015 min
Notice a measurement
Light a paper object and shade it. Describe what changes without moving the object.
Ask: “Did the object leave when it got darker?”
Listen for: “No, the light changed.”
Brightlab
brightlab.Made for curious mindsbrightlab.Made for curious mindsBrightlab
Year 1 · Sensors
The toy is still there. Will the detector spot it when the light gets dim?
Go to the investigation ↓A detector setting with three annotated signal tests.
The toy is still there. Will the detector spot it when the light gets dim?
A light sensor measures a signal. Our detector compares that signal with a threshold. The signal depends on the object and its lighting. If the light changes, the same object may fall below the threshold. A sensor gives a measurement, not a complete account of the world.
Before this lesson: Explain the difference between an object and a picture of it. Useful earlier investigations: The playground planner.
A light sensor measures a signal. Our detector compares that signal with a threshold. The signal depends on the object and its lighting. If the light changes, the same object may fall below the threshold. A sensor gives a measurement, not a complete account of the world.
Explain the difference between an object and a picture of it. Useful earlier investigations: The playground planner.
Use a torch and a paper object to demonstrate light without pointing at eyes. Open the synthetic signal display and identify the fixed object pulse and background readings.
ACARA V9 AI curriculum connection ↗ · Technologies ↗
A complete teaching sequence · 40 minutes
Light a paper object and shade it. Describe what changes without moving the object.
Ask: “Did the object leave when it got darker?”
Listen for: “No, the light changed.”
Children mark whether the detector will still see the object at half brightness.
Ask: “What does the detector compare?”
Listen for: “The signal with the line.”
Move the light control in steps; observe the waveform and detected region. Keep the threshold fixed and record two readings.
Ask: “Which number changed while the object stayed?”
Listen for: “The measured light signal.”
Lower the threshold until the object reappears. Reveal a background spike.
Ask: “Can lowering the line cause a different mistake?”
Listen for: “It might see the background as an object.”
Choose a compromise threshold and test bright, shadowed and empty-background cases. Explain unresolved errors.
Ask: “Can your setting pass all three tests?”
Listen for: “Some signals overlap, so maybe not.”
Draw signal and threshold for one miss and one false alarm.
Ask: “How do we know this was a miss?”
Listen for: “The object was there but its signal was too low.”
Sensors see the whole world.
Lowering the threshold recovers a shaded object but can also detect a background spike that is not the object.
Choose a threshold and document one correct detection, one miss and one false alarm across lighting conditions.
Look for an explanation involving measurement, not an object disappearing. Use the background spike to check whether the learner recognises two kinds of error.
| Criterion | Beginning | Secure | Extending |
|---|---|---|---|
| Fair testing | Moves object and light together | Changes light with object fixed | Records comparable signal values |
| Error explanation | Says the sensor is bad | Connects a miss to the threshold | Explains the miss/false-alarm trade-off |
Use a large printed signal with a movable string threshold. Students can point instead of reading decimal values.
Compare two sensors with independent background spikes and propose an agreement rule.
No cameras or microphones are used. Never shine a torch at eyes; the lab is a synthetic demonstration, not a safety detector.
Shared device? Turn remembering off. A project file lets you continue on another device.
Your browser is the laboratory
The toy is still there. Will the detector spot it when the light gets dim?
The same object occupies readings 9–16. Its bright-light signal is roughly 0.7; background is roughly 0.1. The detector threshold is 0.5. Halving light halves those signals.
The experiment opens after your prediction.
There is no penalty for being surprised.
Apple Silicon · PyTorch MPS
Simulate many independently noisy readings at each illumination level and estimate detection and false-alarm rates with sample counts.
Teacher-operated extension: students predict and interpret the visual report together.
The bundle contains lesson-specific working code, a configuration file, a reactive notebook, a deterministic CPU check and hardware setup instructions. Acceleration is reported only after a tensor operation and result read-back succeed.
Download Mac Studio investigation ↓unzip y1-sensor-mac-pathway.zip -d y1-sensor-mac cd y1-sensor-mac bash setup-mac.sh source .venv/bin/activate python experiment.py --device mps --output results marimo edit notebook.py
CPU and available-device execution status is recorded in the downloaded README and validation report. DGX Spark execution requires that hardware; static validation alone does not establish GPU compatibility or performance. The browser lesson remains fully available without this extension.
NVIDIA DGX Spark · PyTorch CUDA
Simulate many independently noisy readings at each illumination level and estimate detection and false-alarm rates with sample counts.
Teacher-operated extension: students predict and interpret the visual report together.
The bundle contains lesson-specific working code, a configuration file, a reactive notebook, a deterministic CPU check and hardware setup instructions. Acceleration is reported only after a tensor operation and result read-back succeed.
Download DGX Spark investigation ↓unzip y1-sensor-dgx-pathway.zip -d y1-sensor-dgx cd y1-sensor-dgx bash run-dgx.sh # Open the localhost notebook URL printed by the container.
CPU and available-device execution status is recorded in the downloaded README and validation report. DGX Spark execution requires that hardware; static validation alone does not establish GPU compatibility or performance. The browser lesson remains fully available without this extension.
A detector setting with three annotated signal tests.
Download the editable handout →