Effects of High‐Phenolic Extra Virgin Olive Oil on Sensory Processing in Children With Autism Spectrum Disorder: A Pilot Study

ABSTRACT

No approved treatments currently exist for the core symptoms of autism spectrum disorder (ASD), prompting clinicians and families to seek alternative approaches, including dietary and nutritional interventions. The Mediterranean diet (MeDi), characterized by a high intake of antioxidant and anti‐inflammatory components, is associated with improved health outcomes and reduced risk of chronic diseases. A key element of the MeDi is extra virgin olive oil (EVOO), which is rich in bioactive phenolics, including oleocanthal, oleuropein and hydroxytyrosol. Through its antioxidant and anti‐inflammatory actions, EVOO may contribute to regulating pathological processes in both the central nervous system (CNS) and the gastrointestinal tract, offering a promising avenue for pharmacological and nutritional intervention in children with ASD. This pilot study investigated the effects of a three‐month intervention combining the MeDi with EVOO supplementation on ASD symptoms. Ten children diagnosed with ASD aged 5.00–11.7 years, all boys, were recruited from Northern Greece. The changes in social interaction and communication were evaluated using standardized instruments: the Social Responsiveness Scale (SRS), Word Finding Vocabulary Test (WFVT), Autism Diagnostic Interview‐Revised (ADI‐R) and the Sensory Profile (SP). Paired‐sample t‐tests revealed significant improvements post‐intervention in SRS (p = 0.034), ADI‐R A (p = 0.008), WFVT (p = 0.020) and ADI‐R B (p = 0.022) scores. These findings suggest that EVOO supplementation may reduce the severity of social deficits and enhance communication skills in children with ASD.

Following the 3‐month Mediterranean diet and high‐phenolic extra virgin olive oil intervention, significant improvements were observed in social responsiveness, reciprocal social interaction and expressive vocabulary in children with autism spectrum disorder, with the most notable changes occurring in social and communication domains.

1. Introduction

Autism spectrum disorder (ASD) is defined as a lifelong neurodevelopmental disorder typically identified in early childhood, characterized by deficits in social communication and social interaction across multiple contexts as well as restricted, repetitive patterns of behaviour, interests or activities (American Psychiatric Association ²⁰¹³; Bjørklund et al. ²⁰²⁰; Maenner et al. ²⁰¹⁴; Thomaidis et al. ²⁰²⁰). ASD prevalence was reported as 1 in 44 children in 2021, increasing to 1 in 31 (3.2%) by 2025, with notably higher rates among boys (Maenner 2021; Shaw et al. ²⁰²²). In Greece, a study by Thomaidis et al. (2020) reported a prevalence of 1.15% among children aged 10–11 years.

ASD exhibits substantial phenotypic and biological heterogeneity (Lenroot and Yeung 2013; Lombardo et al. ²⁰¹⁹) and is frequently accompanied by a wide range of comorbid conditions, including anxiety, depression, psychiatric disorders, delayed motor milestones, attention‐deficit/hyperactivity disorder (ADHD), epilepsy, gastrointestinal disturbances and sleep disorders (Al‐Beltagi ²⁰²¹; Hawks and Constantino ²⁰²⁰; Lord et al. ²⁰⁰⁰; Mannion and Leader ²⁰¹³; Mansour et al. ²⁰¹⁷; Maski et al. ²⁰¹¹; Noland et al. ²⁰²⁰; Weir et al. ²⁰²⁰). Notably, some studies report comorbidity rates of up to 91% in individuals with ASD (Al‐Beltagi ²⁰²¹; Mannion and Leader ²⁰¹³; Noland et al. ²⁰²⁰). The prevalence and presentation of these comorbidities vary widely across studies, reinforcing the view that ASD is not a single condition but rather a spectrum of conditions with distinct pathophysiological profiles (Casanova et al. 2020). Evidence suggests that autism may involve subgroups in which one or more bodily systems are selectively affected, and this complexity is further supported by findings indicating that multiple systems can be involved, highlighting their variability (Burns et al. 2017; Noland et al. ²⁰²⁰).

Among the most frequently reported comorbidities in ASD are sensory dysfunctions, with prevalence rates ranging from 45% to 95%. Their consistent presence contributed to the inclusion of Unusual Sensory Behaviours in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM‐5) diagnostic criteria for ASD in 2013 (Grant and Nozyce 2013; Maenner et al. ²⁰¹⁴; Van Etten et al. ²⁰¹⁷).

1.1. Oxidative Stress, Mitochondrial Dysfunction and Neuroinflammation

Given that autism is multidimensional and associated with comorbidities involving diverse physiological systems, including the nervous, immune and gastrointestinal systems, fundamental cellular processes may underlie this condition (Balachandar et al. 2021).

Oxidative stress is a pathological imbalance between the production of reactive oxygen species (ROS) and the body's antioxidant defences. ROS are naturally generated in mitochondria during the electron transport chain (ETC) function, where oxygen is converted into energy (ATP). At physiological levels, ROS support signalling pathways involved in growth, repair and immune responses. In excess, however, they damage proteins, lipids and especially mitochondrial DNA, impairing energy production, amplifying ROS generation and perpetuating oxidative stress when antioxidant defences are insufficient (Bhat et al. 2015; Bhatti et al. ²⁰¹⁷; Bjørklund et al. ²⁰²⁰; Liu et al. ²⁰²²; Russell et al. ²⁰⁰²; Usui et al. ²⁰²³; Yenkoyan et al. ²⁰¹⁷). Multiple lines of evidence support the involvement of oxidative stress in ASD. Elevated levels of nitric oxide, thiobarbituric acid‐reactive substances (TBARS) and xanthine oxidase activity have been detected in the red blood cells of individuals with autism, indicating an increased oxidative load (Essa and Qoronfleh 2020).

The central nervous system (CNS) is particularly vulnerable to oxidative stress due to its limited antioxidant defences, high energy demands and elevated levels of lipid/ketones and iron (Bjørklund et al. 2020). Children are more susceptible, given naturally low glutathione levels from conception through infancy. In ASD, studies have reported reduced mitochondrial glutathione and genetic variations in related pathways, correlating with symptom severity (Liu et al. 2022). Notably, low glutathione levels alongside elevated oxidative stress markers have been associated with more severe gastrointestinal symptoms, a common comorbidity in autism (Bjørklund et al. 2020; Essa and Qoronfleh ²⁰²⁰; Liu et al. ²⁰²²; Rossignol and Frye ²⁰¹⁴).

Neuropathological studies provide further evidence of chronic neuroinflammation in ASD. Postmortem analyses reveal widespread activation of astroglia and microglia in the cerebral cortex and white matter, confirming an ongoing inflammatory process (Bjørklund et al. 2020; Noland et al. ²⁰²⁰; Siddiqui et al. ²⁰¹⁶; Yenkoyan et al. ²⁰¹⁷).

Mitochondrial dysfunction is increasingly recognized as a key mechanism in ASD, potentially contributing to its core pathology. Elevated oxidative stress levels resulting from mitochondrial abnormalities have been consistently documented, particularly in the mitochondrial electron transport system of individuals with ASD (Usui et al. 2023). This oxidative stress impacts critical brain regions involved in social behaviour, memory and motor coordination, including the hippocampus, cerebral cortex and cerebellum (Liu et al. 2022).

1.2. Sensory Processing

Sensory processing constitutes a critical function of the nervous system, as it involves the reception, processing and integration of sensory stimuli, thereby forming the basis for adaptive behaviour and the individual's interaction with the environment (Galiana‐Simal et al. ²⁰²⁰). Ayres's sensory integration theory outlines four stages: registration, modulation, discrimination and response. Dysfunction in any stage may result in sensory processing disorder (SPD), characterized by maladaptive behaviours and motor responses (Baker et al. 2007).

Studies indicate that up to 95% of children with ASD experience some form of sensory dysfunction, which is closely associated with behaviour, social interaction and adaptive capacity (Kirby et al. 2022). Individuals with ASD often experience sensory difficulties, ranging from excessive to insufficient responses to sensory stimuli (Maenner et al. 2014). Baranek et al. (2006) identified two primary patterns of sensory dysfunction: hypo‐responsiveness (reduced reaction) and hyper‐responsiveness (exaggerated response).

Tools such as Dunn's Sensory Profile (SP) and Ayres' Sensory Integration and Praxis Tests (SIPT) are widely used to diagnose and assess sensory disorders, offering clinically relevant insights into differences between children with and without such conditions (Katsiana ^2015b).

Sensory difficulties significantly affect psychosocial functioning, motor development and daily life of individuals with ASD. They are linked to secondary behaviours such as anxiety, aggression, sleep problems and emotion regulation challenges, highlighting the need for deeper investigation into their relationship with core ASD symptoms (Baker et al. 2007; Galiana‐Simal et al. ²⁰²⁰; Maenner et al. ²⁰¹⁴).

Mitochondrial dysfunction has been linked to increased vulnerability to sensory disorders, especially in individuals with neurodevelopmental conditions, due to the high energy demands of sensory systems. Oxidative stress may further exacerbate these difficulties by disrupting neuronal balance and promoting inflammation (Silva Santos Ribeiro et al. 2022; Rossignol and Frye ²⁰¹⁴).

1.3. Mediterranean Diet (MediDiet)—Extra Virgin Olive Oil (EVOO)

Oxidative stress, mitochondrial dysfunction and neuroinflammation are frequently observed in individuals with ASD, with their severity correlating with symptom intensity, suggesting a central role in ASD pathophysiology.

Currently, only two FDA‐approved drugs, risperidone and aripiprazole, are available for managing irritability and aggression in individuals with ASD (Baribeau and Anagnostou 2014; Goel et al. ²⁰¹⁸). At the same time, no pharmacological treatments target core symptoms, such as social or communication deficits (Hellings 2023).

Given the lack of effective pharmacological treatments for core ASD symptoms, interest has grown in alternative approaches such as dietary interventions, particularly the Mediterranean diet, which has long been associated with longevity and the prevention of chronic diseases (Gaforio et al. 2019). It is characterized by a high intake of plant‐based foods, fish and seafood; moderate consumption of aged cheese and red wine; limited meat and dairy; and avoidance of processed foods (Guasch‐Ferré and Willett ²⁰²¹; Pollicino et al. ²⁰²³; Sikalidis et al. ²⁰²¹; Summerhill et al. ²⁰¹⁸; Vrdoljak et al. ²⁰²²). Its hallmark is EVOO, the primary fat source (Gaforio et al. 2019).

EVOO contains monounsaturated fatty acids (e.g., oleic acid) and vitamin E, providing significant antioxidant activity (Salucci et al. 2022). Its polyphenols—such as hydroxytyrosol, oleuropein and oleocanthal—have notable anti‐inflammatory, anti‐allergic, antithrombotic, anticancer and antioxidant effects. Hydroxytyrosol protects against oxidative stress, oleuropein reduces inflammation and blood pressure, and oleocanthal acts similarly to ibuprofen (a non‐steroidal anti‐inflammatory drug) (Al Rihani et al. 2019; Gorzynik‐Debicka et al. ²⁰¹⁸; Hrelia et al. ²⁰¹⁹; Jimenez‐Lopez et al. ²⁰²⁰; Presti et al. ²⁰¹⁷; Yubero‐Serrano et al. ²⁰¹⁸). These compounds support the neuroprotective role of EVOO (Gonçalves et al. 2024).

A growing body of literature highlights the protective effects of EVOO in chronic diseases, focusing on its biological and pharmacological–nutritional potential (Gaforio et al. 2019; Vrdoljak et al. ²⁰²²). A study by Pase et al. (2015) demonstrated that a 20% olive oil–enriched diet (OOED) administered during pregnancy, lactation and post‐weaning in rats significantly reduced oxidative stress markers, such as lipid peroxidation and ROS, while enhancing antioxidant defences (GSH) and upregulating neurotrophic factors (BDNF and FGF‐2) in the prefrontal cortex and hippocampus. These findings highlight the neuroprotective potential of EVOO components, particularly hydroxytyrosol and oleuropein, during critical developmental periods (Pase et al. 2015).

EVOO shows neuroprotective effects in aging and early neurodegeneration through its polyphenols, which modulate inflammation and support neural function (Alkhalifa et al. 2024; Chatzikostopoulos et al. ²⁰²²). In MCI patients, Tsolaki et al. (2020) found that combining MedDiet with high phenolic early harvest EVOO improved cognition and symptom stability. Gonçalves et al. (2024) confirmed that oleuropein, hydroxytyrosol and oleocanthal reduce oxidative stress and support neuroprotection in models of Alzheimer's and Parkinson's diseases. Beyond neurodegeneration, EVOO's bioactive compounds are also linked to benefits in immuno‐inflammatory diseases, including inflammatory bowel disease (IBD), rheumatoid arthritis (RA), systemic lupus erythematosus (SLE) and multiple sclerosis (MS), as well as in cancer prevention, Type 2 diabetes management and gut microbiota modulation (Gavahian et al. 2019; Trichopoulou et al. ²⁰⁰⁰).

Disruptions in the microbiome–gut–brain axis in ASD have spurred interest in dietary interventions (Thoma et al. 2023). Rodop et al. (2021) found that the MediDiet, gluten‐ and casein‐free diets and the ketogenic diet may help reduce inflammation, alleviate gastrointestinal symptoms and improve brain function by regulating short‐chain fatty acids and serotonin levels.

The primary goal of the study was to evaluate whether a polyphenol‐rich Mediterranean diet incorporating high‐phenolic EVOO alleviates sensory processing difficulties and related symptoms in children with ASD, potentially mediated by CNS oxidative and inflammatory pathways.

2. Methods

2.1. Study Design

This non‐randomized pilot open intervention involved a three‐month Mediterranean diet combined with polyphenol‐rich extra virgin olive oil (EVOO).

2.2. Participants

Participants were recruited through personal, email or telephone communication, during which they received detailed study information and were screened for eligibility. Eligible and willing participants provided informed consent in accordance with the GDPR guidelines before scheduling their first visit. The study aimed to evaluate improvement in sensory processing and related domains following the dietary intervention.

A total of 37 families were initially approached and received a consent questionnaire along with detailed study information. Of these, 12 families provided written informed consent and completed the preliminary screening materials. However, 27 families ultimately did not participate in the study. Among them, 21 initially expressed interest but chose not to proceed, citing concerns about the feasibility of daily EVOO intake due to their children's restrictive eating habits. Four children were excluded because they did not meet the study's age inclusion criteria. Additionally, two children were excluded after initiating risperidone treatment (Risperdal) with increasing dosage prior to the intervention phase, which met the exclusion criterion regarding the introduction of new pharmacological or nutritional interventions during the study period.

As a result, 10 children with ASD completed the intervention and were included in the final analysis. All children had been diagnosed with either childhood autism or atypical autism. The diagnosis was further confirmed through the Autism Diagnostic Interview—Revised (ADI‐R; Lord et al. ¹⁹⁹⁴). The sample included exclusively boys aged between 5.4 and 11.7 years.

Participants were eligible if they had a clinical diagnosis of ASD based on DSM‐5 criteria, were between 6 and 12 years old, resided in Greece and were enrolled in a public educational or developmental support institution. Children were excluded if they had initiated a new treatment within 1 month prior to the start of the intervention. No inclusion or exclusion criteria were applied on the basis of verbal/non‐verbal status or general cognitive functioning.

The participants were recruited from local special and mainstream schools in Northern Greece through personal and professional contacts, such as psychologists, occupational therapists, doctors and teachers, as well as through parents' associations for people with autism, various special treatment and rehabilitation centres for children and through the distribution of informational flyers to local healthcare professionals. Participants' families selected for the study were provided verbal information and received consent forms detailing all necessary information. Parents or legal guardians of children who wished to participate signed the consent forms. Participation in the study was entirely voluntary and free of charge.

2.3. Measures

2.3.1. Demographic Information Questionnaire

Three questionnaires were developed to gather information: one for demographic data, one for the child's medical history (including the prenatal period, birth, psychomotor development, genetics, comorbidities and current condition) and one for medication and/or dietary supplements.

2.3.2. ADI‐R

The ADI‐R is a standardized, semi‐structured 90‐ to 150‐min interview designed to diagnose ASD in individuals 18 months and older, conducted with a parent or caregiver by a trained interviewer (Baldimtsi et al. 2021; Lebersfeld et al. ²⁰²¹; Wan Azamudin et al. ²⁰²²). The ADI‐R consists of 93 questions covering three functioning areas: communication, social development and play and repetitive and restricted behaviours (Wan Azamudin et al. 2022; Zander et al. ²⁰¹⁷).

The ADI‐R includes algorithms for diagnosis, referred to as Diagnostic Algorithms, and for the examinee's current clinical picture, referred to as Current Behaviour Algorithms (Wan Azamudin et al. 2022; Zander et al. ²⁰¹⁷). In the present study, the Current Behaviour Algorithms were used to compare children's scores at two different time points: before and after the 3‐month intervention with EVOO. The Greek ADI‐R has demonstrated satisfactory diagnostic/discriminant validity, including high sensitivity and satisfactory specificity, in a clinical sample of Greek children and adolescents with a wide range of intellectual abilities (Papanikolaou et al. 2009). However, a comprehensive evaluation of its factor structure and convergent and divergent validity has not, to our knowledge, been reported in a Greek population.

2.3.3. Social Responsiveness Scale (SRS)

The Social Responsiveness Scale—Second Edition (SRS‐2), School Age Version, is a revision of the original SRS. It is a questionnaire that assesses the severity of social dysfunction associated with ASD in children and adolescents aged 4–18 years (Constantino and Gruber 2005, 2012). It is a behavioural assessment scale that can be completed by a parent, caregiver or teacher familiar with the child's or adolescent's everyday behaviour (Nelson 2015). In this study, all questionnaires were completed by a parent. The SRS measures the extent and severity of social deficits and records symptoms and dysfunctions associated with autism in natural settings. It consists of 65 questions about the child's behaviour during the past 6 months. Based on algorithms, each question is re‐scored. Higher scores indicate increased ASD symptomatology and a need for further assessment (Bruni 2014; Channell ²⁰²⁰; Duvekot et al. ²⁰¹⁴; Nelson ²⁰¹⁵; Patti et al. ²⁰²⁴).

2.3.4. SP

The SP Test is designed to evaluate sensory modulation, children's ability to process sensory information and the impact of sensory processing on daily life activities. It is administered to children aged 3–12 years and is completed by a caregiver who is familiar with the child's behaviour across various contexts and periods (Brockevelt et al. 2013; Burns et al. ²⁰¹⁷; Katsiana ^2015b; Katsiana et al. ²⁰²⁰).

The caregiver is asked to respond to 125 behavioural statements (items) using a 5‐point Likert scale, indicating how frequently the child engages in each behaviour. Lower scores indicate better performance. Thus, when children never engage in a behaviour, they receive a score of 1, whereas frequent engagement scores 5 points (Brockevelt et al. 2013; Dunn ¹⁹⁹⁹; Jones et al. ²⁰²⁴; Katsiana et al. ²⁰²⁰).

Prior to 2006, analysis of the SP provided was based on Section and Factor scores. In 2006, a new form of analysis was introduced in the SP manual, incorporating the four sensory Quadrant scores, which represent four distinct sensory behaviour patterns (Katsiana ^2015a).

The Section scores, organized into three main sections, summarize children's abilities in Sensory Processing, Modulation and Behavioural and Emotional Responses (Brockevelt et al. 2013; Dunn ¹⁹⁹⁹; Jones et al. ²⁰²⁴; Katsiana ^2015a). The 9‐Factor scores, which include items from multiple sensory systems, were used to provide information regarding children's reactivity to sensory experiences (under‐responsiveness, over‐responsiveness) (Brockevelt et al. 2013; Dunn ¹⁹⁹⁹; Houwen et al. ²⁰²²; Katsiana ^2015b). The Quadrant scores represent four distinct behavioural patterns. Each quadrant represents the interaction between the child's neural threshold and the behavioural strategy they use to respond to sensory information. They assess the extent to which children fail to notice, process, seek out or are adversely affected by sensory stimuli. The Quadrants include registration, seeking, sensitivity and avoiding (Brockevelt et al. 2013; Houwen et al. ²⁰²²; Katsiana ^2015b; Katsiana et al. ²⁰²⁰).

Studies indicate that SP is a reliable and valid tool for assessing sensory processing abilities. In the present study, children's scores at two different time points—before and after the 3‐month intervention with EVOO—were compared across all SP parameters (sections, factors and quadrants). Caregivers were asked to complete the SP‐GR form twice, with an interval of approximately 3 months between each completion. The data are analysed below.

2.3.5. Word Finding Vocabulary Test (WFVT)

Communication difficulties in ASD exhibit varying degrees of severity, with some individuals being verbal and others remaining non‐verbal or exhibit limited verbal skills. To assess each child's expressive vocabulary, the Greek version of the standardized Word Finding Vocabulary Test (WFVT) was used (Renfrew 1997; Vogindroukas et al. ²⁰⁰⁹).

The WFVT is a naming activity consisting of 50 black‐and‐white images arranged in increasing order of difficulty, designed to assess the level of expressive vocabulary. The child's responses are recorded and subsequently scored as correct or incorrect. Each correct response is awarded 1 point, with a maximum score of 50 (Vogindroukas et al. 2009).

In the present study, the WFVT was also administered to older children, as it was initially considered appropriate for evaluating the limited expressive vocabulary observed in many children with ASD (non‐verbal or minimally verbal) and for estimating the developmental age of their expressive vocabulary. Additionally, it allows for quick re‐evaluation of this skill following the intervention (Kallergi 2015; Rose et al. ²⁰¹⁶; Vogindroukas et al. ²⁰⁰⁸). In the present study, the WFVT was used to assess changes in expressive vocabulary following the intervention and was not intended as a measure of general cognitive functioning. During the initial assessment, older, verbal children achieved the maximum score, whereas three non‐verbal children were unable to provide any responses, resulting in a total score of zero. The Greek adaptation of the WFVT is a norm‐referenced measure of expressive vocabulary, developed and standardized for Greek‐speaking children (Vogindroukas et al. 2009). Although evidence of high internal consistency has been reported, published evidence concerning its latent structure and broader construct validity in Greek samples remains limited.

2.4. Procedure

2.4.1. First Visit

Following the acquisition of written informed consent, the first author met with both the parent and participant during the initial visit. During this session, the parents completed the demographic questionnaire, the Social Responsiveness Scale (SRS) and the SP. Concurrently, the researcher completed a medical history questionnaire, focusing on medication and/or dietary supplements, to ensure the accuracy and completeness of the reported information. The researcher also administered the ADI‐R to confirm the ASD diagnosis. In addition, the child's expressive vocabulary skills were assessed using the WFVT.

Upon completion of all assessments and collection of the questionnaires, the researcher provided the parents with a supply of EVOO, along with verbal instructions for appropriate consumption based on the participants' dietary habits. Parents were subsequently provided with a daily form to record the EVOO intake and the dietary groups included in the child's diet. Finally, a licensed nutritionist provided a four‐page informational brochure outlining general recommendations and appropriate age guidelines for adopting the Mediterranean diet in children.

2.4.2. Study Intervention

After completing the first meeting and receiving the EVOO package, parents and caregivers were instructed to incorporate 30 mL of EVOO into their child's diet. Based on dietary habits and the nature of meals, parents/caregivers were verbally informed by the researcher that they could administer the 30 mL of EVOO as a single dose of 30 mL or divided into two doses of 15 mL each. According to the study protocol, EVOO was to be administered daily with meals for a period of 3 months (90 days). At the end of the intervention, the researcher would collect all updated data. In the recording form, parents recorded daily whether the 30 mL of EVOO was consumed, the number of doses (1 or 2), and the dietary groups consumed by the child during the day (e.g., dairy, meat, fish, cereals and processed foods). After completing the 3‐month intervention with daily EVOO consumption, participants were contacted by the researcher to schedule a second visit.

2.4.3. Second Visit

Parents/caregivers completed the SRS and SP forms again during the second visit. The researcher conducted a follow‐up interview using the ADI‐R, focusing specifically on the sections related to the child's current behaviour. Subsequently, the Greek version of the WFVT, which assesses expressive vocabulary, was re‐administered as part of the follow‐up evaluation.

The results section below thoroughly compares and systematically analyses all the previously described tools and assessments, which were re‐administered before and after the 3‐month intervention to evaluate the potential effects of EVOO on children with ASD. Finally, parents submitted the completed form to the researcher, documenting the daily recording form of EVOO and the child's dietary intake categorized by food groups. In all cases, the forms were fully completed for the 3‐month intervention period.

2.5. Statistical Analyses

Data was entered in Microsoft Office Excel 2016 from a hard copy. IBM SPSS 26 was used to analyse the data. Descriptive statistics were performed using mean (M), standard deviation (SD), minimum (Min) and maximum (Max) values for age. In contrast, percentages and frequencies were used for the remaining nominal variables. Inferential statistics were performed at a significance of 5% (Field 2017). The Cronbach's alpha coefficient was used to measure the internal consistency of the factors, with acceptable values considered to be those greater than 0.7 (Nunnally and Bernstein 1994). The normality of factors was tested via the Shapiro–Wilk test, which is considered more accurate (Razali and Wah 2011). Differences between pre‐ and post‐condition were examined using the paired samples t‐test. Data were presented with mean (M) and SD (Field 2017).

3. Results

3.1. Descriptive Statistics

Table 1 presents the descriptive statistics for the variables that refer to the Demographic Information Questionnaire. All children were boys with autism, with a mean age of 7.70 years (SD = 2.37). Half of them study in kindergarten, and the other half attend primary school. Most (80%, N = 8) study at a typical school, and 50% (5) receive parallel support and have completed the sensory integration sessions. Most of the children do not have comorbidities (80%, N = 8), and the minority receive Risperdal (30%, N = 3) and Depakine (10%, N = 1). Regarding vitamins and food supplements, 40% (N = 4) currently receive them, whereas 90% (N = 9) have received them in the past. Considering therapies, 70% (N = 7) accomplish two occupational and SLP therapies per week, and 50% (N = 5) accomplish two other therapies per week. The effect of ASD on the quality of life of the child and the family was high for the majority (70%, N = 7) of children.

TABLE 1.

Demographic information questionnaire.

Variable	Statistic
Diagnosis	Autism 100% (10)
Gender	Male 100% (10)
Age	M = 7.70
	SD = 2.37
	Min = 5.40
	Max = 11.70
School	Kindergarten 50% (5)
	Primary 50% (5)
Type of school	Typical 80% (8)
	Special 20% (2)
Comorbidity	No 80% (8)
	Seizures ‘E’ 10% (1)
	Obesity 10% (1)
Parallel support	50% (5)
Risperdal	30% (3)
Depakine	10% (1)
Vitamins/food supplements (past)	90% (9)
Vitamins/food supplements (present)	40% (4)
Sessions sensory integration	50% (5)
Occupational therapy per week	One 10% (1)
	Two 70% (7)
	Three 10% (1)
	Four 10% (1)
SLP therapy per week	Two 70% (7)
	Three 30% (3)
Other therapies per week	None 20% (2)
	One 30% (3)
	Two 50% (5)
Effect of ASD on the quality of life of the child and the family	Minimum 10% (1)
	Enough 20% (2)
	Very 70% (7)

Table 2 presents the reliability analysis results for the current study's factors, as measured by the Cronbach's alpha coefficient, which assesses internal consistency. Results indicate excellent reliability (α > 0.9) for ‘WFVT’ and ADI‐R Total B high (0.9 < α < 0.8) for SRS T‐score, ADI‐R Total A and ‘Sensory processing’ and satisfactory (0.8 < α < 0.7) for ‘Behavioral and emotional responses’ and ‘sensitivity to stimuli (Quadrant)’ in pre‐ and post‐condition. ‘Sensation Avoiding (Quadrant)’ indicated high reliability in pre‐condition and satisfactory in post‐condition. ‘Poor registration (Quadrant)’ indicated satisfactory pre‐ and high‐post‐condition reliability. ‘Modulation’ and ‘Sensation Seeking (Quadrant)’ indicated satisfactory reliability in pre‐ and moderate (0.7 < α < 0.6) in the post‐condition.

TABLE 2.

Reliability analysis.

Scale	Variable	Items	Pre	Post
Social Responsiveness Scale (SRS)	SRS T‐score	65	0.869	0.876
Word Finding Vocabulary Test (WFVT)	WFVT	50	0.983	0.983
Autism Diagnostic Interview‐Revised (ADI‐R)	ADI‐R Total A	15	0.859	0.815
	ADI‐R Total B	13	0.918	0.901
Sensory Profile (SP)	Sensory processing	65	0.864	0.806
	Modulation	34	0.763	0.691
	Behavioural and emotional responses	26	0.746	0.792
	Poor registration (Quadrant)	15	0.796	0.877
	Sensitivity to stimuli (Quadrant)	26	0.702	0.762
	Sensation Seeking (Quadrant)	20	0.720	0.618
	Sensation Avoiding (Quadrant)	29	0.867	0.777

According to Table 3, normality was accepted for all study factors, both in the pre‐condition (p ≥ 0.094) and post‐condition (p ≥ 0.198).

TABLE 3.

Test of normality.

Condition	Variable	W	df	p
Pre	SRS T‐score	0.975	10	0.933
	WFVT	0.837	7	0.094
	ADI‐R Total A	0.958	10	0.760
	ADI‐R Total B	0.893	10	0.185
	Sensory processing	0.898	10	0.207
	Modulation	0.968	10	0.875
	Behavioural and emotional responses	0.932	10	0.466
	Poor registration (Quadrant)	0.943	10	0.586
	Sensitivity to stimuli (Quadrant)	0.900	10	0.218
	Sensation Seeking (Quadrant)	0.914	10	0.311
	Sensation Avoiding (Quadrant)	0.966	10	0.852
Post	SRS T‐score	0.896	10	0.198
	WFVT	0.923	7	0.497
	ADI‐R Total A	0.949	10	0.652
	ADI‐R Total B	0.943	10	0.584
	Sensory processing	0.981	10	0.972
	Modulation	0.949	10	0.654
	Behavioural and emotional responses	0.950	10	0.671
	Poor registration (Quadrant)	0.910	10	0.283
	Sensitivity to stimuli (Quadrant)	0.973	10	0.917
	Sensation Seeking (Quadrant)	0.917	10	0.332
	Sensation Avoiding (Quadrant)	0.973	10	0.919

3.2. Inferential Statistics

According to Table 4, there was a significant decrease in post‐condition compared with pre‐condition, for ‘SRS T‐score’ (M _Pre = 78.90 vs. M _Post = 72.90, −8.12% change, p = 0.034) and ADI‐R Total A (M _Pre = 17.90 vs. M _Post = 14.90, −12.67% change, p = 0.008), although there was a significant increase for ‘WFVT’ (M _Pre = 20.57 vs. M _Post = 28.71, 50% change, p = 0.020) and ADI‐R Total B (M _Pre = 7.70 vs. M _Post = 10.90, 50% change, p = 0.022). No significant differences were observed between pre‐ and post‐conditional factors for the SP (p ≥ 0.173).

TABLE 4.

Comparisons between pre‐ and post‐condition (raw values).

Variable	Pre	Post	Statistic	p
SRS T‐score	78.90 (8.48)	72.90 (8.29)	t (9) = 2.491	0.034
WFVT	20.57 (16.05)	28.71 (12.28)	t (6) = −3.127	0.020
ADI‐R Total A	17.90 (6.33)	14.90 (5.28)	t (9) = 3.354	0.008
ADI‐R Total B	7.70 (6.68)	10.90 (6.57)	t (9) = −2.753	0.022
Sensory processing	232.10 (26.18)	245.10 (19.84)	t (9) = −1.363	0.206
Modulation	113.20 (15.55)	118.20 (12.50)	t (9) = −1.004	0.342
Behavioural and emotional responses	82.40 (11.73)	81.40 (12.96)	t (9) = 0.191	0.853
Poor registration (Quadrant)	54.20 (10.55)	57.40 (10.28)	t (9) = −1.191	0.264
Sensitivity to stimuli (Quadrant)	82.50 (12.46)	84.90 (11.83)	t (9) = −0.440	0.671
Sensation Seeking (Quadrant)	72.10 (9.67)	76.50 (8.38)	t (9) = −1.479	0.173
Sensation Avoiding (Quadrant)	105.30 (15.37)	108.10 (11.13)	t (9) = −0.545	0.599

4. Discussion

This pilot study aimed to examine the effect of using the Mediterranean diet in combination with EVOO rich in polyphenols for children with ASD, considering the potential improvement in ASD symptoms and autism‐related traits linked to sensory processing after a three‐month intervention. The sample consisted of 10 boys with autism, aged 5–11 years, the majority of whom attend typical schools, do not have comorbidities, receive two occupational and speech‐language therapy sessions per week and experience reduced quality of life as a result of ASD. The mechanisms described above, including oxidative stress, neuroinflammation and mitochondrial dysfunction, appear to be influenced by interactions between genetic and environmental factors during the prenatal, perinatal and postnatal periods, which may intensify oxidative stress (Bjørklund et al. 2020; Noland et al. ²⁰²⁰).

Results indicated a significant decrease of 8.12% on the SRS and 12.67% on the ADI‐R Total A score after 3 months of intervention with the use of the Mediterranean diet in combination with EVOO, rich in polyphenols, for boys with ASD. The decrease in the SRS indicates that the intervention was effective in reducing the severity of social dysfunction symptoms associated with ASD and beneficial in promoting social awareness, social knowledge, social communication and social motivation, as well as reducing the symptoms of restricted interests and repetitive behaviour. The decrease in ADI‐R Total A score indicates that the intervention was beneficial in reducing abnormalities in reciprocal social interaction, such as the failure to use nonverbal behaviours to regulate social interaction, the inability to develop peer relationships, a lack of shared enjoyment and a lack of socio‐emotional reciprocity. Elevated oxidative stress has been consistently associated with epigenetic dysregulation, chronic neuroinflammation and neurodevelopmental impairment in ASD (Liu et al. 2022; Noland et al. ²⁰²⁰).

Dietary consumption of EVOO may modulate ASD behaviour by influencing gut microbiota composition and resolving systemic inflammation (Millman et al. 2021). The oleuropein content is also a proteasome activator, clearing the cells of protein build‐up. HP EVOOs contain compounds like tocopherol Α, squalene, monounsaturated omega‐9, omega‐7 and polyunsaturated fatty acids such as omega‐3 and omega‐6 that have positive effects on the CNS through synergistic mechanisms. Studies suggest that certain polyphenols might also regulate neurotrophic levels of the nerve growth factor (NGF) and brain‐derived neurotrophic factor (BDNF). Such evidence confirms their involvement in neuroprotection (Wozniak et al. 2023).

Moreover, there was a significant 50% increase in WFVT score and ADI‐R Total B score after 3 months of intervention using the Mediterranean diet in combination with EVOO‐rich polyphenols for boys with ASD. The WFVT increase indicates that the intervention improved the verbal and non‐verbal communication of boys with ASD. The increase in the ADI‐R Total B score indicates that the intervention was beneficial in reducing communication abnormalities, such as lack of or delay in spoken language and failure to compensate through gesture, lack of varied spontaneous make‐believe or social imitative play, relative failure to initiate or sustain conversational interchange and stereotyped, repetitive or idiosyncratic speech.

In a study conducted in Cyprus, Thoma et al. (2023) examined the effects of high‐phenolic extra virgin olive oil (HP‐EVOO) supplementation in six individuals with ASD, with a focus on the potential modulation of the gut‐brain axis. After 6 weeks of intervention with HP‐EVOO combined with a special diet, participants exhibited significant improvements in ASD‐related symptoms, including communication skills.

The strength of the current study is the presence of significant effects after a 3‐month intervention using a Mediterranean diet in combination with EVOO rich in polyphenols for boys with ASD, considering the reduction in the severity of social dysfunction symptoms, abnormalities in reciprocal social interaction, improvement in verbal and non‐verbal communication and resolution of communication abnormalities, although the sample size was small. These preliminary findings suggest that, when combined with EVOO rich in polyphenols, the Mediterranean diet may offer beneficial effects for boys with ASD. However, further studies with larger sample sizes and greater internal validity are needed to confirm and generalize these results (McLeod ²⁰¹³).

5. Limitations

Despite the promising findings, this study presents several limitations that must be acknowledged. First, the small sample size limited the statistical power, increasing the risk of a Type II error—that is, the inability to detect statistically significant results even when an effect may exist (Cohen 2013). This limitation was particularly evident in the results derived from the SP questionnaire.

Second, although Greek‐language versions or adaptations of the instruments were used, the extent of psychometric evidence available specifically for Greek populations differs across measures. In particular, comprehensive evidence regarding factor structure and convergent/divergent validity is limited for the Greek ADI‐R and WFVT. In addition, the SRS‐2 used in the present study was translated from the English version into Greek and was not based on a formally validated Greek version. Moreover, the small sample of this pilot study precluded an independent examination of dimensionality through exploratory or confirmatory factor analysis (Kline 2014) and did not permit stable estimates of internal consistency. Future adequately powered studies should examine these properties in Greek‐speaking populations.

Third, the study relied on total scores from the SRS rather than analysing the contribution of its subscales (Awareness, Cognition, Motivation, Communication, Restricted Interests and Repetitive Behaviour), which could have offered a more nuanced understanding of symptom changes.

Fourth, convenience sampling restricts the generalizability of the findings (Creswell 2014). The sample consisted exclusively of boys aged 5–11 years with ASD, most of whom attend mainstream schools, have no comorbidities, receive two weekly occupational and speech‐language therapy sessions and present reduced quality of life due to ASD‐related challenges. These characteristics limit the extent to which the findings can be applied to more diverse ASD populations.

Fifth, the study did not investigate which participants benefited most from the intervention, as subgroup analysis or bivariate statistics were not conducted. Future studies with larger samples should consider stratification according to verbal/non‐verbal status and cognitive functioning, in addition to demographic characteristics such as age and educational level. Lastly, progress was only assessed at the end of the 3‐month intervention period; no intermediate or follow‐up evaluations were performed to track symptom progression.

6. Conclusion

New research is recommended to strengthen the results of the current pilot study. To ensure the necessary statistical power (Cohen 2013) to examine the current intervention's effect, new researchers should use a sample of at least 54 children with autism. They should also use a sample of 300 children with autism to confirm the validity of the scales used through confirmatory factor analysis (Kline 2014) and remove items with low positive or negative loadings (Tavakol and Dennick 2011).

Moreover, future studies should examine the intervention's effect on the individual factors of the Social Responsiveness Scale (SRS) (Awareness [AWR], Cognition [COG], Motivation [MOT], Communication [COM], Restricted Interests and Repetitive Behaviour [RRB]) and determine the profile of children for whom the intervention is effective. Furthermore, future meta‐analyses can compare the effectiveness of the Mediterranean diet intervention, combined with EVOO rich in polyphenols, across different age groups or genders.

Finally, it would be helpful to examine the progress of children with autism across different time points, such as monthly assessments.

To sum up, the 3‐month intervention using a Mediterranean diet in combination with EVOO rich in polyphenols for boys with ASD was found to help reduce the severity of social dysfunction symptoms and abnormalities in reciprocal social interaction, improve verbal and nonverbal communication and reduce abnormalities in communication.

Funding

The authors have nothing to report.

Ethics Statement

Ethical approval for this study was obtained from an institutional Bioethics Committee prior to study commencement. The study was conducted in accordance with applicable ethical standards and the General Data Protection Regulation (GDPR).

Consent

Written informed consent was obtained from the parents or legal guardians of all participating children following the provision of detailed verbal and written information about the study procedures. Assent was obtained from the children where developmentally appropriate. Participation in the study was entirely voluntary, and no financial incentives or costs were involved.

Conflicts of Interest

The authors declare no conflicts of interest.

Papastavridou, D. , Baldimtsi E., Katsiana A., and Tsolaki M.. 2026. “Effects of High‐Phenolic Extra Virgin Olive Oil on Sensory Processing in Children With Autism Spectrum Disorder: A Pilot Study.” International Journal of Developmental Neuroscience 86, no. 6: e70181. 10.1002/jdn.70181.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.