Adults With Profound Autism May Retain Room to Grow in Daily Living Skills

Adults with profound autism are often assumed to have reached the limits of what they can learn, but a new study suggests that assumption may be selling them short. Researchers analyzing item-level data from adaptive behavior interviews found that every one of the 36 autistic adults studied had at least one daily living skill they could plausibly perform if given enough opportunity to learn and practice it. The finding, published in the Journal of Autism and Developmental Disorders, challenges the widespread belief that skill development effectively stops in adulthood for people with the highest support needs.

The study focused on daily living skills, the everyday abilities that underpin independence: personal care such as hygiene and health routines, domestic tasks like cooking and cleaning, and community activities such as handling money or crossing the street safely. These skills are closely tied to how much support a person needs in adult life, yet few studies have examined them in adults with profound autism, a subgroup defined by intellectual disability and, in many cases, minimal or no spoken language.

To get past the limitations of standard summary scores, the research team applied a coding framework developed by Teh and colleagues to interview data from the Vineland Adaptive Behavior Scales, Third Edition. Parents and staff members of the participants, who averaged 35 years of age, answered detailed questions about each skill the adult was rated as completing independently never or seldom. For every such item, coders assigned one of five reasons explaining why the skill was not achieved, ranging from genuine cognitive inability to lack of opportunity or simple reliance on reminders.

The results revealed strikingly different barriers across skill domains. The distribution of reason codes differed significantly across the Personal, Domestic, and Community subdomains, with a moderate-to-large effect size. Domestic skills, such as locking doors, washing food before cooking, and operating appliances, were significantly more likely to be rated as achievable with reminders, suggesting that prompts, timers, or visual schedules could unlock independence in these areas. Community skills, by contrast, were significantly more likely to be coded as beyond current cognitive ability, and five participants received that code for every community item.

Personal skills fell somewhere in between, but with an important twist. Items in this domain were more often coded as unable for other reasons, pointing to sensory sensitivities, co-occurring conditions such as anxiety or obsessive-compulsive disorder, and motor challenges as hidden obstacles to independence in hygiene and health-related behaviors. The skills most frequently rated as never or seldom performed independently included health awareness and appropriate hygiene when coughing or sneezing, and for these items cognitive inability was the most common explanation offered by informants.

When the researchers recalculated scores to reflect skills that informants believed were attainable with adequate support, the so-called target scores rose significantly above original scores for all three subdomains, with moderate-to-large effects. The gains translated into small increases in v-scale scores and small-to-moderate gains in age equivalents, though after correction for multiple comparisons only the target Personal v-scale scores and target Domestic age equivalents remained statistically significant. Domain-level standard scores showed a small-to-moderate gain that did not reach significance, underscoring how blunt summary measures can be.

That bluntness is itself a central finding. More than three-quarters of the sample had a daily living skills standard score of 20, the lowest possible score, before target points were assigned, and accounting for potential improvement raised the standard score in only 13 percent of cases. Yet adjustments to subdomain age equivalents changed results for the vast majority of participants, affecting 97 percent of Personal age equivalents, 82 percent of Domestic, and 61 percent of Community. In other words, the standard scoring system was masking meaningful variation that item-level analysis could reveal, variation that may be exactly what clinicians need to plan realistic, individualized goals.

Three case studies illustrate just how varied those profiles can be. Lionel, a 33-year-old man with flexible phrase speech and a nonverbal ratio IQ of 56, received mostly codes indicating lack of opportunity or need for reminders; his parents believed the only skills truly beyond his cognitive reach were setting long-term goals, managing household finances, and using technology to communicate. His target standard score rose by 10 points, and his Personal age equivalent jumped by more than six years. Blake, a 26-year-old minimally speaking man with a nonverbal ratio IQ of 77, showed an even spread of codes, with nearly all cognitively impossible items confined to the community domain. Velda, a 35-year-old non-speaking woman with epilepsy and a ratio nonverbal IQ of 10, received mostly codes of cognitive inability, yet her parents identified skills she could still learn, such as spreading food with a table knife and sweeping, and reported that she had even lost some previously mastered abilities, possibly from lack of practice.

The authors caution that the target-score approach is exploratory and rests entirely on informant reports, primarily from parents, without independent verification of the reported barriers. The sample was also small, predominantly White, and included few female participants, and no data were available on the services participants were receiving or whether skill improvement was an explicit goal of those services. Because participants could not self-report due to co-occurring intellectual disability, the study also could not capture the adults’ own perspectives on their abilities and goals. Future intervention studies, the researchers note, are needed to test whether adults with profound autism can actually acquire the skills flagged as attainable.

Even with those limitations, the study carries a pointed message for clinicians, educators, and families: meeting criteria for profound autism should not preclude efforts to teach new skills. The finding that quality-of-life scores showed no association with daily living skills scores, contrary to earlier literature, likely reflects measurement tools that were never designed for this population, the authors argue, calling for better instruments and more explicit guidance on when existing measures are appropriate. Meanwhile, the item-level coding method offers a relatively brief way to extract clinically useful information from a measure many families already complete. For a group long excluded from intervention research, the message is that learning does not necessarily end at adulthood, and that the right question is not whether a person can be taught, but which specific skill, with which specific support, should come next.

Subject of Research: Daily living skills and intervention targets in adults with profound autism

Article Title: Potential for Life Skills Growth in Adults With Profound Autism

Article References: Clarke, E. B., Wellmann, M., Kubis-Badilla, J., Lord, C., & Bal, V. H. (2026). Potential for Life Skills Growth in Adults With Profound Autism. Journal of Autism and Developmental Disorders. https://doi.org/10.1007/s10803-026-07488-w

Image Credits: AI Generated

DOI: 10.1007/s10803-026-07488-w

Keywords: profound autism, daily living skills, adaptive behavior, Vineland Adaptive Behavior Scales, adults with autism, intellectual disability, intervention targets, item-level analysis, autonomy, quality of life, caregiver report, Journal of Autism and Developmental Disorders