Adaptive Learning Environments 2011/12 (UG4/MSc)

1. Adaptive Learning Environments 2011/12(UG4/MSc) Helen Pain, helen@inf.ed.ac.uk Adaptive Learning Environments

2. ALE Lectures 1 and 2:Introduction Adaptive Learning Environments

3. What does adaptation mean? • Personalisation to the learner • In the specific context (learning) • In relation to their (or others) goals Adaptive Learning Environments

4. What might we adapt to? • Age • Special Needs • Physical environment • Knowledge and background of student • Motivation and affective state • Domain or curriculum area Adaptive Learning Environments

5. What do we adapt? • How the curriculum is presented, represented, structured • The teaching approach and strategies that we use • The physical environment and the interface • The communication and form of interaction Adaptive Learning Environments

6. 1. An on-line elearning example Adaptive Learning Environments

7. Example from CBBC Schools • Alien Cookbook - teaches numeracy (5 to 7 year olds) • http://www.bbc.co.uk/schools/starship/maths/aliencookbook.shtml • In what ways does it adapt to the learner? • What problems do you see with the design of this software? Adaptive Learning Environments

8. 2. Example 1: modelling subtraction Adaptive Learning Environments

9. Example problem: subtraction • a. 73 b. 32 c. 164 d. 187 e. 19763 • -11-16- 37- 99-16824 • How do you do each of them? • What methods do you use? • Do you use the same method for all of them? • How did you learn to do it? • How would you teach someone else to do it? • What would they need to know to do so? • What would you need to know to teach them? Adaptive Learning Environments

10. What if their answers were: • a. 73 b. 32 c. 164 d. 187 • -11-16- 37- 99 • S1 62 24 133 112 • S2 62 26 137 198 • S3 62 24 214 817 • S4 61 14 130 89 • or no response at all.... Adaptive Learning Environments

11. Modelling learners • trying to find out what the student knows, believes, can do • looking for evidence that user fails to exploit some knowledge • looking for inconsistent beliefs, differences between student and domain models • teach accordingly Adaptive Learning Environments

12. Diagnosing Student Models • If the teacher believes student has a different model from their own (correct) one: • make list of common errrors and match to it • reason about what student would believe in order to exhibit behaviour indicating this • Representation of student's current state of knowledge = STUDENT MODEL • Inferring the Student Model = DIAGNOSIS Adaptive Learning Environments

13. Possible Diagnoses • a. 73 b. 32 c. 164 d. 187 • -11-16- 37- 99 • S1 62 24 133 112 • take higher from lower • S2 62 26 137 198 • give 10 but don't pay back • S3 62 24 214 817 • work l to r, higher from lower • S4 61 14 130 89 • guess… Adaptive Learning Environments

14. Design an ALE that can: • Do subtraction for itself • Know how to teach subtraction • Determine what the student knows already • Adapt its teaching to the learners ability • Infer how the student learns as they do it • Diagnose errors • Know how and when to talk to student • Give appropriate feedback • Suggest other exercises Adaptive Learning Environments

15. Adapting to errors and misconceptions • What are we adapting to? • Hypothesised misconceptions of students, inferred from their behaviour and knowledge of common problems • Knowledge of the students ability in arithmetic and literacy level • How do we adapt? • Customise feedback to: • the specific errors a child makes • literacy level • Select further problems to fit students ability Adaptive Learning Environments

16. 3. Example 2: modelling spelling Adaptive Learning Environments

17. 1. Specific difficulties in spelling • A child types • e.g. “neiz” • “wen” • What did they intend? Adaptive Learning Environments

18. Identifying and Correcting Errors • e.g. “neiz” -> knees/niece • “wen” -> when/went/we/win • “fiknusiz” -> Adaptive Learning Environments

19. Identifying and Correcting Errors • e.g. “neiz” -> knees/niece • “wen” -> when/went/we/win • “fiknusiz” -> thicknesses Adaptive Learning Environments

20. Possible Inferences • Work back from the misspelling to the correction • k w i c quick • r i d red, rid, ride, write, rite, read • f a t fate • Letter as its name?i for i_ea for a_e Adaptive Learning Environments

21. Phoneme-Grapheme Correspondences Phoncode (Pain, 1985) • Based on phoneme-grapheme grammar: consider what phonemes error was intended to represent, then see if any word matches • ‘neiz’ • ‘n’ ‘ei’ ‘z’ /i/ kEEp /IE/ EAr /aI/ hIGH /I/ bIt /eE/ Air /eI/ dAY /z/ Zoo /zh/ pleaSure /s/ Sit /sh/ buSH /n/ No /ng/ siNG /m/ jaM knees /n/ /i/ /z/ niece /n/ /i/ /s/ Adaptive Learning Environments

22. 4. Some other examples … Adaptive Learning Environments

23. Computer games in education • Children find computer games motivating and spend a lot of time playing them * • Teachers and parents recognise that children develop collaboration, thinking and discussion skills through computer games * • BUT – there is a mismatch between curriculum and commercial game content which prevents games from being used in the classroom • Purpose made educational games can be developed to fit the curriculum • Pupils can learn from making their own games • * McFarlane, A., Sparrowhawk, A. and Heald Y. (2002) Adaptive Learning Environments

24. Adventure Author: storymaking through computer game design • Cathrin Howells, Creative Contexts • Judy Robertson, Heriot Watt University • “Creating a 3D interactive immersive game provides new opportunities for authors. • We see conventional writing at work, but also gain insights into ways pupils use the visual, spatial and interactive elements of game design to contribute to the storymaking process, carrying meaning beyond words.” (Judy Robertson) • ‘Adventure Author’is a freely available game creation tool for children aged 10-14, a modification of the Neverwinter Nights 2 game-making toolset for designing and building interactive stories. • (slides edited from those of Judy Robertson)

25. Adventure Author…www.adventureauthor.org Aims to study the creative process learners go through when making their own computer games Team has worked with schools in Edinburgh and Dundee, and holiday workshops, to design and refine game making software Children must consider: How do I build a story spatially? Under what circumstances should I try to tell the story through words and when should I use visual design? How can I anticipate what my reader might do? What makes a good writer-designer???

26. Neverwinter Nights 2

27. Game making toolset Toolset comes free with Never winter Nights 2 game

28. Dialogue as it appears in the game world

29. Conversation Writer: dialogue

30. Dialogue: choices

31. Crystal Island: Outbreak • James Lester, intelliMEDIA group, • North Carolina State University • http://www.intellimedia.ncsu.edu/ci8.html • Crystal Island is an intelligent game-based learning environment • Focus of middle science education (Grade 8) emphasizing microbiology • Students interacting with the Outbreak episode of Crystal Island solve a science mystery that unfolds on a recently discovered island. • The project focuses on narrative-centered learning and modeling students’ problem-solving activities. Adaptive Learning Environments

32. Video clips • Crystal Island overview: • http://www.youtube.com/watch?v=aduzGkj8J2k&feature=mfu_in_order&list=UL • Testing an egg • http://www.youtube.com/watch?v=r1waTnT4Y5s&feature=mfu_in_order&list=UL • I think I ate something bad…. • http://www.youtube.com/watch?v=giwBQqSG-yc&feature=mfu_in_order&list=UL Adaptive Learning Environments

33. Alelo: Foreign language and culture training • Alelo's Virtual Cultural Awareness Trainer (VCAT) • http://www.youtube.com/watch?v=hZ2CLv6JyXo&feature=BF&list=ULLR-c8JEL1J0&index=5 • Alelo FITE-Pashto demonstration • http://www.youtube.com/watch?v=5_VSzC6jbz4&feature=related • Alelo's Virtual Role Players (VRP) for Bohemia's VBS2 Mission Rehearsal software • http://www.youtube.com/watch?v=JjZd34_RF0g&feature=mfu_in_order&list=UL • Scenario demonstration from Alelo's Mission to France (pilot) language and culture training program • http://www.youtube.com/watch?v=_BDR1y2Dc2Q&feature=BF&list=ULLR-c8JEL1J0&index=6 Adaptive Learning Environments

34. Vicarious Learning and Case-Based Teaching of Clinical Reasoning Skillshttp://www.tlrp.org/proj/phase111/cox.htm • “Vicarious Learning” (VL) is learning from the learning experiences of others: • one student is the focus of tutorial attention • others present also benefit from observing the interaction. • Capture and re-use of learning dialogues • Evaluation of how, when and why VL works • Collaborative project with Sussex, Sheffield and Newcastle Adaptive Learning Environments

35. “PATSy” case-based clinical training system • “PATSy” helps students (speech, language therapy, medicine and psychology) practice diagnostic skills through audio clips, video clips and interactive tests, which they can 'administer' to virtual patients. • Initial focus on speech pathology - clips of tutors and students dialogues on difficulties experienced • Future learners will be able to select, retrieve and use the clips for vicarious learning support when they encounter similar difficulties. • Planned extensions to general teacher education and creative practices Adaptive Learning Environments

37. 5. Supporting Language Play in Childrenhttp://www.csd.abdn.ac.uk/research/standup/ Standup: (based on JAPE, Binsted & Ritchie 1994, 1997): Facilitating language play in non-speaking children through computer-supported joke construction* - Built support tool that enables children with complex communication needs (CCN) to experience language play (through humour) - Current AAC tools are primarily functional, focusing on needs-based communication Interactivity and customizability User-centred design Brings together a number of aspects of AI, Cognitive Science, Human Computer Interaction, Education *EPSRC Grants GR/S15402/0 and GR/R83217/01 Adaptive Learning Environments

38. Play through humour • What do you get when you cross a monkey and a peach? • An ape-ricot. • What do you call a murderer with fibre? • A cereal killer. • What kind of vegetable can jump? • A spring onion. • What do you get when you cross cars and sandwiches? • Traffic Jam • How does a whale cry? • Blubber blubber. • How is a car like an elephant? • They both have trunks. • Based onJAPE (Binsted & Ritchie 1994, 1997) and developed in Standup project Adaptive Learning Environments

39. Exercise: Consider what knowledge is needed to understand such jokes - what different types of knowledge do you have to have? What might the educational function of playing with such jokes be? Can we automate this - what would we need to know and represent to do this? What would be the purpose of this? Adaptive Learning Environments

40. Would need a word list = Lexicon • Part-of-speech (POS) tags • Phonetic spelling, for computing: • homophonestime thyme • rhyme pub tub • spoonerismbare/spank spare/bank • Compound nouns and their components • e.g. long time, traffic jam • Distinct senses of a word/phrase, • e.g. match=sporting event, match=ignition stick • Semantic relations: • synonymsstrange bizarre • hypernymsthyme herb • meronymstraffic car Adaptive Learning Environments

41. e.g. Augmenting Wordnet Starting point: WordNet (200k senses, synonyms, hypernym hierarchy, meronyms) Phonetic forms: Unisyn: pronunciation dictionary, phonetic strings assigned to >115k word forms (Edinburgh accent) Pictures: Widgit “conceptcodes” already linked to two picture libraries; linked to WordNet senses, manually. Familiarity ratings: data from spelling lists, SemCor, Widgit conceptcode set, MRC psycholinguistic database, BNC Topics: adopted hierarchy supplied with Widgit coding. Excluded items: anything in Shorter Oxford “coarse slang” or “racially offensive”, plus a few from personal knowledge. (Thanks to Widgit Software and Mayer-Johnson for pictures) Adaptive Learning Environments

42. How could we automate this?JAPE: example of structure synonym What do you call a strange market ? describes It searches a general purpose dictionary to find words that fit pre-defined structures called schemas and templates. A bizarre bazaar. homophone synonym Adaptive Learning Environments

43. Purpose: need for language play • Word play is critical part of language development • typically-developing (TD) children enjoy jokes and riddles • provide opportunity to practise language, conversation and social interaction skills. • Jokes • are a type of conversational narrative • play an important role in the development of storytelling skills. • Role of punning riddles in language development • pragmatics => turn taking, initiation etc. • vocabulary acquisition • Children with speech and/or language disabilities do not always have language play opportunities • Pre-stored rather than novel jokes • little opportunity for independent vocabulary acquisition and word play Adaptive Learning Environments

44. Standup goals • To build a tool that helps children with complex communication needs (CCN) to play with language: • generate novel puns using familiar vocabulary, • experiment with different forms of jokes. • provide social interaction possibilities • go beyond the “needs” and “wants” of AAC* • *AAC: augmentative or alternative ways to communicate for people with limited or no speech Adaptive Learning Environments

45. Adaptive Learning Environments

46. References • see webpage Adaptive Learning Environments