Find in Picture Widget
Have learners tap every thing you ask for in a picture, like every rectangle or every animal, and grade whether they found them all.
Find in Picture asks learners to tap every thing you ask for in a picture: "Tap every rectangle in the playground", "Find all the animals on the farm", "Tap each thing that starts with B". It is graded, so you see whether each child found them all and what they tapped by mistake.
You add one picture and draw regions on it, one around each thing in the picture that matters. Each region is either:
- Find it: one of the things learners should tap, like the door, the window and the sandbox in a picture of rectangles.
- Not this one: a near-miss you expect children to tap, like a pair of scissors whose blades look almost like a rectangle. Tapping it counts as a wrong tap.
Taps on the picture outside every region count for nothing, so a learner who taps the sky isn't marked wrong. If regions overlap, a tap goes to the smallest region under the finger, so a window drawn inside a wall still works.
Find in Picture is the graded cousin of the hotspot. See Find in Picture or Hotspot? below.
Set it up
Select the widget on the canvas and click the floating ⚙ Configure button. The picture, the regions and every setting are there, with a live preview.
- Click Choose a picture and pick one from your media library.
- Draw a region around each thing that matters, using the tools above the picture:
- Rectangle and Oval: drag across the thing.
- Trace: click around its outline point by point. Click the first point or press Enter to finish, and press Escape to start over. Use it for odd shapes, like a tree or a puddle.
- Select: click a region to select it, drag it to move it, drag its corner handle to resize it, or drag the points of a traced region. Arrow keys nudge the selected region (Shift + arrow nudges further), and Delete removes it.
- Below the editor, each region has a row:
- A name, like "the clock". Screen readers read it when a learner reaches the region, and you see it on the student's attempt page. Name what the thing is, never whether it's one to find: "the scissors", not "wrong answer".
- A Find it / Not this one switch.
- A delete button.
Every region starts as Find it. Switch the near-misses to Not this one.
Settings
| Setting | What it does |
|---|---|
| Draw regions | Off (the default) keeps regions invisible over the picture. Outlined or Filled draws them as shapes. With it on, you don't need a picture at all, which suits exact shapes like a design made of pattern blocks. |
| Check | Right away (the default) marks each tap as it happens. At Done lets learners mark freely and checks the whole set when they tap Done (see below). |
| Correct when (Right away) | Every region found (the default), or Every region found with no wrong taps. Either way, wrong taps show on the student's attempt page. |
| Show "2 of 4 found" (Right away) | On by default. Shows a row of dots and the count under the picture, so learners know how many are left. |
| Marks (Right away) | Check marks or Circles on the things a learner found. |
| Speak region names | Says a region's name when a learner taps it ("the clock"), using the device's built-in voice. Good for pre-readers. |
How learners use it
Right away:
- Tapping a region to find marks it with a check mark (or a circle) and it stays found.
- Tapping a Not this one region makes it wiggle, shows a cross and the words "Not this one", then lets it go, so it doesn't stay selected. With reduced motion on, the cross appears without the wiggle.
- The learner is done when every region to find is found.
At Done:
- Each tap circles a region or uncircles it, with no feedback, like circling things on a paper worksheet.
- The Done button checks the whole set. It's right only when the circled regions are exactly the ones to find. A wrong Done says "Not yet. Look again." and the learner can change their circles and try again.
Use Right away for young learners who need feedback as they go. Use At Done when you want to see what a child believes without hinting along the way.
Worked example: "Tap every rectangle"
A playground picture with 4 rectangles (the door of the shed, a window, the sandbox and a bench seat) and 2 near-misses (a pair of scissors on the bench and a triangle-shaped sign).
- Add a text widget that says Tap every rectangle. and, if you like, a text to speech widget that reads it aloud.
- Add Find in Picture from the toolbox and make it big: most of the scene. Small regions are hard for young fingers.
- Click ⚙ Configure, then Choose a picture, and pick the playground picture.
- With Rectangle, drag a region over the door, the window, the sandbox and the bench seat. Name them "the door", "the window", "the sandbox" and "the bench".
- With Trace, click around the scissors and the sign. Name them "the scissors" and "the sign", and switch both to Not this one.
- Leave Check on Right away and Show "2 of 4 found" on.
- Set the scene to move on when correct, or add a submit button to grade it like any other question.
A child who taps the scissors sees them wiggle and "Not this one", then keeps looking. On the student's attempt page you see one line: Found 4 of 4 · 1 wrong tap (Scissors). With At Done it reads like Marked 3 of 4 · also marked Scissors.
To give a hint, add a behavior on On Wrong Tap with a Display Message that says "Count the sides. A rectangle has 4."
Grading
A new Find in Picture starts graded. Points are all or nothing. The taps are sent to Berdee, and the grade is worked out there from the lesson, so it can't be faked from the learner's device.
The Attempts column on the attempt page counts:
- Right away: one attempt per tap on a region, found or not. A perfect run is the number of regions to find (4 in the example), and each wrong tap adds one. Taps outside every region, and tapping a region that's already found, count nothing.
- At Done: one attempt per Done tap. A perfect run is 1.
What blocks publishing
The lesson can't be published until you fix:
- No region marked Find it.
- No picture, with Draw regions set to Off (learners would see nothing to tap).
- A region with no area, like a trace with every point in a line.
- A region that reaches past the edge of the picture, where no one can tap it.
- More than 40 regions.
The studio also warns, without blocking, about a region smaller than 44 pixels at the widget's size (too small for young fingers: make the widget bigger or the region larger), a region with no name, and a region mostly covered by a smaller region (taps there go to the smaller one). A tap that just misses a small region still counts for it: a tap a fingertip's width outside every region goes to the nearest one.
Behaviors and logic
Find in Picture reports these for conditions and variables:
- Found: how many regions to find the learner has found (or circled, at Done)
- To find: how many regions are marked Find it
- Wrong taps: how many times the learner tapped a Not this one region
- All found
- Is correct and Is answered
Its events depend on Check:
- Right away: On Found runs when a learner finds a region, On Wrong Tap when they tap a Not this one region, and On All Found when the last one is found.
- At Done: On Done runs on every Done tap, and On All Found when a Done tap is right.
Its actions are Reset, which clears every mark, and Reveal the answers, which marks every region to find. Put Reveal the answers on a "Show me" button to demonstrate before learners try. See Events, Actions, and Conditions.
Accessibility
- Every region is a button named by its region name. A region with no name is read by position, like "Thing 3 of 9", so name them all.
- Regions are in reading order: top to bottom, then left to right.
- Tab reaches the picture, arrow keys move between regions with a visible outline, and Enter or Space taps the region. Tab again moves on to Done.
- Screen readers hear what happened, for example "Clock. Found. 2 of 4 found." or "Scissors. Not this one."
- Color is never the only cue: found regions get a check mark or circle, wrong taps a cross.
- Animations respect the device's reduced-motion setting.
Find in Picture or Hotspot?
- Use Find in Picture when there's a right answer: learners should find particular things, and you want to know whether they did. One widget holds the picture, every region and the answer key.
- Use a **hotspot** when learners explore: tap a part of the heart to hear what it does, tap a city to see its name. A hotspot has no wrong answer and isn't a graded question.
Not in this version
- Two kinds of mark in one question, like "circle the rectangles and cross out the triangles". Make it two questions, one per scene.
- A magnifier for phones. Make the widget big; the studio warns you about regions that are too small to tap, and a tap just outside a small region still counts for it.
- Counting what was found as the answer. To ask "How many rectangles did you find?", follow it with a number input, which can check its answer against Found.