The research

The Evidence Behind Our Approach

Families travel a long way for this therapy, so you deserve to see what the science actually says — including where it is still thin. Every study below is summarised in plain English, with its limits stated, and linked so you can read the original yourself.

Reviewed by the Bright Beginnings Pediatric Therapy team·Last reviewed: July 22, 2026

10 studies 2 randomized controlled trials 1 meta-analysis Published 2010–2025

How to read this page. Not all evidence carries the same weight. A randomized controlled trial compares treatment against a placebo and is the most reliable form here. An open-label study has no comparison group. A review summarises other people's work. A case report describes a handful of individuals and proves the least. Each study is labelled, and we have said what it doesn't show as well as what it does.

Laser Therapy for Autism

The strongest evidence for our signature service. Two randomized trials, one open-label study, and the reasoning behind them.

4 studies

Randomized controlled trial 30 children · ages 2–6

Near-Infrared Light Lowered Autism Symptom Scores Against a Sham Control

Children received pulsed near-infrared light (850 nm, 40 Hz) twice a week for eight weeks. Symptom scores on the CARS-2 scale fell 7.2 points further in the treated group than in the sham group (95% CI 2.4–12.1, p=0.011). 87% of treated children improved by at least 4.5 points — the threshold considered clinically meaningful — against 35% given the sham. No moderate or severe side effects were reported.

What it doesn't show: 30 children is a small trial. The authors call for larger studies to confirm it.

Frontiers in Neurology · 2024 Read the study

Open-label study 22 children · ages 2–7

An Eight-Week Course of Light Therapy Tracked a 7-Point Drop in Symptom Scores

Using the same 850 nm, 40 Hz protocol, average CARS-2 scores moved from 36.5 to 29.7 (p<0.0001). EEG recordings showed a matching shift in brain activity, and the size of that shift tracked the number of sessions a child received.

What it doesn't show: No control group, so normal development over the eight weeks cannot be separated from the treatment effect. There was no follow-up to see whether gains held.

Frontiers in Child and Adolescent Psychiatry · 2025 Read the study

Randomized controlled trial 40 children and teens · ages 5–17

Eight Brief Laser Sessions Reduced Irritability, and the Effect Held at Follow-Up

Eight five-minute sessions over four weeks, applied at the base of the skull and the temples. Irritability scores on the Aberrant Behavior Checklist fell 15.2 points relative to placebo (p<0.0001), and the improvement was still present at the week-8 follow-up.

What it doesn't show: A small trial measuring behaviour only — no brain-based measures were taken.

Advances in Experimental Medicine and Biology · 2018 Read the study

Review article Review of existing evidence

Why Researchers Think Light Therapy May Help — the Proposed Mechanisms

A review setting out three ways photobiomodulation might act in autism: rebalancing connectivity between brain regions, reducing inflammation and neuronal loss, and shifting the gut microbiome. Useful for understanding the reasoning behind the therapy.

What it doesn't show: This is a hypothesis paper, not evidence of benefit. The authors note no animal studies had yet tested transcranial light therapy in autism models, and call for proof-of-concept work.

Neurology International · 2022 Read the study

Laser Therapy and the Brain

How photobiomodulation is thought to work, and what is — and is not — established in brain injury and rehabilitation.

2 studies

Review article Review · brain injury and stroke

How Light Reaches the Brain, and What Is Known About Injury Recovery

A thorough review of transcranial low-level laser therapy for traumatic brain injury and stroke. It explains the core mechanism — light absorbed by cytochrome c oxidase in mitochondria, raising ATP production — and finds animal evidence strongest at 660 and 810 nm with pulsed delivery.

What it doesn't show: Human evidence is thin: mostly case reports and one 11-person open study. Of three large stroke trials, the third was stopped early for futility. Optimal dosing is still unsettled.

Photomedicine and Laser Surgery · 2016 Read the study

Review article Review · neurorehabilitation

A Broad Look at Low-Level Laser Therapy in Neurological Rehabilitation

Covers stroke, brain injury, spinal cord injury and nerve regeneration. Pooled acute-stroke data across 778 patients showed significant improvement (p=0.003). The authors highlight a favourable safety record, reporting no adverse events directly attributable to the therapy.

What it doesn't show: Leans heavily on animal studies and small human series. Wavelength and dosing vary so widely between studies that they are hard to compare.

PM&R · 2010 Read the study

Primitive Reflexes and Development

Why we assess retained reflexes, and how honestly the research supports it.

3 studies

Meta-analysis 4 studies · 229 participants

Retained Tonic Neck Reflexes Are Consistently Linked with ADHD Symptoms

Pooling four studies, retained symmetric tonic neck reflexes correlated moderately with ADHD symptoms (r=0.39, 95% CI 0.25–0.52) — a consistent association across the available research.

What it doesn't show: Correlation only. The authors state plainly that this cannot establish that retained reflexes cause ADHD symptoms, and too few studies existed to analyse subtypes or medication effects.

Frontiers in Psychiatry · 2023 Read the study

Controlled study 60 participants · ages 5–35

Reflex Patterns Tracked with Measurable Differences in Brain Connectivity

Using EEG across 19 scalp sites, participants with autism and retained reflexes showed long-range underconnectivity alongside short-range overconnectivity. After an intervention aimed at reducing those reflexes, connectivity measures moved toward more developmentally typical patterns.

What it doesn't show: A modest sample spanning a very wide age range, and the paper does not set out its own limitations in detail.

Brain Sciences · 2023 Read the study

Case report — weakest evidence 3 individuals

Three Detailed Cases of Reflex Integration, Reported with Cognitive Gains

A twelve-week programme of reflex exercises, nerve stimulation and sensory work. All three individuals moved from retained to fully integrated on the reflexes tested, with IQ gains recorded alongside changes in EEG organisation.

What it doesn't show: Three people, no control group. The authors themselves note that best practice would leave a year between testing and retesting rather than twelve weeks. Treat this as a description of what happened for three individuals, not as evidence of what to expect.

BMJ Case Reports · 2023 Read the study

Why We Look at the Whole Body

The basis for a brain-body assessment that considers inflammation, sleep and gut health alongside behaviour.

1 study

Review article Review · post-mortem and imaging studies

Evidence of Immune Activation in the Autistic Brain

Reviews post-mortem findings of activated microglia — the brain's immune cells — in the cortex and cerebellum, and links these to the loss of long-distance neural connections seen in autism. This is part of why our assessment looks beyond behaviour to inflammation, sleep and gut health.

What it doesn't show: Built on a small number of post-mortem samples. The authors note that therapies targeting this inflammation have not been tested in autism.

Neuron Glia Biology · 2011 Read the study

Questions About Any of This?

If you would like to talk through what the research means for your child specifically, the 15-minute consultation is free and there is no obligation.

Free 15-Minute Consultation

These studies are shared for information and are not a promise of results. Research on photobiomodulation in children is still at an early stage — most trials are small, and several of the papers above are reviews or case reports rather than controlled trials. Nothing here is medical advice or a substitute for discussing your child's care with a qualified clinician. Outcomes differ from child to child.