The citation is on the card, not in a footnote.
Twenty-two studies across retrieval practice, spaced practice, elaboration, dual coding and metacognition. Every one of the 150 cards names which it runs on.
Every card names the research it rests on.
Turn any card over and the study is printed on the back, with the strategy it runs on named in plain words. Nothing here rests on a hunch about what works.
What students can retrieve from memory, without prompts or notes, is what they actually know. Everything else is still sitting in the classroom.
Retrieval
practice.
SPARK cards turn retrieval into a classroom event. The activity is the learning, rather than a check on learning that happened earlier. Every card pulls information back from memory, which strengthens the trace far more than re-reading or being told something again.
- Cards
- 75
- Range
- K-01 → K-75
- Phase
- Open / encode
Formative
reflection.
SHIFT cards ask students to look at their own understanding: what they know, how sure they are of it, and what they will do about the gaps.
- Cards
- 75
- Range
- S-01 → S-75
- Phase
- Close / consolidate
Seventeen traditions.
Four in depth.
Every card names the specific research behind it, from retrieval practice and interleaving to Chi & Wylie's ICAP framework, metacognition, learning by teaching, and formative assessment. 17 research traditions in all, each matched to what the card actually does. These four get the deep dive; the citation for every card appears on its back.
The testing effect.
Roediger & Karpicke, 2006. Cognitive psychology · Washington University in St. Louis.
Roediger and Karpicke demonstrated that retrieving information from memory is itself a learning event, not simply a measure of learning that already happened. Students who tested themselves on material remembered significantly more than students who re-read the same material for the same amount of time.
Peer learning &
tuakana-teina.
Mazur, 1997 on peer instruction · Nuthall, 2007 on the hidden lives of learners.
Eric Mazur showed that students learn deeper from peers wrestling with the same misconception than from a clean teacher explanation. Graham Nuthall's research found most of what students actually retain comes from peer-to-peer talk, not teacher-to-student talk. Tuakana-teina, the reciprocal relationship between a more-expert and less-expert learner, names what these cards enact: knowledge held by students, treated as worth defending and explaining.
Metacognition &
self-regulation.
Education Endowment Foundation · Cowie on assessment agency · Bjork on desirable difficulties.
The EEF rates metacognition as one of the highest-impact interventions available to teachers. Bronwen Cowie's work on assessment agency frames students as the primary agents in the feedback loop: generating, acting on, and evaluating feedback, not just receiving it. Robert Bjork's research on desirable difficulty explains why mistakes, properly framed, are not failure signals but learning conditions.
Formative assessment
as a loop.
Hattie · synthesis of 1,000+ studies · Timperley, 2014 · Spiral of Inquiry.
Hattie identifies feedback as one of the most powerful influences on student achievement, provided it is specific, actionable, and timely. Helen Timperley frames feedback as a loop: receive, act on it, then reflect on whether the action changed your understanding. Several SHIFT cards build this loop explicitly so students aren't waiting for a teacher to close it for them.
Nine researchers
we kept coming back to.
These are the people whose work most directly shaped what a card does and how to tell whether it worked. Citations appear on the back of every card where their work applies.
Seven tikanga concepts,
and where they run.
Every card carries a te reo Māori term alongside its title. The terms are not ornament. Each names something real about what the activity asks of students, often something the English title does not capture. These seven concepts are woven through specific cards in the deck. Listed with the cards they live in.
What a term
with the deck
looks like.
After about four weeks of consistent use, the time it takes to get a card going drops to almost nothing. Students know the format. They start to have preferences and request cards by name. They anticipate the write-before-you-speak prompt. They know what a debrief is for.
Four common
mistakes.
These are the patterns we watched experienced teachers fall into during testing. None of them are dealbreakers, but all four quietly cut the learning value of the cards by a lot.
Using cards as fillers
An unplanned card produces less retrieval benefit than a card chosen to retrieve something specific. The cards aren't time-killers. They are sharp instruments for a specific cognitive job. Choose them on purpose.
Only using Spark
Without SHIFT, you have limited information about what students actually understood. SPARK shows what students can retrieve. SHIFT shows what they know about what they retrieved. You need both halves.
Always choosing high energy
High-energy cards are engaging, but low-energy individual reflection is often where the deepest metacognitive work happens. A steady diet of high-energy cards can mask gaps that quieter reflection would surface.
Skipping the debrief
The debrief is where the learning gets consolidated. Without it, a card is just an activity. Even 60 seconds of debrief significantly increases the learning value of any card in the deck.
Read the deck
for yourself.
Every card names the study it rests on and the command word it practises.