A session log can tell you a child answered 12 of 16 trials correctly. It cannot tell you that the child spent a third of the session looking at the ceiling — and that the adult sitting alongside spent the session gently steering their gaze back. That difference matters clinically: attention is the first prerequisite of learning.
Interlaza’s new Attention Monitor closes that gap, and it does so with a privacy stance we consider non-negotiable for an app used by toddlers.
What it does
When you switch it on for a student, the tablet’s front camera checks — several times per second, during sessions only — one simple thing: is the child’s face oriented toward the screen? From that single signal, three features follow:
Gentle auto-pause. When attention drifts away and stays away for several seconds, the session pauses on a calm break screen instead of burning trials nobody is attending to. The moment the child looks back, the session resumes by itself. (A quick glance at you doesn’t count — the monitor waits for genuinely sustained disengagement.)
Trials that mean something. Trials no longer run while the child is looking elsewhere, so accuracy reflects ability, not attendance.
An objective engagement measure. After each session, instructors see what percentage of the session the child was engaged, how many attention breaks happened, and how engagement evolved across the session — the “before and after” picture that’s so hard to judge from memory. The signals we use — face orientation toward the screen — are the same ones validated by researchers measuring attention in autistic toddlers on tablets (Scientific Reports, 2023).
The privacy part, in plain words
Everything happens on the tablet. The camera frames are analyzed by a small model running on the device itself and are discarded instantly — they are never recorded, never stored, never uploaded. There is no server that could see them, because nothing is sent. What remains after a session is a handful of numbers, such as “engaged 84% of the session, 2 attention breaks” — and even those numbers stay on your device and are excluded from cloud backup entirely.
We built it this way because anything less would be unacceptable for children this young. If you never enable the feature, the camera is never touched at all.
How to turn it on
The Attention Monitor is opt-in, per student, from the student’s profile (Attention Monitor section). Switching it on is itself the consent step — the setting explains exactly what is processed and what is kept, in one paragraph, before you decide. If a device has no camera or permission is declined, sessions simply run as they always have.
One more thing it quietly improves: the Program Advisor reads the same engagement picture when suggesting whether a child would benefit from a period of real-object practice away from the screen. The two features were designed together — noticing the problem, and knowing what to do about it.