What does the research say about animals and anxiety in autistic children?

The most striking evidence is physiological rather than just reported. In O'Haire and colleagues' 2015 study, published in Developmental Psychobiology and funded in part by the US National Institutes of Health, 114 children aged 5 to 12 wore wristbands that measured skin conductance — how easily a tiny electrical charge passes through the skin, which rises with excitement, fear and anxiety. Autistic children showed higher arousal than their typically developing peers while reading silently, reading aloud and playing with toys. When two guinea pigs were brought into the room, the autistic children's arousal dropped significantly. The researchers suggest animals may act as 'social buffers' because, unlike people, they offer acceptance without social judgement.

Can horses help autistic children with emotional regulation and irritability?

A randomised controlled trial by Gabriels and colleagues (2015), published in the Journal of the American Academy of Child & Adolescent Psychiatry, randomised 127 autistic children and adolescents aged 6 to 16 to 10 weeks of therapeutic horseback riding or to a barn-activity control group with no horse contact. The riding group showed statistically significant improvements in irritability (effect size 0.42) and hyperactivity (effect size 0.45), with effects beginning by the fifth week. A 2018 follow-up found that about 73 percent of the irritability improvement was still present six months later (effect size 0.32), and that gains in social and communication behaviour were sustained. The researchers note that autistic children often have particular difficulty with emotion regulation, and hypothesise that time with the horse may dampen stressed or anxious states.

Is animal therapy a treatment or cure for anxiety in autism?

No. Animal-assisted and equine therapy are best described as a promising complementary support for wellbeing, confidence and self-regulation — not a treatment, an assessment or a cure for autism or for an anxiety disorder. The research base, while encouraging, is still developing: sample sizes are often small, and some studies rely on parent or teacher reports. Anxiety in an autistic child that is severe, persistent or affecting daily life should be discussed with a GP, paediatrician or CAMHS. Animal therapy can sit alongside that support, and many families find it a gentle, low-pressure way to help a child feel calmer and more able to cope.

Guides & Answers → Evidence

Does animal therapy help with anxiety and emotional regulation in autistic children?

Q: Does animal therapy help with anxiety and emotional regulation in autistic children?

The evidence is promising but still developing, and it is best understood as complementary support rather than a treatment or cure. Several studies point the same way: being with a calm animal appears to lower physiological signs of anxiety and to help with the irritability and dysregulation that many autistic children experience. In a study by O'Haire and colleagues (2015), autistic children aged 5 to 12 showed a significant drop in skin conductance — an objective, unconscious measure of anxious arousal — the moment two guinea pigs were introduced into a social situation, having been more aroused than their peers while reading aloud and playing with toys. In the largest controlled trial to date (Gabriels et al., 2015; 127 autistic children and teenagers), 10 weeks of therapeutic horseback riding reduced irritability (effect size 0.42) and hyperactivity (effect size 0.45) compared with a no-horse control group, with changes appearing by the fifth week. These are real, measured effects, but the field is young and animal therapy works best alongside school, NHS and family support, not instead of it.

Q: Do animals actually lower stress hormones like cortisol?

There is direct evidence in children, though not yet in an autism-specific trial. Pendry and Vandagriff (2014) ran a randomised trial of an 11-week equine-facilitated learning programme with 131 children (school-referred and community-recruited, not selected for autism). Children randomised to the programme had significantly lower levels of the stress hormone cortisol across the day, and in the afternoon, than children on the waiting list. It was the first evidence-based study in human-equine interaction to measure a change in participants' cortisol. This helps explain the calming effect families describe, and points to a biological pathway, but because the sample was not autism-specific it should be read as supporting evidence rather than proof for autistic children.

What the research actually measures — including one study where you can watch a child's anxiety fall the moment an animal walks in.

Short answer: The evidence is promising but still developing, and it is best understood as a complementary support, not a treatment or cure. Being with a calm animal appears to lower the body's signs of anxiety and to ease the irritability and dysregulation many autistic children live with. In one study, autistic children's physiological arousal dropped significantly the moment guinea pigs were introduced; in the largest controlled trial, 10 weeks of therapeutic riding reduced irritability (effect size 0.42) and hyperactivity (effect size 0.45), with changes showing by the fifth week. Real effects — but it works best alongside school, NHS and family support, never instead of it.

Why this question comes up so often

For a lot of autistic children, the hardest part of the day isn't a specific task — it's the anxiety underneath everything, and how quickly a small upset can tip into a much bigger one. Difficulty with emotion regulation is common in autism, and it is exhausting for the child and for everyone around them. So when parents hear that time with a horse, a dog or a guinea pig seems to help a child feel calmer, they quite reasonably want to know: is that real, or is it a nice story?

The honest answer is that it is partly measured and partly still being worked out. Below is what the strongest studies have actually found — including their limits.

You can see the anxiety fall — the skin-conductance study

The most persuasive single finding in this area is physiological, not just reported. In a 2015 study by Marguerite O'Haire and colleagues, published in Developmental Psychobiology and funded in part by the US National Institutes of Health, 114 children aged 5 to 12 wore wristbands measuring skin conductance — how easily a tiny, unnoticeable electrical charge passes through the skin. It rises when a person feels excited, fearful or anxious, so it is an objective, unconscious window on arousal that a child cannot fake or mask.

The children were arranged in small groups, each with one autistic child and two typically developing peers. While reading silently, reading aloud and playing with toys, the autistic children showed higher arousal than their peers — consistent with the social anxiety that families and teachers describe. Then two guinea pigs were brought into the room. For the autistic children, arousal dropped significantly. The researchers suggest animals can act as social buffers: because a guinea pig offers acceptance without social judgement, its presence appears to make an anxious child feel safer in a social setting.

What a controlled trial found on irritability and regulation

The strongest test of horses specifically is still a randomised controlled trial by Gabriels and colleagues (2015) in the Journal of the American Academy of Child & Adolescent Psychiatry. It randomised 127 autistic children and adolescents aged 6–16 to 10 weeks of therapeutic horseback riding, or to a barn-activity control group that came to the same centre and learned about horses but had no direct horse contact.

Against that control, the riding group improved significantly on measures closely tied to emotional regulation:

Irritability — effect size 0.42 (p = .05)
Hyperactivity — effect size 0.45 (p = .04)
Improvements on irritability and hyperactivity began by the fifth week of the programme

A 2018 follow-up in Frontiers in Veterinary Science checked whether the calm lasted. In the subset re-assessed six months later, roughly 73% of the irritability improvement was still present (effect size 0.32), and gains in social and communication behaviour were sustained. The authors are careful in their language — they hypothesise that time with the horse dampens stressed and anxious states — but the direction of travel is consistent.

The stress-hormone evidence

There is even direct biological evidence that structured time with horses lowers stress, though not yet from an autism-specific trial. Pendry and Vandagriff (2014) ran a randomised trial of an 11-week equine-facilitated learning programme with 131 children (referred by school counsellors and recruited from the community, not selected for autism). Children randomised to the programme had significantly lower levels of the stress hormone cortisol across the day, and in the afternoon, than children on the waiting list. It was the first evidence-based study in human–equine interaction to measure a change in participants' cortisol.

Because the children weren't specifically autistic, this is supporting evidence rather than proof for autistic children — but it points to a real biological pathway behind the calm that families describe, and it is exactly the kind of mechanism the horse trials above suspect is at work.

The evidence at a glance

Study	Who & design	What it measured	What it found
O'Haire et al., 2015	114 children aged 5–12 (autistic + peers) in small groups	Skin conductance — a physiological measure of anxious arousal	Autistic children's arousal dropped significantly when guinea pigs were introduced
Gabriels et al., 2015	Randomised trial, 127 autistic children/teens 6–16, 10 weeks, riding vs no-horse control	Irritability & hyperactivity (Aberrant Behavior Checklist)	Irritability improved (ES 0.42) and hyperactivity (ES 0.45); effects began by week 5
Gabriels et al., 2018	6-month follow-up of the same trial (subset of 64)	Whether the gains lasted	~73% of the irritability gain preserved (ES 0.32); social/communication gains sustained
Pendry & Vandagriff, 2014	Randomised trial, 131 children (general, not autism-specific), 11-week equine programme vs waitlist	Salivary cortisol (stress hormone)	Programme group had significantly lower cortisol across the day
O'Haire et al., 2014	64 autistic children, 15 schools, 8 weeks of classroom animals	Parent- and teacher-rated social skills	Significant gains in social skills and less social withdrawal versus a control period

Effect sizes here are small-to-moderate. That is normal for gentle, complementary supports and does not mean "no effect" — it means a real, measurable shift on top of everything else a child has going on.

How to read all this honestly

Two things are true at once. First, more than one type of study — physiological, behavioural and hormonal — points the same way, which is more convincing than any single result. Second, the field is still young: samples are often small, some measures rely on parent or teacher reports, and much of the horse research blends riding with ground-based work, so we can't always separate the two. Newer trials, including a 2024-funded study comparing riding with grooming, are designed to tighten exactly these gaps. Our full evidence guide goes deeper →

What that means for a family is simple: animal therapy is a reasonable, gentle thing to try for anxiety and regulation, with genuine evidence behind it — provided you hold it as one support among several, not a replacement for medical help when a child needs it.

This page is general information, not medical advice. Animal-assisted and equine activities are a promising complementary support for wellbeing, confidence and self-regulation — they are not a treatment, an assessment or a cure for autism or for an anxiety disorder. If a child's anxiety is severe, persistent or affecting daily life, speak to a GP, paediatrician or CAMHS. We will update this page as new trials report.

Sources

O'Haire ME, McKenzie SJ, Beck AM, Slaughter V — "Animals may act as social buffers: Skin conductance arousal in children with autism spectrum disorder in a social context", Developmental Psychobiology, 2015 (114 children aged 5–12; NIH/NICHD-funded) — doi.org/10.1002/dev.21310 · plain-English summary via ScienceDaily / NIH
Gabriels RL et al. — "Randomized Controlled Trial of Therapeutic Horseback Riding in Children and Adolescents With Autism Spectrum Disorder", Journal of the American Academy of Child & Adolescent Psychiatry, 2015 (N=127; irritability ES 0.42, hyperactivity ES 0.45; effects by week 5) — PMC4475278 · PubMed 26088658
Gabriels RL et al. — "Long-Term Effect of Therapeutic Horseback Riding in Youth With Autism Spectrum Disorder: A Randomized Trial", Frontiers in Veterinary Science, 2018 (6-month follow-up; irritability ES 0.32, ~73% of gain preserved) — PMC6054954
Pendry P, Vandagriff JL — "Randomized Trial Examines Effects of Equine Facilitated Learning on Adolescents' Basal Cortisol Levels", Human-Animal Interaction Bulletin, 2014 (N=131; significantly lower cortisol) — CABI Digital Library
O'Haire ME et al. — "Effects of Classroom Animal-Assisted Activities on Social Functioning in Children with Autism Spectrum Disorder", 2014 (64 children, 15 schools, 8 weeks) — PMC3948601

Last reviewed: 23 July 2026 · Stable Ground

A calmer hour, at the pace your child sets

Stable Ground is building gentle, one-to-one animal-assisted and equine sessions for autistic children across East Renfrewshire, Renfrewshire and North Ayrshire — designed around regulation, safety and trust. No diagnosis needed.

For families & carers For professionals