Mi Board Metacognitive Training Through Gaming in i

Mi. Board: Metacognitive Training Through Gaming in i. START Justin F. Brunelle, Irwin B. Levinstein, and Chutima Boonthum, Member, IEEE

Agenda Introduction The i. START Project Overview of Mi. Board ◦ Origin ◦ Rules Game play of Mi. Board Technical Aspects of Mi. Board Future Work 4/9/2009 2

Introduction - Names Justin Brunelle ◦ ODU Computer Science Graduate Student ◦ jbrunelle@cs. odu. edu Interactive Strategy Trainer for Active Reading and Thinking (i. START) Multiplayer Interactive Board Game (Mi. Board) 4/9/2009 3

Introduction - i. START Web Based Trains high school students to become better readers ◦ Science texts Animated Agents Instruction provided in Modules: ◦ Guide and instruct students in each module ◦ Genie, Merlin ◦ Introduction ◦ Demonstration ◦ Practice Will Utilize Mi. Board 4/9/2009 4

Introduction - Practice and Extended Practice Problem with Extended Practice ◦ Repetitive and tedious ◦ Leads to disengagement and lack of interest 4/9/2009 5

Introduction - Mi. Board Extension of i. START Practice Goals: ◦ Provide opportunity to practice ◦ Increase engagement ◦ Provide collaborative/social learning environment Mi. Board Experiment Begins Fall 2009 4/9/2009 6

i. START Funded by National Science Foundation Online Tutoring system Focuses on science texts ◦ Complex! Teaches active reading ◦ Metacognitive strategies When used, increase comprehension of a text 4/9/2009 7

i. START - Strategies Metacognitive Strategies ◦ Comprehension monitoring being aware of how well one understands what one is reading ◦ Paraphrasing restate the sentence content in their own words ◦ Prediction predicting what will come next in the text ◦ Elaboration linking information in the sentence to information you already know ◦ Bridging linking different parts of a text together 4/9/2009 8

i. START – Self Explanation (SE) Students use strategies to create self explanations Creating self explanations makes students ◦ More successful at solving problems ◦ More likely to generate inferences ◦ Able to construct more coherent mental models ◦ Develop a deeper understanding of the concepts covered in the text [2], [3] 4/9/2009 9

Example Self Explanation Text: ◦ The smallest possible particle of an element is called an atom. Atoms are more than a trillion times smaller than a human cell. Self Explanation: ◦ This text says atoms are really small. I learned in science that everything is made of atoms. I don’t understand what that has to do with elements. I bet this text will talk about molecules. 4/9/2009 10

i. START - Modules Instruction occurs in stages ◦ Introduction ◦ Demonstration ◦ Practice & Extended Practice 4/9/2009 11

i. START – Introduction Module Three animated agents ◦ Provide instruction on each strategy ◦ Provide information ◦ Pose questions ◦ Provide explanations of SEs and the reading strategies Includes for each strategy: ◦ Example text and SEs ◦ Quiz 4/9/2009 12

i. START – Demonstration module Two Agents ◦ One instructor, one trainee ◦ Interact to show to compose SEs (And consequently, use the software. ) Agents read the text, then make the self explanation Student selects which strategy the trainee agent used ◦ Feedback given by the instructor agent Mimics the practice module 4/9/2009 13

i. START – Practice Module Instructor Agent ◦ Reads text Student creates own self explanation Instructor provides feedback Student may be required to modify his SE Student identifies the strategy he used Leads to Extended Practice ◦ Same as practice, but over time to allow practice creating SEs 4/9/2009 14

i. START - Findings Students who have used i. START have increased comprehension over those who do not i. START online training is as effective as live training Readers of all skill levels benefit ◦ Skilled readers make more connections within a text (bridging) ◦ Less skilled readers learn more basic skills (paraphrasing) [5], [6], [7], [8], [9], [10] 4/9/2009 15

i. START – Extended Practice Research indicates need for extended practice ◦ Effects of i. START taper over time ◦ Less skilled readers need additional practice Current extended practice is repetitive ◦ Leads to disengagement and lack of interest among students [11] 4/9/2009 16

Mi. Board Provide an engaging method of Extended Practice ◦ Meant as a supplement to Extended Practice, not replacement Accommodate all students Allow further practice of i. START strategies Based on a physical game system 4/9/2009 17

i. START: The Board Game Developed by Dr. Mike Rowe Implemented i. START Practice in a game environment ◦ Goal is to move around a board Roll dice and draw cards (like Monopoly) ◦ A student creates a SE from a text included with the game ◦ Other students guess what strategy was used ◦ A discussion is initiated to resolve disagreement Refer to Appendix A for the rules of Rowe’s game 4/9/2009 18

i. START: The Board Game - Findings Effective form of Extended Practice Increased engagement; more enjoyable Rowe’s dissertation mentions a digital version would further increase engagement [13] 4/9/2009 19

Mi. Board Virtual version of i. START: The Board Game Random text chosen from a database Text revealed gradually over the course of the game Varying the texts aims to increase engagement by increasing variety 4/9/2009 20

Mi. Board - Rules were adapted from Rowe’s game ◦ Digitalization allows for simplification New features added to ensure educational environment ◦ Limiting availability of the chat Rules are outlined briefly over the next three slides 4/9/2009 21

Mi. Board – Game Rules Everyone will take turns being a reader (starting with Player 1) Being the READER: ◦ Given text with a target sentence in BOLD ◦ Given a strategy and a point value ◦ Provide a SE of the target sentence using the given strategy After the READER finishes his/her selfexplanation all players are directed to the GUESSER screen All players read the SE created by the READER and indicate the strategy used 4/9/2009 22

Mi. Board – Game Rules Once all players have entered his/her GUESS, the SUMMARY screen will present the points scored for all players The READER then ◦ Rolls the dice to moves his/her token ◦ Then draws an event card The next player will then take his/her turn being the READER 4/9/2009 23

Mi. Board – Game Play A list and description of the game screens can be found in Appendix B Mi. Board can be played here: Mi. Board Login 4/9/2009 24

Mi. Board - Experimentation Testing begins Fall 2009 ◦ Used in three high schools using i. START in Memphis, TN Control group will use i. START with regular Extended Practice Test group will use Mi. Board in addition to regular Extended Practice 4/9/2009 25

Mi. Board – Technical Innovations Developed with: ◦ Flash programming language Action. Script 3. 0 ◦ Java. Script ◦ Java Server Pages (JSP) ◦ My. SQL ◦ Electro. Server 4/9/2009 26

Mi. Board - Electroserver Multiplayer server product ◦ Specializes in multiplayer Flash games ◦ Has own set of Action. Script abstract data types Provides infrastructure for Chatting Public messages ◦ Sent to all connected clients within a room or zone 4/9/2009 27

Mi. Board - Electroserver Room and Zone ADTs ◦ Zones contain rooms ◦ Rooms contain a number of players participating in a single game A Mi. Board game is contained in a room of 3 to 4 players 4/9/2009 28

Mi. Board – Control Passing Participating clients have round-robin master-slave relationship ◦ Each client contains code to run the entire game When a client is a reader ◦ Client controls the game by passing public messages to each client in the room When a client finishes its turn, it relinquishes control to the next player 4/9/2009 29

Mi. Board – Game-Chat Relationship Mi. Board Game Movie: ◦ User interacts with the movie ◦ Relevant messages are sent to the Chat Movie ◦ Receives message from the chat and reflects the message content in the Game Movie Chat Movie ◦ Broadcasts the message as a Public Message to all other Chat Movies, or… ◦ Receives a message and sends it to the Game Movie 4/9/2009 30

Mi. Board – Infrastructure Action. Script 3. 0 ◦ Not meant for database communication ◦ Can’t communicate with other non-movie entities ◦ Can only reference the calling entity The web browser in this case Called External. Interface 4/9/2009 31

Mi. Board - External. Interface has a call property External. Interface. call( “my. Func”, “my. Param” ) ◦ invokes the calling entity’s my. Func function with the parameter my. Param 4/9/2009 32

Mi. Board - External. Interface Mi. Board web page is JSP Contains Chat Movie and Game Movie External. Interface calls reference Java. Script functions Movies call Java. Script functions that call Action. Script functions in the opposite movie 4/9/2009 33

External. Interface - Diagram 4/9/2009 34

Mi. Board - External. Interface Example: Player 2 has moved 3 spaces Game Movie ◦ Tells the Java. Script to tell the chat Player 2 moved 3 spaces Chat Movie ◦ Broadcasts the public message to all connected players ◦ The receiving chat movie tells its Java. Script to tell the board movie the passed message Game Movie ◦ Parses the message, and moves Player 2’s token 4/9/2009 35

Mi. Board – Future Work Next version release Fall 2010 Will include automatic SE analysis using LSA Ability for users to buy Skins for the game ◦ Change the board theme, color, etc. Difficulty of text and strategy will adapt to each player’s ability level 4/9/2009 36

Acknowledgements Special thanks to the i. START Game Team at the University of Memphis under Dr. Danielle S. Mc. Namara, and Dr. Mike Rowe for his invention Thank you VMASC for hosting the Capstone Conference Thank you to my audience for attending 4/9/2009 37

References D. S. Mc. Namara, I. B. Levinstein, and C. Boonthum, “i. START: Interactive strategy trainer for active reading and thinking, ” in Behavior Research Methods, Instruments, and Computers, vol. 36, 2004, pp. 222 -233. M. T. H. Chi, M. Bassok, M. Lewis, P. Reimann, and R. Glaser, “Self-explanation: How students study and use examples in learning to solve problems, ” Cognitive Science, vol. 13, 1989, pp. 145 -182. M. T. H. Chi, N. De Leeuw, M. Chiu, and C. La. Vancher, “Eliciting self explanations improves understanding, ” Cognitive Science, vol. 18, 1994, pp. 439 -477. D. S. Mc. Namara, C. Boonthum, , I. B. Levinstein, and K. K. Millis, “Evaluating self-explanation in i. START: Comparing word-based LSA systems, ” in T. Landauer, D. S. Mc. Namara, S. Dennis, and W. Kintsch eds. , Handbook of Latent Semantic Analysis, Lawrence Erlbaum, Mahwah, NJ, 2007, pp. 227 -241. D. S. Mc. Namara, “SERT: Self-explanation reading training, ” Discourse Processes, vol. 38, 2004, pp. 1 -30. T. O’Reilly, G. P. Sinclair, and D. S. Mc. Namara, “Reading strategy training: Automated versus live, ” Proceedings of the 16 th Annual Meeting of the Cognitive Science Society, Austin, TX: Cognitive Science Society, 2004, pp. 1059 -1064. 4/9/2009 38

References (cont’d) T. O’Reilly, R. Best, and D. S. Mc. Namara, “Self-explanation reading training: Effects for low-knowledge readers, ” in K. Forbus, D. Gentner, and T. Regier eds. , Proceedings of the 26 th Annual Meeting of the Cognitive Science Society, Mah. Wah, NJ: Erlbaum, 2004, pp. 1053 -1058. T. P. O’Reilly, G. P. Sinclair, and D. S. Mc. Namara, “i. START: a web-based reading strategy intervention that improves students’ science comprehension, ” in Kinshuk, D. G. Sampson, and P. Isaias eds. , Proceedings of the IADIS International Conference on Cognition and Eploratory Learning in the Digital Age: CELDA, Lisbon, Portugal: IADIS Press, 2004, pp. 173 -180. R. S. Taylor, T. O’Reilly, G. P. Sinclair, and D. S. Mc. Namara, “Enhancing learning of expository science texts in a remedial reading classroom via i. START, ” in S. Barab, K. Hay, and D. Hickey eds. , Proceedings of the 7 th International Conference of the Learning Sciences, Mahwah, NJ: Erlbaum, 2006. T. O’Reilly, R. S. Taylor, and D. S. Mc. Namara, “Classroom based reading strategy training: Selfexplanation vs. reading control, ” in R. Sun and N. Miyake eds. , Proceedings of the 28 th Annual Conference of the Cognitive Science Society, Mahwah, NJ: Erlbaum, 2006, pp 1887 -1892. J. P. Magliano, S. Todaro, K. K. Millis, K. Wiemer-Hastings, H. J. Kim, and D. S. Mc. Namara, “Changes in reading strategies as a function of reading training: A comparison of live and computerized training, ” Journal of Educational Computing Research, vol. 32, 2005, pp. 185 -208. M. E. Gredler, “Games and simulations and their relationships to learning, ” in D. H. Jonassen ed. , Handbook of research on educational communications and technology, Mahwah, NJ, US: Lawrence Erlbaum Assoc. , 2 nd ed. , 2004, pp. 571 -582. M. Rowe, “Alternate forms of reading comprehension strategy practice and game-based practice methods, ” Doctoral Dissertation, Psychology Department, the University of Memphis, 2008 4/9/2009 39

Thank You! Questions? 4/9/2009 40

Appendix A – i. START: The Board Game Rules Pieces: ◦ 4 Game boards ◦ 6 Player Tokens ◦ 1 Monster Token ◦ 120 event cards ◦ 6 sets of 5 strategy cards ◦ 20 Task cards ◦ 20 power cards 4/9/2009 41

Appendix A – i. START: The Board Game Rules Player Turn: Take the top card off the Task Card deck (Do not show other players) Read a passage from the text selected aloud ◦ Read at least one sentence. For more advanced players, multiple sentences can be read. ◦ If using the same text as other players Continue where last reader left off, or If you are the first reader, select a place to begin reading ◦ If using a different text than other readers Continue where you left off, or If you are just beginning, select a place to begin reading Self-explain the text aloud, using one or both strategies on the Task Card so the other players can hear you ◦ If the reader uses one strategy correctly, the reader gets all the points listed next to the strategy. ◦ If the reader uses both strategies correctly, the reader gets double the larger point value on the card. 4/9/2009 42

Appendix A – i. START: The Board Game Rules All the other players will attempt to guess what strategy the reader used ◦ Other players (guessers) will place one of their Strategy Cards face down in front of them All guessers will turn over their Strategy Cards at once. ◦ Beginning to the reader’s left and continuing clockwise each guesser should state what their guess is ◦ If there is no disagreement, score points If the strategy matches how the reader self-explained, and is on the Task Card, the guesser gets half the points listed next to the strategy rounded down If the strategy matches how the reader self-explained, but is NOT on the Task Card, the guesser gets 1 point If the strategy does not match how the reader self-explained, the guesser gets no points. ◦ If there are disagreements, do not score points until disagreement is resolved 4/9/2009 43

Appendix A – i. START: The Board Game Rules All players discuss whether the strategy use and guesses were correct. ◦ Beginning with disagreements about the reader’s strategy use. A majority of players must agree that the reader did not use a specific strategy. The reader can attempt to explain their self-explanation and show it was a correct use. If a majority still disagrees, the reader can try again to use the strategy again for half points ◦ After the reader’s strategy use has been resolved, continue with the guesser to the reader’s left. A majority of players must agree that the reader did not use the guessed strategy. The guesser can attempt to explain why their guess is correct and where it was used in the self-explanation. If a majority still disagrees, no points are scored After the disagreement is resolved continue clock-wise to the next disagreement. 4/9/2009 44

Appendix A – i. START: The Board Game Rules The player may now use any Power Card they have Roll dice Move token all the spaces on the die Take an Event Card ◦ Perform the action on the event card After all players have completed one (1) turn the round ends ◦ Roll 1 die for monster movement ◦ Move the monster half the number shown rounded down [13] 4/9/2009 45

Appendix B – Mi. Board Screens Game Board ◦ The basic game board of Mi. Board includes the playing field, 4 player tokens, a message box, a list of players with associated scores and tokens, a button for drawing event cards, seeing the text, and getting help. The event cards cannot be drawn until after the player rolls. 4/9/2009 46

Appendix B – Mi. Board Screens Chat ◦ The chat is used for the idle players to converse and for sending messages between connected players. The chat is also the medium in which players discuss disagreements in voting. The chat is only enabled during discussions and when the players are idle. In order to retain the attention of the idle players, they are allowed to chat with other idle members of the game. 4/9/2009 47

Appendix B – Mi. Board Screens 4/9/2009 The Discussion includes a set of rules (in red) and enabling of the chat room. This player has forfeited his responses by clicking the “Pass” button. After the discussion, the players see the Guesser screen, where they may select as many strategies as they like. 48

Appendix B – Mi. Board Screens Reader Screen ◦ At the Reader Screen, the Reader reads the sentence for which he is to provide a SE, and types his SE, focusing on the provided strategy. He has the option of choosing a random, new strategy or a random, new point value by clicking on the appropriately labeled buttons. 4/9/2009 49

Appendix B – Mi. Board Screens Guesser Screen ◦ At the Guesser Screen, players select the strategy they think was focused on by the Reader. The Guesser may only choose one such strategy at this stage in the game. 4/9/2009 50

Appendix B – Mi. Board Screens 4/9/2009 The Cascading Menu Block is part of the Guesser Screen. It is called cascading because each time a user clicks on a check box, a new screen appears. A use is asked to click a strategy, then a reason for that selection (such as, Linked to a specific sentence), and then is asked to highlight the part of the SE in which that particular strategy was used. 51

Appendix B – Mi. Board Screens Summary Screen ◦ The Summary Screen provides a summary of the explanations built by the Cascading Menu Block, as well as a summary of points earned in the round. 4/9/2009 52

Appendix B – Mi. Board Screens Power Cards ◦ A user may use a power card by clicking on the blue power card button to bring up the power card screen 4/9/2009 53