A research team led by Paige E. Cervantes of Virginia Commonwealth University, together with Robert D. Gibbons of the University of Chicago, Cecilia B. Becker, Yehuda Cohen, and Sarah M. Horwitz of NYU Grossman School of Medicine, has now published an evaluation in the Journal of Autism and Developmental Disorders of how one promising tool, the K-CAT, performs with autistic youth. The K-CAT is a computerized adaptive test originally developed to close the detection gap in the general population, particularly in primary care and schools where specialized psychiatric expertise is scarce. Its designers built it on multidimensional item response theory, a psychometric framework that models multiple latent symptom traits simultaneously, allowing the test to measure several distinct but correlated dimensions of psychopathology at once and to yield more nuanced scores than traditional unidimensional questionnaires.

The technical elegance of the K-CAT lies in its adaptive algorithm. Rather than administering a fixed questionnaire, the platform draws a small but statistically optimal, individualized set of items from a large item bank and continues until a target precision level is reached, with each scale taking roughly two minutes or less. Both a child and a parent version exist, covering depression, anxiety, ADHD, mania, conduct disorder, and oppositional defiant disorder, while the child version adds a suicide scale plus modules for post-traumatic stress disorder and substance use. Items are written at a fourth-grade reading level and can be presented auditorily. In its original validation against gold-standard structured clinical interviews such as the Kiddie Schedule for Affective Disorders and Schizophrenia, the tool achieved area under the curve values from 0.83 for generalized anxiety disorder to 0.92 for major depressive disorder, and an extraordinary 0.996 for suicidal ideation. Crucially, however, autistic youth were excluded from that validation sample.

That exclusion matters because prior research has documented that measures built for the general population often lack psychometric evidence and raise accessibility problems when used with autistic people. Common pitfalls include confusing terminology and figures of speech, imprecise response options, anxiety about giving a perfectly accurate answer, and failure of items to capture the intended domain in the context of autism. Earlier phases of this research program, including a pilot study of the K-CAT suicide scale and a mixed-methods evaluation of community perspectives, found largely positive impressions but recurring concerns about linguistic clarity and symptom overlap, meaning items that may describe autism features rather than the mental health condition they are meant to detect. Differential item functioning analyses went further, flagging nearly a quarter of evaluated items as showing questionable or serious DIF, indicating they do not discriminate between high and low symptom levels in autistic youth the way they do in non-autistic youth.

The new study extended this work by asking how the full K-CAT battery performs across the heterogeneity of the autism spectrum. One hundred fifty-five autistic youth aged 7 to 17 without intellectual disability, recruited primarily from an academic safety-net health system in the Southeastern United States as part of a National Institute of Mental Health-funded effort, completed the K-CAT child version along with comparison measures, while their caregivers completed the parent version. Receptive language was estimated with the NIH Toolbox Picture Vocabulary Test, an adaptive measure in which participants hear a spoken word and select the matching picture from four options, and autism features were rated with the Autism Spectrum Rating Scale short form. All assessments were administered remotely through secure platforms, with the K-CAT results flowing into the REDCap data system via an application programming interface.

Administration data revealed a clear pattern. Fifty-nine percent of youth completed the K-CAT with a research team member rather than independently, and twenty-seven percent received accommodations such as having items read aloud. Youth who needed accommodations were significantly younger, averaging about 10.4 years versus 12.5 years, and had significantly lower receptive vocabulary scores. Younger participants also took longer to finish, with age negatively correlated with completion time. Children averaged just over ten minutes and about seventy items for the full battery, while caregivers averaged under six minutes and about forty-seven items. When the researchers adjusted for the later-added PTSD module and compared with the original validation study, completion times were essentially unchanged, but autistic youth and their caregivers were presented with more items, a median of sixty versus fifty-one for children and forty-seven versus thirty-eight for caregivers. Because the algorithm stops once precision is reached, needing more items suggests the items themselves carry less information about symptom severity in this population.

The K-CAT did identify a broad and plausible range of mental health concerns. On paired assessments combining parent and child report, moderate or severe ratings were most common for ADHD symptoms at just over sixty-three percent, followed by mania at about fifty-four percent, anxiety at forty-seven percent, oppositional defiant disorder at forty-three percent, and conduct disorder and depression at roughly thirty-four percent each. Child-only results placed ADHD, suicide, mania, conduct disorder, and PTSD symptoms at the top, while parent-only results again led with ADHD, anxiety, and mania. With the exception of the notably high mania rates, this pattern mirrors existing prevalence research, and the elevated suicide and trauma symptom rates align with prior findings. The authors stress, however, that these results were not validated against clinical diagnoses in this sample.

More troubling were the subgroup differences. On the child version, language level emerged as the dominant factor: youth scoring in the moderate or severe range on five of eight modules, including ADHD, suicide, mania, conduct disorder, and depression, had significantly lower receptive vocabulary scores. Because language level was unrelated to parent-version results and low language predicted poorer parent-child agreement, the authors suspect comprehension difficulties rather than genuinely higher symptom burden. Conversely, autism severity, measured by the ASRS, was significantly associated with every parent-version module but with none of the child-version modules, and higher ASRS scores predicted fewer parent-version items administered. This asymmetry hints at insufficient differentiation between autism features and psychiatric symptoms within the instrument, consistent with community feedback about symptom overlap. Parent-child agreement itself was low to moderate overall, ranging from a correlation of 0.17 for ADHD to 0.43 for oppositional defiant disorder, and it fractured along subgroup lines: agreement was stronger for externalizing modules in adolescents, but for youth with lower vocabulary scores only the depression module showed a significant correlation.

There were genuine bright spots. Convergent validity held up well, with the K-CAT depression module correlating at 0.74 with the child Mood and Feelings Questionnaire and 0.73 with the parent version, and the anxiety module correlating at 0.63 and 0.72 with the corresponding SCARED versions. The K-CAT suicide scale agreed with the four-item Ask Suicide-Screening Questions on suicide risk in about eighty-one percent of cases, and when the two disagreed, the K-CAT flagged risk more often. These results were stable across age, language, and autism severity subgroups. Still, the authors caution that none of the comparison measures were developed specifically for autistic youth, and no gold-standard psychiatric assessment currently exists for this population, so encouraging correlations cannot substitute for autism-specific validation.

The team is candid about limitations, including caregiver-reported diagnoses, the absence of multiplicity correction across some sixty statistical tests, self-selected participation, and the exclusion of youth with intellectual disability or significant communication needs, whose complex mental health needs future tools must also serve. The path forward is already mapped: the researchers plan to systematically revise the K-CAT at the module and item level with an advisory panel of autism community members, pilot a modified K-CAT Autism Version, and validate it against clinicians trained in co-occurring autism and mental health conditions. If those adaptations succeed in improving accessibility, linguistic clarity, and diagnostic specificity, the authors argue the platform could transform scalable and equitable mental health screening for autistic youth, combining its wide coverage of psychopathology, minimal respondent burden, and integrated multi-informant diagnostic predictions in a way no current instrument can match.

Subject of Research: Evaluation of the K-CAT computerized adaptive test for assessing psychiatric symptoms in autistic youth

Article Title: Assessing Psychiatric Symptoms in Autistic Youth Using the K-CAT®

Article References: Cervantes, P. E., Gibbons, R. D., Becker, C. B., Cohen, Y., & Horwitz, S. M. (2026). Assessing Psychiatric Symptoms in Autistic Youth Using the K-CAT®. Journal of Autism and Developmental Disorders. https://doi.org/10.1007/s10803-026-07454-6

Image Credits: AI Generated

DOI: 10.1007/s10803-026-07454-6

Keywords: autism, K-CAT, psychiatric assessment, mental health screening, computerized adaptive testing, item response theory, suicide risk, anxiety, depression, ADHD, differential item functioning, autistic youth

The research, published in the Journal of Child Psychology and Psychiatry, examined whether a technique called cognitive reappraisal could alter neural reactions to aversive sensory stimulation in autistic youth. Cognitive reappraisal is a cornerstone strategy of cognitive behavioral therapy, a widely used psychological treatment best known for helping people manage anxiety and difficult emotions. The idea behind reappraisal is deceptively simple: by deliberately changing the way you think about something, you can change the way it feels. Instead of suppressing an emotional reaction or simply enduring an unpleasant stimulus, a person reframes the situation mentally, shifting its meaning and, with it, its emotional weight.

To test whether this strategy works on raw sensory experience rather than just emotional interpretation, the UCLA team designed an experiment with real physical stakes. The study included 47 autistic children and teenagers and 45 typically developing youth, all between the ages of 8 and 15. While lying inside a magnetic resonance imaging scanner, participants had genuinely unpleasant textures, including strips of artificial grass and scratchy wool, rubbed gently against their forearms. The scanner captured activity throughout the brain as the textures were applied, allowing researchers to observe in real time how the participants’ brains reacted to the uncomfortable sensations under different mental conditions.

The key twist was how the children were prepared before entering the scanner. Each participant was taught the reappraisal technique through what the researchers called the reporter game, a child-friendly framing of the same cognitive strategy adults learn in therapy. On some trials, children were asked to simply focus on how the texture felt, attending fully to the sensation. On other trials, they were instructed to describe the texture objectively, sticking strictly to the facts the way a news reporter would: this is artificial grass, it is stiff, the blades are rough and plastic-like, it is moving across my arm. The goal was to shift attention from the felt experience of the texture to a detached, factual account of it, changing how the brain processed the incoming sensory information.

The results were striking. When autistic participants played the reporter game, activity in the brain’s sensory regions dropped significantly compared with trials in which they focused on how the texture felt. In other words, the simple act of thinking like a reporter measurably quieted the neural circuits that respond to aversive touch. This finding matters because it suggests that sensory over-responsivity in autism, often assumed to be a fixed feature of the nervous system, may be more malleable than previously thought, and that top-down cognitive strategies can reach down into sensory processing itself.

Both groups of children also showed increased activity in frontal brain regions, areas at the front of the brain known to help regulate emotional responses. This pattern is consistent with what cognitive reappraisal is supposed to do: recruit executive control networks to modulate lower-level reactions. Interestingly, this frontal activation was stronger in older autistic children compared with younger autistic participants, hinting that the regulatory machinery engaged by reappraisal continues to mature through adolescence and that developmental timing may shape how well the strategy works.

That developmental pattern carried an important caveat. Senior author Shulamite Green, associate professor-in-residence of psychiatry and biobehavioral sciences at UCLA, noted that typically developing children may begin using frontal regions of the brain to regulate sensory responses at younger ages, while autistic children may show a developmental delay in this type of regulation. The implication is that older autistic youth may be able to deploy the strategy more effectively than younger ones, and that clinicians and parents working with younger children might need to adjust expectations or adapt the approach to fit the child’s developmental stage.

An equally intriguing discrepancy emerged when the researchers looked at how unpleasant the children actually rated the textures. Only the typically developing youth reported that the textures felt less unpleasant when they used the reporter game strategy. The autistic children showed significant changes in brain activity but did not report the sensations as any less aversive. Green offered a possible explanation: difficulties with cognitive flexibility, common in autism, may make it hard to shift one’s rating of the same sensation within a short span of time. She also suggested that longer-term intervention may be needed before changes in brain activity translate into a subjective change in how a sensation feels. The brain may be adapting beneath the surface before the person becomes consciously aware of any difference.

The practical appeal of the finding lies in its accessibility. Cognitive reappraisal is already a standard component of cognitive behavioral therapy, which means clinicians would not need to invent or validate an entirely new treatment to begin incorporating sensory-focused reappraisal into existing care. As Green put it, cognitive behavioral therapy interventions are already known to work for anxiety and emotional responses, so there is no need to reinvent the wheel. Adapting a proven, widely available technique for one of autism’s most burdensome symptoms could be far faster and cheaper than developing a therapy from scratch.

The researchers were careful to note the limits of the evidence. The measured responses came from a single scan session, so the study cannot show whether practicing the reporter game repeatedly over time would ease sensory difficulties in the messy reality of daily life, at school, at home, or among friends. Green said the next step is to test the strategy as a proper intervention, following children over weeks or months to see whether the neural changes observed in the scanner grow into lasting, real-world relief. If that research succeeds, a game familiar to any child who has watched a news report could become a routine part of helping autistic children navigate a world that often feels too loud, too rough, and too bright.

Subject of Research: Cognitive reappraisal training and its effects on brain responses to aversive sensory stimulation in autistic children and teenagers.

Article Title: Teaching autistic children to ‘think like a reporter’ calms overly sensitive brain activity

Article References: Teaching autistic children to ‘think like a reporter’ calms overly sensitive brain activity. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: autism, sensory over-responsivity, cognitive reappraisal, cognitive behavioral therapy, UCLA Health, MRI, Journal of Child Psychology and Psychiatry, sensory processing, frontal brain regions, reporter game, autistic youth, Shulamite Green