
the estimate.Some children answered without estimating. I made moving the marker part of the task.
Cause and lasting learning unproven.
Explore the woodland or school.
Tools for gathering and making sense of evidence.
Build the test around the experience. Follow a fictional task through to the evidence available for review.
Stage 1 of 4: The task.
Illustrative explanation. Fictional tasks, screens and responses. No research data is collected or sent. These are two separate examples, not a shared implementation.
Try a fictional field note. Keep what happened separate from what you think it means.
Loading the logger…
These are the current demo record’s values. Empty fields stay empty.
Poke, pull and discover what changes. An experiment in touch and cause and effect.
Loading the board…

The control and the object it changes light up together. Try the weather buttons and notice how the penguin reacts.
The board brings pressing, pulling, sliding and turning into one place. The question is whether children notice the connections and explore them together with an adult.
A working prototype intended for a young child and an adult to explore together. It hasn’t been tested with children yet.
Find four letters, build a spell and wake the sleeping pen.
Loading SpellQuest…
A word met in the forest becomes part of the adventure.
Spelling difficulty and willingness to keep trying were both part of the research. I planned the learning, story and emotional journey together.
The forest introduces words. The workshop asks children to construct and try a spelling. Waking the pen gives the task a place in the adventure.
Design intentions. The prototype still needs testing with children.

Pip uses curious in conversation. The word arrives with a meaning.
Replay the conversation or listen to the word on its own.
Return to the forest. The encounter can be revisited.
Five research interviews informed the early direction. The research highlighted the need for context and repeatable audio.
Characters introduce words through short conversations, with written text and speech. Children can listen again, hear the word on its own, or leave and return. The intention is to make help available without making every encounter a test.
These are design intentions. The prototype still needs testing with children.
Confidence and the response to mistakes mattered in the research. At the bench, children can move, swap and return letter blocks before trying their spell.
The interaction developed from selecting the next letter into rearranging the blocks directly. I kept judgement out of construction so children could inspect and revise their attempt.
Still to test: do children understand how to revise without an adult stepping in?

Tracing follows the constructed letters. The movement is forgiving, without an accuracy score, and children release between letters. Keyboard input offers another way through.
After a successful attempt, a question mark becomes an ink cat before returning to the pen. The reward connects to the meaning of curious.
Tracing and narrative are design choices here. Their contribution to spelling retention still needs testing.

Earlier work explored a parent-led spelling app. I removed teacher-assigned lists from that direction because depending on updated school lists was a parent pain point.
I also explored reducing parent setup with the Eliminate, Automate, Simplify framework. Those onboarding flows and the wider multi-stage learning plan are earlier explorations, outside this playable level.
The current prototype opens directly, without this account flow. The earlier idea included a broader age range; the current build is aimed at Years 5–6.
Research and learning design by James Spiller. Prototyped with AI coding assistance. These are design intentions and prototype decisions, not measured learning outcomes.
Loading the app…


Research and learning design by James Spiller. Prototyped with AI coding assistance. Images show the actual apps.