throwing balls from the three partners were gradually developed to minimize the risk of reciprocal "fast-paces" between P and partner A ("good" profile) or blocking situations in which P might be excluded from the game for more than 10 tests.

 The generation of the test sequences and the development of the profiles of the fictitious players was carried out as follows:

max is the total number (100%) of balls exchanged between all players in the game. pass (x, y) means a single ball (or exchange) between two players X (sender) and Y (receiver), max is set at 80 ball passes (bp) and therefore each pass (x, y) is equal to 1 bp / 80b.p. = 1, 25% of max. pass (x, y) is not reflected so can not be both sender and receiver. Fictitious partners are assigned a separate profile and their participation in the game reflects that profile.

The participant is presented as the player D through the game. The player D is the only autonomous player able to guide the game through his way of playing. Fictitious partners to be compliant with their corresponding profiles, the upper and lower limits of their participation (described as worst and best possible scenarios) are evaluated in order to estimate their participation in the game. In the worst case scenario, only two partners play the 80 balls of the game, which corresponds to situations where the player D always plays with the same partner. In fact, none of the fictitious partners will always play with the player D, nevertheless this theoretical calculation is taken into account to be the upper limit, maxindividual represents this upper limit, and corresponds to 50% of max. Obviously for each player p (= partner A, B or C) the worst scenario corresponds to an identical ball pass between p and D. xyx

 Therefore, Σ pass (p, D) + Σ pass (D, p) = 100% results inΣpass (p, D) i = 1 i = 1 i = 1

= 50%. Since max was set at 80b. p., maxindividual = 50% of max = 40b. The best scenario estimate could be neutral social behavior where all players play identically with each other. The participation of each player is equal to max I (number of players) = 100% / 4 = 25%. Therefore, the b.p. per player is equal to 25% of 80b. p. = 20b. p .. A game of 40 balls (upper limit) has been planned for each profile, while keeping the lower limit, that is to say 20b. p., as the breakpoint to refresh the behavior of profiles.

 These distributions were performed as follows:

The computer-generated behaviors of partners A, B, and C were based on expectations from other past behaviors. Each of the three partners was assigned a particular profile: "good" for player A, "neutral" for player B, and "bad" for player C, specified by different percentages to circulate the ball between each of three potential players. Since each player must playing with the other three players, X, Y and Z are the percentages of playing the ball where X is the least appreciated player, X + Y + Z is the preferred player, and X + Y is the intermediate player. Are defined X> 0, Y> 0, Z> 0 where (Z + Y + X> Y + X>X> 0 at startup). The progression of bp per player during the game has been defined as a round of ball swapping for each player in a random sequence. If p represents a dummy partner chosen from the set P formed by> 4 and C, then:

- pass (p, a) _{t = 0} = Χ, α Θ {{A, B, C} - {p}}

- pass (p,>) _{t = Q} = Χ + Υ, β C {{A, B, C} - {p}}

- pass (p, b) _{t = 0} = Χ + Υ + Ζ, δ 6 {{A, B, C} - {p}}.

 Therefore a propagation for a player p, from time t to time t + 1, can be expressed (if X> 0, Y> 0, Z> 0) as:

prog (p) _M = {pass (p, a) ₁ - 1, 25%) + (pass (p, 1) ₁ - 1, 25%) + (pass (p,) ₁ - 1, 25%) From a sender's point of view, the application of this progressive behavior during the game is presented in a distribution cycle to the remaining players by applying the prog function (with particular heuristics). Ball exchanges with other players were randomized to minimize the likelihood of having the same sequence of ball distribution between two propagations for the same player. In order to allow the participants a non-intuitive deduction of each profile behavior, the profiles A and C were developed in order to gradually appear as "neutral", then "partially neutral" and finally "discriminatory" behaviors. Player B behaves "neutral" throughout the game. "Neutral" behavior occurs if and only if X> 0, Y> 0, Z> 0, the "partially neutral" behavior occurs if and only if X = 0, Y> 0, Z> 0, the "discriminatory" behavior occurs if and only if X = 0, Y = 0, Z> 0. In order to instantiate the values for X, Y and Z for each fictitious player, the concepts of inclusive and exclusive loops must be kept in mind. An exclusive loop corresponds to a cycle of at least 10 passes where the participant does not play (ieΣ pass (p, p "), where ρ ', ρ" Θ P and P Θ {A, B, C} and p' ≠ p ") ^{i = 1}

 Specifically player A (the "good" profile) spends 70% of his balls played on player D (the participant), 20% on player C (the "bad" profile) and 10% on player B (the "neutral" profile) ); player C (the "bad" profile) passes 10% of his balls played to player D (the participant), 20% to player A (the "good" profile) and 70% to player B (the "neutral" profile) ; player B (the "neutral" profile) passes 30% of his balls played to player D (the participant), 40% to player C (the "bad" profile) and 30% to player A (the "good" profile) . These percentages were defined during the pilot test and selected because they were more robust with respect to inclusive and exclusive loops, and ensured a constant total number of b.p. per player D (the participant), regardless of his own player profile. A direct illustration of profile specification and game sequence is as follows:

 - for player A: [Σ pass (A, C) = 20% of 20 balls = 4 balls, Σ pass (A, B) = 10% of 20 balls = 2 balls andΣ pass (A, P) = 70 % of 20 balls = 14 balls];

 - for player B: [Σ pass (B, C) = 40% of 20 balls = 8 balls, Σ pass (B, A) = 30% of 20 balls = 6 balls andΣ pass (B, P) = 30 % of 20 balls

= 14 balls];

 - for player C: [Σ pass (CA) = 20% of 20 balls = 4 balls, Σ pass (CB) = 70% of 20 balls = 14 balls andΣ pass (C, P) = 10% of 20 balls = 2 balls].

Each throwing sequence of 20 balls is divided into sets of 3 or 2 or 1. These proportions could be obtained with player sequences such as these (hereinafter is specified the random randomization function of the blocks):

- for player A: {rand [BCD] rand [BCD] rand [DC] rand [DC] [D] [D] [D] [D] [D] [D] [D] [D] [D] [D]}; for player B: {rand [ACD] rand [ACD] rand [ACD] rand [ACD] rand [ACD] rand [ACD] [C] [C]};

 - for the player C: {rand [ABD] rand [ABD] rand [AB] rand [AB] [B] [B] [B] [B] [B] [B] [B] [B] [B] }.

 The task was designed and executed using the Presentation software

(NBS Inc., Albany CA, version 10.1). The graphics and animations were taken from the video game Cyberball Williams et al. [33]. The calendar of events (decision time, throwing balls) performed by computer-generated players has been modified from trial to try to increase the impression of a realistic behavior.

 After completion of the game, participants were asked to rate their feelings of "trust" and "preference" for each of the fictitious partners on a scale of 1 to 7 points: How would you rate you trust for a player X, from total mistrust (1) to total trust (7)? How would you rate your level of preference for a player X compared to other players, from least favorite (1) to preferred (7)?

In order to reinforce the observations on the effects of oxytocin on the processing of socially relevant information, it was studied how participants looked at pictures of human faces. Participants examined pictures of faces presented one at a time on a computer screen while their eye movements were recorded. After calibration of the 1750-Tobii ™ binocular free-eye tracking system (participant's eyes on a series of blue circles appearing and disappearing in the center and at each corner of the 17-inch video player), the images of the faces have been presented on the system drive, with a maximum resolution of 1280x1024 pixels, 0.5 degree of accuracy, and a sampling rate of 50 Hz (manual Tobii technology). Recordings of gaze movements were analyzed offline using the ClearView ™ software. Two random sequences of facial stimuli were presented in separate pieces with different task instructions. In one task, the participants were instructed to identify the gender (Gl) of the faces. The stimuli consisted of 24 color images of male (12) and female (12) neutral faces, selected from the Nimstim database [Tottenham et al., 2002, cited above] [33]. Each image was presented on the computer screen for 2 seconds, followed by two labels ("male", "female") until the participant responded by pressing one of the two answer keys on the display. the computer keyboard (Figure 2A). In a second task, they were asked to detect the direction of gaze (GD) of faces. The stimuli consisted of 24 gray-scale images of male faces (selected from the Wicker B. database). Half of the faces presented a averted look (12) and the other half a direct look (12). Each image was presented for 3 seconds, followed by two labels ("look at me", "do not look at me") until the participant answered by pressing one of the two answer keys on the keyboard. computer (Figure 2B). The programmed stimulus sequence and presentation were performed using Presentation Software (NBS Inc., Albany CA, version 10.1). MATLAB software was also used. For each face, the gaze fixation time (in milliseconds) and the number of jerky eye movements were recorded for each of the 6 regions of interest (ROIs). Five areas of interest (ROI) of the face were defined: the two eyes, the nose, the region of the mouth, the forehead, and the two cheeks. A sixth ROI consisted of the portion of the image outside the contours of the face. For each image, the gaze fixation time (in milliseconds) was calculated for each of the 6 ROIs. The number of movements by jerking of the eyes (rapid changes in the direction of gaze) made for two ROIs, inside and outside the contours of the face was calculated (an increasingly fine grind of the face results in too much few eye movement samples), and converted to jerk frequency corresponding to (number of saccades) / (total fixation time in the ROI).

 Healthy subjects were tested on a single visit. Prior to the actual start of the experiment, the instructions were presented to the participants who declared confidentiality agreements. They were informed that their participation was voluntary and that they could stop at any time.

 The participants arrived in Lyon one day before the experimental program. The tests were conducted in the same way for all participants (Table 2).

 Table 2

 by a registered nurse. After a short rest period, a nasal spray was administered to the participants (24 IU, 3 sprays per nostril, placebo or Syntocinon®). Ten minutes later, a second blood sample (6 ml EDTA tube) was taken. The tests started 50 minutes after nasal intake. Each participant served as their own control, and received oxytocin and placebo at one week intervals. The order was balanced between the participants and kept fixed within the participants.

Blood samples (6 ml) were frozen within 15 minutes of collection. The plasma was separated by centrifugation at 2000 xg (for 10 minutes at 4 ° C) and then stored in a freezer at -70 ° C until assayed. Centrifugation and storage of EDTA tubes were performed at the hospital's Neurobiotec center Neurological Institute of Lyon. The Center for Immunology and Neuroendocrinology of the University of Liège has carried out the analyzes of the immunoprotective oxytocin in the plasma. Prior to the assay procedure, plasma samples were filtered on Centricon YM-3 (3000 Da cleavage potential, Millipore USA) to remove interfering plasma proteins. After filtration, the plasma oxytocin assay was performed using a specific enzyme immunoassay [Péqueux et al., Cancer Res., 69: 4623-2629, 2002] [34]. The plasma oxytocin assays obtained were compared to the normative data range published by Péqueux et al. [Scand. J. Clin. Lab. Invest., 61: 407, 2001] [35].

 Blood samples and oxytocin administration were performed at the Lyon Neurological Hospital near the CNRS Cognitive Neuroscience Center (CNC) where plasma oxytocin assays and behavioral tests were performed.

 General affect was measured after taking oxytocin and following placebo for each participant using the PANAS scale [Watson et al., J. Pers. Soc. Psychol., 54 (6): 1063-1070, 1988] [36] to evaluate the possible effects of mood alteration of oxytocin. PANAS is a standardized test used to measure emotional states by questioning participants to assess their moods and feelings through various adjectives covering a range of positive emotional states (eg, happy, relaxed, proud, confident, etc.). or negative (irritable, frightened, ashamed, shy, etc.).

In order to minimize the effects of learning between two visits, different versions of the "throwing balls" game have been created in which the position on the screen of the three partner profiles has been swapped and the associated photos have been modified. The order of the versions was randomized among the participants, as were the oxytocin and placebo assignments, thus eliminating any possible experimental biases. The healthy participants (control) were distributed at random on one of the two versions of the game. In the tasks of perception of the face, the order of presentation of the tasks (sex or direction of the look) was randomized between the participants and counterbalanced between the visits. Statistical analysis

 Statistical analyzes were performed on the behavioral data of the social game of "throwing balls", measurements of eye movements and levels of plasma oxytocin.

 Comparisons between groups and drug treatment conditions were made using the non-parametric ANOVA test (Friedman) as well as the sum of the Wilcoxon and Mann-Whitney tests.

 The effect of the treatment sequence was tested using the Mann-Whitney test. All tests were evaluated against two bilateral probabilities. Statistical analyzes were performed using Statistica software (statistical package for the social sciences version 7.1) and MATLAB (the MathWorks version 7.0) for Windows.

 Preliminary analyzes showed no effect on the order or session versions, so the variables were not considered further. Due to a synchronization problem with the Tobii system, the eye movement analysis data for facial perception tasks of a patient with autism could not be included.

Results

Social game of "throwing balls"

 The behavioral decision variable of interest in this game is the participant's ball throw choices. However, the data were reworked according to Partner A's lap intervals, since it was by observing A's behavior that the participant could learn to cooperate more with him than with the other two partners.

So the first 6 shots of partner A were impartial, so the probability of the participant receiving the ball p [A → P] = 0.33. On the next 4 shots, partner A increased the proportion of balls sent to the participant receiving the ball p [A → P] = 0.50. From the 1 ^st throw, partner A sent all his balls to the participant, p [A → P] = 1. At the same time, partner C started to exclude participant P, revealing himself as the "bad" profile. The effect of these biases on the behavior of the participants was illustrated in Figure 3, commented below. Since preliminary analyzes in healthy participants and autistic individuals failed to reveal differences in behavior with respect to the "neutral" versus "bad" profile, the focus of the study was focused on "good" and "bad" partners.

 Under placebo treatment, HFA participants showed little evidence that they discriminated against the cooperation profiles of the 3 fictitious partners. While healthy participants sent significantly more bullets to the "right" partner than to the "bad" partner (Wilcoxon test, z = 3, p <0.003) or the "neutral" partner (z = 2.76, p <0.005) (Figure 1 B, left), placebo-treated HFA participants responded in the same way to all partners (z = 0.36, p = 0.72, z = 0.2, p = 0, 84) (Figure 1 B, right).

 In contrast, taking oxytocin led participants to engage more often with the "right" partner and to send significantly more balls to this partner than to the "wrong" partner (z = 2.04, p < 0.041, bilateral) (Figure 1B, medium). When the effects of placebo and oxytocin were compared directly, a significantly greater difference in the number of bullets sent to the "right" partner compared to the "bad" partner (z = 1, 99, p <0.047, bilateral) has also been highlighted.

Finally, the difference in performance (number of balls sent to the "right" partner versus the "wrong" partner) between the "healthy" control participants and the HFA participants, which was significant placebo (Mann-Whitney test: z = 3.1, p <0.0021), disappeared when the comparison was made under treatment with oxytocin (z = 1.62, p <0.1 1).

 An increasingly fine picture of the participant's decision was obtained by examining the distribution of throwing balls over time. Healthy and autistic participants under oxytocin started to cooperate preferably with the "good" partner at about the same time, with the cumulative number of bullets sent to the "good" and "bad" partners respectively, which diverged significantly in the range 15-17 (Figures 3A and 3B, first of two significant series in a row on the difference between "good" and "bad" partners, healthy participants: z = 2.7, p <0.007, HFA partners under oxytocin: z = 2.04, p <0.041, bilateral Wilcoxon test). On the other hand, under placebo, the cumulative bale throwing curves of HFA participants with respect to "good" or "bad" partners did not differ significantly (Figure 3C).

In addition, during the post-experimental questionnaire, HFA participants reported that they were more confident with and had a stronger preference for the "right" partner than the "bad" partner, after playing with oxytocin (ANOVA test). Friedman, respectively, χ ² = 17.89, p <0.0002; χ ² = 13.63, p <0.001, post-hoc pairwise comparison test p <0.05, Figure 4), while their Feelings for the three fictitious partners did not differ with placebo (χ ² = 2.39, ρ = 0.3, χ ² = 1, 19, p = 0.55).

Visual exploration of the visapes

Tables 3A and 3B provide data on the total glare time (in milliseconds) and the frequency of eye jerking (milliseconds) in HFA and healthy participants, both inside and outside the contours of the eye. face, during the identification of sex (A) and the direction of gaze (B). (mean and standard deviation of the mean, ^* means a significant difference p <0.05 or better on post-hoc pairwise comparisons). Table 3

 Total fixation time (Msec) Frequency of saccades (bag / sec)

Face contours Face contours Face contours External internal external facial contours

AT.

B.

 Healthy participants focused their attention preferably on face contours (gender identification, Gl: 80% ± 2.09: gaze direction, GD: 84% ± 1, 74).

 In contrast, placebo-treated HFA participants spent significantly less time directly examining the faces compared to healthy participants (53% ± 7.24 in Gl: Mann-Whitney test, z = 2.67, p <0, 0077, 41% ± 8.04 in GD: z = 4.08, p <0.00005). A more detailed analysis of the visual exploration pattern of faces by region of interest showed that HFA participants particularly avoided the eye area (z = 2.88, p <0.004, z = 3.48, p <0 , 0005 for Gl and GD).

Interestingly, during glance direction judgments, HFA participants also produced more jerk movements than healthy participants (GD: z = 2.45, p <0.015). This increase in saccadic rate was only present during periods when HFA participants examined the faces directly but not when they explored the rest of the photo (GD: z = 0.71, p> 0.47) . A such eye-to-eye exploration, associated with high jerk frequency, implies that HFA participants scanned these images hastily through many brief fixations, possibly with high levels of anxiety and discomfort. The frequency of saccades did not increase during the gender identification task (Gl: z = 1, 36, p> 0.17), perhaps because, unlike the identification task of the direction of gaze, it does not depend critically on having to examine the area of the eyes.

 Oxytocin has changed the way HFA participants respond to face shots compared to placebo. Total fixation time on the face increased significantly in both tasks (Gl: z = 2.27, p <0.023, GD: z = 2.19, p <0.029, bilateral Wilcoxon test, Figures 5A and 5B, left). The effects of oxytocin can largely be explained by an increased fixation time on the eye region (Gl: z = 2.12, p <0.04, significant trend for GD: z = 1.88, p = 0.059, Figures 5C and 5D). No effects of oxytocin were observed on the other regions of interest (mouth and nose, Gl: z = 1, 18, p> 0.23, GD: z = 1, 41, p> 0.15 brow and cheeks, Gl: z = 39, p> 0.69, GD: z = 0.86, p> 0.38) (Figures 5C and 5D, right). Finally, oxytocin significantly reduced the abnormally high saccadic frequency observed with placebo during the task of identifying gaze direction (z = 2.12, p <0.03, bilateral).

Although oxytocin significantly increased the visual exploration of the faces by the HFA participants, compared to a placebo-treated state, the eye fixation time on the face and eye area remained significantly lower than that of the healthy participants. for all comparisons except for whole-face exploration in sex identification (Mann-Whitney test, face: GD: z = 3.21, p <0.002, Gl: z = 1, 96, p> 0.05, eyes: Gl: z = 2.77, p <0.006, GD: z = 2.99, p <0.003). This could be explained by the very low basal plasma oxytocin concentrations in autistic participants, and the fact that the relatively low dose that was administered failed to increase oxytocin levels to the point of restoring normal oxytocin function in the brain (Figure 2).

 In summary, HFA participants under oxytocin spent more time looking at facial photos and, specifically, the eye area. The concomitant decrease in the frequency of saccades, ie the increase in the average duration of fixation of individuals, suggests that oxytocin may reduce the fear or anxiety induced by visual stimuli in these autistic individuals.

 A potential effect of the treatment order was tested between the two visits of the participants in the placebo group and in the oxytocin group. No difference was observed between the two visits to one of the dependent variables measured in the bullets and visual exploration tasks (Mann-Whitney test, z = 0, p = 1).

 No statistical difference was observed in the measurements

PANAS mood states for oxytocin groups versus placebo (mean and standard deviation: 2.59 ± 0.63 versus 2.71 ± 0.52, z = 1, 42, p> 0.15, for positive affect , and 1, 27 ± 0.26 versus 1, 32 ± 0.2, z = 0.66, p> 0.5 for negative affect, Wilcoxon rank test).

Levels of oxytocin plasmatigue

 The basal plasma oxytocin concentration measured in HFA participants (1.08 μg / ml ± 1.04) was significantly below the values observed in a normative group of healthy participants (7.28 μg / ml ± 4.49). (Mann-Whitney test, z = 4.69, p <0.0001). A second measurement was performed 10 minutes after the nasal administration of a 24 IU dose of a Syntocinon® spray, and showed a significant increase in plasma oxytocin concentration (2.66 μg / ml ± 2.2) (Wilcoxon test, z = 1.88, p <0.02, Figure 6), indicating the successful assimilation of the substance.

The relationship between oxytocin concentration changes in plasma and brain has not been documented but is unlikely to be direct because, by inhalation, the neuropeptide molecules can penetrate directly into the cerebrospinal fluid, without passing through the bloodstream [Born et al., Nat. Neurosci., 5: 514, 2002] [37]. With this mode of administration, oxytocin is known to be available and remain in the brain for 30 to 80 minutes [Born et al., 2002, cited above] [37], and behavioral effects have been observed in this window. in healthy human subjects [Kosfeld et al., Nature, 435: 673-676, 2005] [38]. EXAMPLE 2: STUDY IN HEALTHY SUBJECTS OF PERSONALITY TRAITS VS PLASMATIC OCYTOCIN LEVELS AND VOLUME AND CONCENTRATION IN GRAY AND WHITE SUBSTANCES OF THE BRAIN

participants

 Thirty adult right-handed men (15 men), without any history of psychiatric or neurological diseases, participated in the study. These adults had no history of head trauma, reported no drug or alcohol abuse in the past 6 months, and were free of psychiatric medication.

 All the participants gave their informed consent and all the procedures were approved by the Local Ethics Committee (Léon Berard Center, Lyon IV) and by the French Agency for Sanitary Safety of Health Products (AFSSAPS) competent for the tests. clinics on a drug for human use.

Experimental protocol

 Revised Personality Inventory NEO (NEO PI R)

On the first day, all participants completed the NEO PI R questionnaire [Costa et al., 2001, above] [9], which includes 240 items divided into 30 facets (six facets by domains) and 5 domains (extraversion, neuroticism, agreeability, awareness, openness). Personality data has been noted to represent T scores. In particular, extraversion and neuroticism dimensions were evaluated as representing two personality traits of interest for this study.

Determination of oxytocin concentration

 On the second day, participants received a blood sample (6 ml in an EDTA tube) from a registered nurse. The blood samples were transmitted to the neurological hospital of Lyon near the cognitive neuroscience center (CNC) of the CNRS where the behavioral tests took place.

 Blood samples (6 ml) were frozen within 15 minutes of collection. Plasma was separated by centrifugation at 2000 x g (for 10 minutes at 4 ° C), then stored in a freezer at -70 ° C until assayed. Centrifugation and storage of EDTA tubes were performed at the Neurobiotec Center of the Neurological Hospital of Lyon. The Center for Immunology and Neuroendocrinology of the University of Liège has carried out the analyzes of the immunoprotective oxytocin in the plasma. Prior to the assay procedure, plasma samples were filtered on Centricon YM-3 (3000 Da cleavage potential, Millipore USA) to remove interfering plasma proteins. After filtration, the plasma oxytocin assay was performed using a specific enzyme immunoassay [Péqueux et al., 2002, cited above] [34]. The plasma oxytocin assays obtained were compared to the normative data range published by Péqueux et al. [2001, cited above] [35].

 Magnetic resonance imaging acquisition

 On the third day, participants were scanned at

Center for Multimodal and Multidisciplinary Research and Study in Living Imaging (CERMEP, Lyon, France), using a Siemens 1, 5-T scanner. A juxtaposition of T1-weighted images 1 mm thick, high resolution (TR = 3000 ms, flip-flop angle = 7 °, matrix = 256x256) was acquired.

Analysis Statistical analysis

 Statistica 8 software was used for statistical analysis. In order to test the correlation between personality and oxytocin, extraversion and neuroticism score scores and their facets were correlated with plasma oxytocin levels, with age and sex adjustment, according to a literature review. multiple regression. Personality dimensions and facets were considered dependent variables, and oxytocin levels, age and sex as independent variables.

Voxel-to-voxel morphometry (VBM)

 Voxel-to-voxel morphometry is a structural analysis technique designed to evaluate the structural features of the brain, in order to quantify the localized volume and concentration of white matter and gray matter [Ashburner and Friston, Neuroimage, 1 1 (6 Pt 1): 805-821, 2000] [39]. Structural images of each participant were pre-processed [with SPM2 software (Wellcome Department of Imaging Neuroscience, London, UK); with MATLAB 7.2 software (Mathworks Inc., Sherborn, Massachusetts, USA)] according to an optimized voxel-to-voxel morphometry protocol [Good et al., Neuroimage, 14 (1 Pt 1): 21 -36, 2001] [ 40]. A personalized model, based on the magnetic resonance images of the participants, was used to optimize the results knowing the non-uniformity of the intensity of the image of each scanner.

Segmentation analysis was performed using a two-step approach. In a first step, the custom template was created by averaging all the segmented images of the entire sample. In a second step, all the images were segmented again using the created sample. The images were smoothed with a full-width 8mm filter at mid-height to increase the signal-to-background ratio. The smoothed images were analyzed according to multiple regression analysis using SPM2 software with MATLAB 7.4 software. T scores for extraversion and neuroticism and each of their facets, and plasma oxytocin concentration, were modeled in multiple regression analysis as principle covariates, and by age and sex. The analysis included the whole brain without taking into account any region of interest. Activations were considered significant when groups corrected with a non-stationary correction (ns) exceeded the significant level p <0.001.

Results

 Behavioral data

 The analysis of the completed questionnaires revealed a significant positive correlation between oxytocin and extraversion (r = 0.54, p <0.0068, Figure 7A), its facets gregariousness (r = 0.59, p <0, 05, Figure 7C), sensation search (r = 0.523, p <0.05, Figure 7B), and positive emotions (r = 0.43, p <0.05, Figures 7D). There was also a tendency for a negative correlation with the vulnerability facet of neuroticism (r = -0.43, p <0.05, Figure 7E).

Data from the VBM technique

 Plasma oxytocin concentrations were negatively correlated with gray matter concentration in the lower right temporal gyrus (coordinates of groups: x = 63, y = -32, z = -23, 2282 voxels, p <0.02, with correction ns) and in the right lower limbic lobe, specifically in the parahippocampus gyrus (coordinates of groups: x = 34, y = -16, z = -35, 751 voxels, p <0.05) (Figure 8A ).

In addition, a negative correlation was found between the sensation seeking aspect of extraversion and the gray matter concentration mainly in the Orbito frontal cortex (group coordinates: x = 1 1, y = 16, z = - 28, 1906 voxels, p <0.03) and the frontal frontal cortex (coordinates of groups: x = 21, y = 13, z = -24, 1271 voxels, p <0.03). No significant correlation was found for neuroticism and its facets (Figure 8B). These results show that the level of oxytocin has a great influence on personality traits, and modulates the activity (as measured by brain volume) of specific regions of the brain.

 Specifically, high levels of oxytocin have been correlated with extroverted and highly sociable personality patterns, while low levels of oxytocin have been demonstrated in individuals with oriented personality traits tending toward neuroticism and social vulnerability. .

 In addition, the results also indicated that individual differences in oxytocin and personality traits are also reflected in brain volume.

These results could provide a potential clinical contribution for healthy individuals who are severely socially disabled. Thus, healthy individuals with personality traits that can potentially affect their daily lives, such as fear of interaction with others, could benefit greatly from the administration of oxytocin.

List of references

 1. Klin et al., Arch. Gen. Psychiatry, 59 (9): 809-816, 2002

 2. Baron-Cohen et al., J. Child Psychol. Psychiatry, 42: 241-251, 2001

3. Frith and Miller, J. Child Psychol. Psychiatry, 45: 672-686, 2004

4. Durand et al., Nat. Genet., 39: 25-27, 2007

 5. Jamain et al., Nat. Genet., 34: 27-29, 2003

 6. Waterhouse et al., Psychol. Rev., 103: 457-489, 1996

7. Peterson et al., NeuReport, 17 (12): 1289-1292, 2006

8. Dalton et al., Biol. Psychiatry, 61 (4): 512-520, 2007

 9. Costa et al., J. Pers. Assess., 57 (3): 393-398, 1991

 Huber et al., Science, 308: 245-248, 2005;

he . Landgraf and Neumann, Front. Neuroendocrinol., 25: 150-176, 2004]

12. Carter, Psychoneuroendocrinol., 23: 779, 1998

13. Ferguson et al., Front. Neuroendocrinol., 23: 200-224, 2002

 14. Insel and Young, Nat. Rev. Neurosci., 2: 129-136, 2001

 15. Lim and Young, Horm. Behav., 50: 506-517, 2006

 16. Uvnàs-Moberg, Psychoneuroendocrinol., 23 (8): 819-836, 1998 17. Savaskan et al., Psychoneuroendocrinol., 33 (3): 368-374, 2008 18. Insel et al., Biol. Psychiatry, 45: 145-157, 1999

 19. Bartz and Hollander, Horm. Behav., 50: 518-528, 2006

 20. Insel, Am. J. Psychiatry, 154: 726-735, 1997

 21. Lim et al., Int. J. Dev. Neurosci., 23: 235-243, 2005

 22. Modahl et al., Biol. Psychiatry, 43: 270-277, 1998

23. International Application WO 2004/030524

 24. International Application WO 2008/042452

 25. International Application WO 98/43660

 26. International application WO 2007/050353

 27. International Application WO 02/102832

28. Costa and McCrae, Assessment, 7 (4): 325-327, 2000

29. Gillberg et al., Autism, 5 (1): 57-66, 2001 30 . Lord et al., J. Autism dev. Dis., 24: 659-685, 1994

 31. Hansson et al., Br. J. Psych., 187: 262, 2005

 32. Williams et al., J. Pers. Soc. Psychol., 7: 748-762, 2000

 33. Tottenham et al., J. Cognitive Neurosci., 14: S74, 2002

34. Péqueux et al., Cancer Res, 69: 4623-2629, 2002

 35. Péqueux et al., Scand. J. Clin. Lab. Invest., 61: 407, 2001

3 6. Watson et al., J. Pers. Soc. Psychol., 54 (6): 1063-1070, 1988 37. Born et al., Nat. Neurosci., 5: 514, 2002

 38. Kosfeld et al., Nature, 435: 673-676, 2005

3 9. Ashburner and Friston, Neuroimage, 1 1 (6 Pt 1): 805-821, 2000

4 0. Good et al., Neuroimage, 14 (1 Pt 1): 21-36, 2001

Claims

1. Use of oxytocin or an oxytocin analogue for the manufacture of a medicament for the prevention or treatment of one or more behavioral characteristics associated with autism and pathological shyness, said behavioral characteristics being chosen in the group consisting of social deficits, and avoidance of the gaze.

The use according to claim 1, wherein the oxytocin analogue is selected from the group consisting of 4-threonine-1 -hydrocodeaminooctocin, 9-deamidoocytocin, carbetocin, injectable pabal, methylergometrine (methergin), vasopressin, serotonin .

Use according to claim 2, wherein the oxytocin analogue is vasopressin or serotonin.

4. Use according to any one of claims 1 to 3, wherein the oxytocin or oxytocin analogue is administered intranasally.

Use according to any one of claims 1 to 4, wherein said behavioral characteristics are associated with high-level autism.