Eight Serious Games at ICT Lessons Learned and

Eight Serious Games at ICT: Lessons Learned and Challenges Identified Michael van Lent USC Institute for Creative Technologies 19 March 2018

Business of Games • 60% of Americans play video games • $25 Billion dollar industry worldwide (2004) • $7. 3 Billion dollars in the US (2004) • $6. 1 billion in 1999, $5. 5 billion in 1998, $4. 4 billion in 1997. • $12 billion in 2006? • One day sales records • Halo 2: $125 million in a single day • Harry Potter (Half-blood Prince): $140 million single day • Consoles dominate the industry • Most of the sales (Xbox, Play. Station 2, Game. Cube) • Average of game players is 30 • Average of game buyers is 37 • 55% of game players are men

Games vs. Learning = Learning

Serious games • An entertaining virtual experience that’s purpose goes beyond entertainment • Serious game purposes include: • • Education Training Communications Public Policy Marketing Mental Health Therapy Medical Diagnosis

Full Spectrum Command Boardgame • Training objectives • Originally just to validate the game design • Focus of attention for Company Commander • Commercial Partners • Legless Productions • Research technologies • None • Notes • 4 month production cycle

Full Spectrum Command 1. 0 • Training objectives • Military Decision Making Process • Course of action development • Course of action adaptation • Commercial Partners • Quicksilver Software • Legless Productions • Research technologies • e. Xplainable AI • Notes • 60% of the CPU goes to AI

Full Spectrum Warrior • Training objectives • Squad-level command • Squad-level team building • Commercial Partners • Pandemic Software • Sony Imageworks • Legless Productions • Research technologies • e. Xplainable AI • Notes • 2004 E 3 • Best original game • Best simulation game

Full Spectrum Command 1. 5 • Training objectives • Military Decision Making Process • Course of action development • Course of action adaptation • Commercial Partners • Quicksilver Software • Legless Productions • Research technologies • e. Xplainable AI • Adaptive Opponents • Notes • Funded by Singapore

Full Spectrum Leader • Training objectives • Platoon-level command • Combined arms coordination • Commercial Partners • Quicksilver Software • Legless Productions • Research technologies • Adaptive Opponents • Notes • First game that allows the player to shoot

Every Soldier a Sensor System (ES 3) • Training objectives • Presence patrols • Military intelligence gathering • Commercial Partners • Warner Bros. Online • Research technologies • None • Notes • 90 days from project start to fielded training system • Do. D 2006 Annual Modeling & Simulation Awards • Training

Joint Fires & Effects Training System • Training objectives • Forward observer missions in urban environments • Close-air support (CAS) calls for fire • Commercial Partners • Buzz. Monkey Software • Game Production Services • Research technologies • Adaptive Opponents • Notes • Installed at Ft. Sill • Over 10, 000 soldiers trained to date

ELECT ATO • Training objectives • Bi-lateral negotiation • Cultural awareness • Commercial Partners • Real Time Associates • Game Production Services • Research technologies • e. Xplainable AI • Intelligent Tutoring • Social & Cultural Simulation • Notes • Boardgame prototype

Lessons learned • What is a video game vs a simulation? • Video game: A virtual experience carefully designed to be entertaining (among other things) • Simulation: A recreation of key aspects of reality in a virtual environment • The two are not mutually exclusive • Identify the learning objectives first • Seems obvious in retrospect • Guided learning is better than unguided discovery • But both have their place • Games must be part of a larger curriculum • Practice and maybe demonstration

Lessons learned (con’t) • Calling it a game just gets you in the door • Games aren’t motivating, but games are fun which can be • Calling it a game sets an initial expectation • But you still need to make sure it’s a fun game • Immersion helps learning • A article of faith • Best opponent isn't the strongest opponent • Serious games have different requirements • Graphics (FSW vs SLIM-ES 3) • AI

Adaptive Game Intelligence • Three inspiring occurrences: • FSC game designers “tricking” the SMEs

Adaptive Game Intelligence • Three inspiring occurrences: • FSC game designers “tricking” the SMEs • Looking behind the AI curtain in Age of Empires

Static Game AI ; The AI will attack once at 1100 seconds and then again ; every 1400 sec, provided it has enough defense soldiers. (defrule (game-time > 1100) => (attack-now) (enable-timer 7 1100)) (defrule (timer-triggered 7) (defend-soldier-count >= 12) => (attack-now) (disable-timer 7) Age of Kings (enable-timer 7 1400)) Microsoft

Adaptive Game Intelligence • Three inspiring occurrences: • FSC game designers “tricking” the SMEs • Looking behind the AI curtain in Age of Empires • B-training at Ft. Sill

Adaptive Game Intelligence • Most entertainment game AI is static and scripted • Each level is played a small number of times • Learning to beat the script is fun • Variability is expensive • Quality assurance is manageable • Adaptation is risky • Designer can’t control the player’s experience • Serious games required variable and adaptive AI • Each level is played a large number of times • Prevent gaming the game • Variability & adaptation are essential • Address student’s specific needs • Give instructors “sufficient” control

What does “Adaptive” mean? • Adapts to the student’s history • Variability • Adapts to the student’s needs • Pedagogical reasoning • Adapts to the instructor’s input • Semi-directed

Variable Behavior • Previous efforts have varied individual behaviors • Behavior parameters, behavior libraries… • Usually less effective variations on a base behavior • Our approach: Variability at the strategy-level • Generate a new strategy for each game session • Keep track the strategies the student has already seen • Strategic behavior vs. Tactical behavior • • Two-level AI system: Strategy Planner + Execution System Planner: Generate an abstract strategy Execution System: Instantiation, execution & plan tracking Planner: Replanning

Planning Approaches • Single agent, automated planning (LPG, SHOP…) • Planning vs. Scripted student model

Planning Approaches • Single agent, automated planning (LPG, SHOP…) • Planning vs. Scripted student model • Multi-agent, adversarial planning (AP, alpha beta. . . ) • Planning vs. Planning student model

Planning Approaches • Single agent, automated planning (LPG, SHOP…) • Planning vs. Scripted student model • Multi-agent, adversarial planning (AP, alpha beta. . . ) • Planning vs. Planning student model • Multi-agent, collaborative planning (ASD & ITS) • Opponent works with the player but appears as an adversary • Intelligent tutor helps steer the player

Collaborative Planning

Planning Approaches • Single agent, automated planning (LPG, SHOP…) • Planning vs. Scripted student model • Multi-agent, adversarial planning (AP, alpha beta. . . ) • Planning vs. Planning student model • Multi-agent, collaborative planning (ASD & ITS) • Opponent works with the player but appears as an adversary • Intelligent tutor helps steer the player • Pedagogic key framing • Instructor provides key frame world states • Planner fills in the gaps providing a complete plan