Introduction
What is this about?
If robots and AI-controlled devices are to act meaningfully in everyday life, how do they know what to do? Are they told what to do all the time? Do they always carry out the instructions without thinking independently? Instead, are machines also able to learn, and if so, how does this learning take place? Can a robot use what it has learned to react flexibly to situations? The educational professionals support the children in their thinking about the question of how robots and AI-controlled devices learn and how cleverly the machines can interact?
Children‘s point of view
Questions from children
What we know
Linguistic dimension
Firstly, it should be clarified with the children what is meant by calling somebody or even something smart oder clever.
Is somebody clever who knows al lot? Does smart mean being good at cheating? Do they kow the colloquial sayings: “clever as a crow” or “clever as a pig”? The point is to understand what children mean by smart and how they would describe a smart robot
Mathematical scientific level
Data collection: counting, ordering, representing/displaying
Combinatorics: sorting, reassembling, pattern recognition
Machine learning, deep learning
Algorithm + abstraction, whereby humans cannot interpret individual „learning steps“
Machine Learning
Goals
Pedagogical professionals
Review and expansion of knowledge base.
Design work spaces to promote systematic thinking.
Explore stuctures and patterns in nature and arts
Foster problemsolving referring to their own mental images.
Children
Distinguish between clever, smart and intelligent
Recognise the limits of a robot’s cleverness
Identifyng structures and delineate pattern recognition
Create stories in which someone acts smart
Reflect on critical contents of image-based media
Exercises
#5 Matching colours & shapes
Materials
- Tablet + APP Preschool
Preparation
Charge the tablet and download the apps in advance. Read the description of the app and think about how to introduce this game. The children should play individually against the app.
Implementation
The app will help the child to sort all kinds of geometric shapes and colours, such as vegetable or fruit. If the wrong vegetable is selected, a sound is given and the piece goes back to the garden. As soon as the child matches the vegetables according to the picture depicted on the basket, the next basket appears and the game continues.
Reflection
Why does the robot (app) know what is right or wrong?
#5 Recognising patterns, deriving rules
Materials
- Sample Pattern Pieces
Preparation
Create different workstations.
There are task cards at each workstation.
On the left hand side, there is the target picture.
On the right hand side, the individual pattern.
pieces needed for the task.
Implementation
Lay out the sample cards.
Clarify with the children how the picture is constructed.
Let the children assemble a picture from the pattern masks.
The correct pattern only emerges when all the cards have been placed on top of each other accordingly.
Reflection
Discuss how knowledge develops from individual experience. Human teaching develops from trial and error or logical thinking, combining different pieces of knowledge and reasoning.
#5 Face Recognition
Materials
- Take phots of faces from a magazine or newspaper
Preparation
Cut photos into 3 stripes: forehead + eyes, nose, mouth + chin.
Cut photos into 5 stripes: forehead, eyes, nose, mouth, and chin.
Implementation
Present the mixed-up stripes of faces to all children in a museum walkway.
Ask them why the compilation fit or is not appropriate.
Let the children hypothesise and think about their suggestions together.
Reflection
Can a robot, an AI recognise a face and parts of a face, for example the eyes?
How does a robot/an AI do that?
What does the robot need to recognise this?
Try out a mobile phone/tablet with face recognition
Can any face unlock the phone or only the owners?
Which robots/AIs have a face recognition sensor?
About this Toolbox
Toolbox #5 was created in 2022 by Susanne Schumacher, Ulrike Stadler-Altmann, Brigit Brunner, Katrin Crazzolara, Michael Schlauch, Christian Laner, Birgit Pardatscher
The project "I'm not a Robot: working with artificial intelligence in early childhood education" is co-funded by the Erasmus+ program of the European Union. The European Commission´s support for the production of this publication does not constitute an endorsement of the contents which reflect views only of the authors. The commission cannot be held responsible for any use which may be made of the information contained therein.
This work is licensed under: Attribution-NonCommercial-ShareAlike 4.0 International