Assessment & Research

Understanding Self-Compassion in Autistic Adults: Validity Evidence and Its Links to Loneliness and Depression Across Autistic and Non-Autistic Individuals.

Cai et al. (2026) · Autism research : official journal of the International Society for Autism Research 2026
★ The Verdict

The Self-Compassion Scale is valid for autistic adults, yet self-kindness does not shield them from loneliness-driven depression like it does for neurotypicals.

✓ Read this if BCBAs writing mental-health protocols or doing ACT with autistic adults.
✗ Skip if Clinicians who only serve young children or focus on skill acquisition.

01Research in Context

01

What this study did

Ying and colleagues asked the adults—half autistic, half non-autistic—to fill out the Self-Compassion Scale.

They used math tests to see if the scale works the same way in both groups.

They also checked whether being kind to yourself softens the link between feeling lonely and feeling depressed.

02

What they found

The scale keeps its two-factor shape in both groups, so you can trust the numbers.

For non-autistic adults, harsh self-talk made loneliness turn into depression faster.

For autistic adults, self-compassion did not act as a buffer—loneliness still led to depression no matter how they scored.

03

How this fits with other research

Werner et al. (2025) showed that autistic youth feel lonely because social moments feel bad, not because they spend less time with peers. Ying moves the lens to adults and adds that self-kindness does not protect them from that loneliness turning into depression.

Howard et al. (2023) found poor sleep and low social connection both predict depression in autistic young adults. Ying agrees that social pain matters, but shows an inner trait—self-compassion—fails to moderate that pain, pointing to a different intervention target.

Hedley et al. (2023) validated a new suicidal-thought scale for autistic adults. Like Ying, they confirm that existing mental-health tools can be used with confidence, yet warn that risk factors may operate differently in this group.

04

Why it matters

If you run CBT or ACT groups, do not assume teaching self-compassion will block the depression path for autistic clients. Keep the scale to track self-talk, but pair it with social-belonging or sleep interventions shown by L et al. and Natalie et al. Target real-world inclusion and concrete coping skills instead of inner warmth alone.

Free CEUs

Want CEUs on This Topic?

The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.

Join Free →
→ Action — try this Monday

Add a peer-belonging goal or sleep-hygiene plan next to any self-compassion exercise for your autistic client.

02At a glance

Intervention
not applicable
Design
other
Sample size
573
Population
autism spectrum disorder, neurotypical
Finding
mixed

03Original abstract

Autistic adults face higher rates of loneliness and depression than non-autistic adults. Self-compassion may offer a protective buffer against mental health difficulties, but its measurement validity and interaction with loneliness have not been studied in autistic populations. This two-part study examined (1) the dimensional structure of the Self-Compassion Scale (SCS) in a global sample of autistic (n = 377) and non-autistic (n = 196) adults, and (2) whether self-compassion moderates the relationship between loneliness and depression in both groups. Confirmatory factor analyses tested multiple models of the SCS, and multigroup regression models tested moderation effects using loneliness and depression scores. The SCS was best represented by two factors-compassionate and uncompassionate self-responding-in both autistic and non-autistic groups. Measurement invariance was supported. In moderation analyses, uncompassionate self-responding significantly moderated the relationship between loneliness and depressive symptoms among non-autistic adults, but not autistic adults. Uncompassionate self-responding was significantly associated with greater depression symptoms in both groups. These findings support using a two-factor structure of the SCS in autistic samples and suggest that reducing uncompassionate self-responding may benefit mental health broadly. However, self-compassion did not buffer the loneliness-depression link for autistic adults, highlighting the need for alternative protective factors tailored to this population.

Autism research : official journal of the International Society for Autism Research, 2026 · doi:10.1002/aur.70150