Digital DTT efficacy with FirstWork
A controlled single-subject study conducted with Possibilities ABA and Augustana University examined how FirstWork affects the rate and quality of receptive-identification learning in a 3-year-old learner with ASD. Presented at ABAI 2025.
What the study examined
A 3-year-old male with ASD and limited listener discrimination skills was enrolled in a multiple-baseline study across three receptive-identification targets: "foot," "dog," and "bus." Under traditional flashcard-based DTT (baseline), the learner showed near-zero correct responding across all three targets. The study then introduced FirstWork as the delivery medium and tracked acquisition, efficiency, and response latency to mastery.
The study was implemented by trained RBTs and BCBAs from Possibilities ABA in standard ABA clinic therapy rooms during scheduled DTT sessions.
Design and intervention
The study used FirstWork's receptive-identification modality. The app delivered the SD via replayable audio, presented a 4-item photographic array, and ran full error correction automatically on every trial.
Baseline condition
- ■Traditional picture-card DTT
- ■No digital supports
- ■0-10% correct responding across all targets
Intervention condition
- ✓FirstWork app on tablet
- ✓4-item photographic array per trial
- ✓Replayable audio SD
- ✓Full automated error correction
- ✓Multiple exemplars per target
Research questions
- 1.Acquisition: How does FirstWork affect the rate and level of correct responding for novel receptive-identification targets?
- 2.Efficiency: What are the trial-count, session-duration, and latency metrics associated with mastery?
- 3.Maintenance: Once acquired, how rapidly does the learner respond to mastered targets?
Baseline to mastery across all three targets
Squares represent traditional flashcard baseline. Circles represent FirstWork intervention. Each target introduced sequentially per the multiple-baseline design.
All three targets showed near-zero responding at baseline (squares). Following FirstWork introduction (circles), each target progressed to 90%+ accuracy. "Foot" mastered fastest; "Dog" required the most sessions due to its introduction with a broader array of distractors.
Lesson speed and mastery rate
Total trials to mastery, total minutes of learning, and per-trial completion time (mean and median) across all three targets.
"Bus" was mastered in under 7 minutes and 60 trials. "Dog," the most challenging target, required approximately 27 minutes and 145 trials. Median trial completion time was 3 seconds across all three targets, indicating fluent responding post-mastery.
| Target | Total trials to mastery | Minutes to mastery | Mean trial time (s) | Median trial time (s) |
|---|---|---|---|---|
| Foot | 105 | 19.8 | 11.3 | 3 |
| Dog | 145 | 26.7 | 11.9 | 3 |
| Bus | 60 | 6.3 | 6.3 | 3 |
Session-level accuracy and response time per target
Blue line shows accuracy per session. Orange bars show response time in seconds. Elevated mean latencies early in learning reflect error-correction trials; the 3-second median indicates fluent responding once mastered.
Low median latencies (3 seconds) across all targets indicate fluent, confident responding following mastery. Elevated mean latencies during early sessions reflect error-correction trials rather than slow responding. "Bus" shows the most compressed learning curve.
What the data shows
All three targets moved from 0-10% baseline accuracy to 90%+ mastery within 8-12 FirstWork sessions — a trajectory not achieved under traditional card-based DTT.
Median response latency of 3 seconds across all targets post-mastery. Elevated means reflect instructional error-correction trials, not slow or uncertain responding.
Automated error correction ensured identical implementation across every trial, every session — eliminating variability introduced by manual RBT delivery.
What this means for ABA practice
Rapid skill acquisition
Digital DTT with automated prompts accelerates learning compared to typical paper-card formats, reducing therapist prep time and trial-delivery load without sacrificing clinical rigor.
High procedural fidelity
Automated error correction ensures consistent implementation, mitigating human variability across RBTs and sessions — a persistent challenge in manual DTT delivery.
Resource optimization
Clinics can reallocate saved prep time and material costs to expanded caseloads or program development — directly addressing growing ABA waitlists.
Data reliability
Automatic trial-by-trial logging eliminates manual recording errors. Cross-verification with RBT/BCBA records confirmed accuracy of app-generated data throughout the study.
Scope and replication
Single-subject design
Results require replication across multiple learners, ages, and diagnostic profiles to establish generality. A scaled-up study is currently in progress.
Short-term assessment
Maintenance and generalization measurements could be expanded over a longer time horizon to evaluate skill retention beyond the mastery window.
Future research directions
- → Group designs and RCTs — Compare FirstWork directly to traditional DTT across larger, more diverse samples to establish statistical generality.
- → Expanded skill domains — Evaluate efficacy for matching-to-sample, feature analysis, intraverbals, and foundational language programs beyond receptive identification.
- → Parent implementation — Digital programs can extend learning into the home, increasing daily instructional opportunities and accelerating generalization.
See the research in action.
Start a free trial and run your first session today, or book a demo to see how FirstWork fits your team's clinical approach.
No charge until day 8. Cancel anytime.
