Innovative Software Developed by Parent and Statistician Enhances Autism Skills Therapy

In a remarkable intersection of personal experience and scientific expertise, Mark Ramos, an assistant research professor at Pennsylvania State University, has created a free software tool aimed at improving therapy outcomes for children with autism spectrum disorder (ASD). Inspired by his daughter Mabel's developmental journey, Ramos designed the software to help therapists more accurately assess and verify skill mastery during therapy sessions.

Mabel, now five years old, was diagnosed with ASD and initially faced significant developmental challenges, including the inability to speak or sit independently at age two. Through dedicated therapies such as physical, occupational, and Applied Behavior Analysis (ABA), her development has been steadily progressing, reaching milestones typical of a younger age.

A key part of Mabel's therapy involves Discrete Trial Training (DTT), a systematic method that breaks down skills into small, manageable steps. For example, teaching hand washing involves sequential components such as approaching the sink, turning on water, applying soap, and drying hands—each practiced repeatedly until mastered.

Mark noted a disconnect between the performance thresholds that define mastery and the actual level of skill demonstrated by children. In traditional settings, success is often measured by a child satisfying performance criteria like 80% accuracy, but this doesn't necessarily reflect true mastery. Recognizing this, he applied statistical methods to better estimate a child's likelihood of future success—transitioning from performance percentiles to a probability-based understanding of mastery.

To address this, Ramos developed MIEBL (Measurement of Individualized, Evidence‑Based Learning), a free software that employs Bayesian probability models. When users input specific performance goals and trial numbers, MIEBL calculates the estimated mastery level, guiding therapists in setting appropriate benchmarks. For instance, to ensure a child consistently performs a skill at a 80% mastery level, the software can recommend adjusting success criteria to 90%, considering statistical margins.

Ramos emphasizes that this tool helps verify whether children truly meet targeted skills, allowing therapists to optimize therapy plans based on data-driven insights. He shared the software with Mabel's therapists, hoping it will become a routine part of autism intervention practices.

While acknowledging that DTT already incorporates careful skill thresholds, Ramos sees this as an enhancement rather than a critique. His ultimate goal is to help more children like Mabel achieve their developmental potential through more precise and scientifically grounded therapy practices.

Reflecting on his personal motivation, Ramos expressed that working on the software has been a meaningful experience, both professionally and personally. Seeing his daughter grow into a lively, music-loving girl who enjoys dancing brings him great joy, as he continues to refine tools that may benefit other children with ASD in the future.