Interactive simulations for teaching physics Work supported by

Interactive simulations for teaching physics Work supported by: NSF, Hewlett Foundation, Kavli Foundation, Univ. of Colorado, me and Sarah

Physics Education Technology Project (Ph. ET) Develop interactive simulations Research on simulation design and effectiveness When simulations carefully tested and refined : • Highly engaging • Very effective for learning • Work with very wide range of students (“grade school to grad school”) Goals for talk Examples of good simulations Little about research on what makes them useful principles to keep in mind when using.

show website, sim list, balloons and sweater, moving man, elctromag Ph. ET (phet. colorado. edu) • ~ 60 interactive simulations • Intro physics, modern physics, some chemistry, bit of math, starting to expand into geo and bio, … • Phet-based activities database on website--Trish Loeblein run phet sims (all free!): • directly from web (regular browser, platform independent) • download whole website to local computer for offline use 2006 -- 1 Million sims launched off website; 50, 000 full site downloads Extensive development and testing process--teams (faculty, software engineers, sci. ed. specialists)

Physics faculty: Michael Dubson Noah Finkelstein Kathy Perkins (manager) Carl Wieman Phet Staff Postdocs: Sam Mc. Kagan Linda Koch (Chem) Software Engineers: Ron Le. Master Sam Reid Chris Malley Michael Dubson Grad students: Wendy Adams Danielle Harlow Chris Keller Noah Podolefsky HS Teacher: Trish Loeblein ~6 full time equivalents Staff: Mindy Gratney, Linda Wellmann

Design Features and Criteria • Engaging and productively fun (interface design, appearance, …) • Connection to real world • Highly interactive- stuff happening, user controls • Explicit visual & conceptual models (experts’) • Explore and discover, with productive constraints • Connect multiple representations K. K. Perkins, et al, “Ph. ET: Interactive Simulations for Teaching and Learning Physics”, Physics Teacher (Jan 2006)

Most important element--testing with students 1. Think aloud interviews (~200 hours) Explore with guiding question surprisingly consistent, always revealing 2. Observations of use in lecture, recitation and lab, homework solving sessions.

Example- of what revealed by interview studies. Radio waves. Initial startup. Experts- - really like. Students--Watch without interacting. Don’t like. Misinterpret.

Start with curve view, manually move electron. Very different result. Later move to full field view, manipulate, like, and understand. Correctly interpret. Why do you think starting this way works so much better? briefly discuss with neighbors, then will collect ideas

Why starting this way works so much better?

Why starting this way works so much better? Matches research on learning. • Cognitive demand. Novices don’t know what to focus on. treat everything equally important. Much more than short-term working memory can handle, overwhelming • Construction of understanding. Other important features: Visual model-electrons in transmitting and receiving antennas, display of waves Interactivity

Example illustrates important principle: students think and perceive differently from experts Good teaching is presenting material so novice students learn from it, not so looks good to experts! Violated by most simulations (and many lecture demonstrations, figures in texts, …) Ph. ET sims almost never right the first time. Test and modify to get right.

Simulation testing microcosm of education research Routinely see examples of principles established in very different contexts. • cognitive load • construction of understanding • build on prior knowledge • connections to real world • exploration and deep understanding transfer • motivation--factors affecting and connections to learning • perceptions based on organizational structures, structures change and develop, changes perception. • …

Sims useful in variety of settings Pre-class or pre-lab Activity Lecture/classroom Visual Aids, Interactive Lecture Demos, & Concept tests Labs/Recitations Group activities Homework Need some structure--activities database advice and workshops Trish Loeblein ploeblei@jeffco. k 12. co. us bits of examples of effectiveness in different settings

show wave on string Lecture (Non-science Majors Course) Standing waves-- Sim vs. Demonstration Wave-on-string sim vs Tygon tube demo Follow-up Concept Tests: A B C snapshots at different times. 1. When the string is in position B, instantaneously flat, the velocity of points of the string is. . . A: zero everywhere. B: positive everywhere. C: negative everywhere. D: depends on the position. Correct : 2002 demo: 27% 2003 sim: 71% 2. At position C, the velocity of points of the string is. . . Correct : A: zero everywhere. B: positive everywhere. 2002 demo: 23 % C: negative everywhere. D: depends on the position. 2003 sim: 84%

Features that make a differenceexperts hardly notice, BIG difference for novices 1. Green beads on string that show moves up and down, not sideways. 2. Speed set so novice brain can absorb and make sense of it. 3. Can do controlled changes, most in response to student requests. Sort out what makes a difference and why.

What else does simulation provide over usual figure and explanation in lecture or textbook? 1. see direct cause and effect relations 2. explicit visual models example-electromagnet, friction, gas, cck

show cck Standard Laboratory (Alg-based Physics, single 2 hours lab): Simulation vs. Real Equipment DC Circuit Final Exam Questions p < 0. 001 N D. Finkelstein, et al, “When learning about the real world is better done virtually: a study of substituting computer simulations for laboratory equipment, ” Phys. Rev: ST PER 010103 (Sept 2005)

f. A few important interface characteristics* • Intuitive interactivity vital • Controls Intuitive when most like hand action – Grab-able Objects – Click and Drag – Sliders to change numeric values • Representations – Cartoon-like features scale distortion OK – Good at connecting multiple representations, but proximity and color coding helps (energy sktprk) *more than want to know in Adams et al. papers

Conclusions: Interactive simulations powerful new technology for learning science. But not automatically good. Trish Loeblein ploeblei@jeffco. k 12. co. us Phet. colorado. edu booth in APS show- play with, discuss. references on website under “research”

Surprising differences in motivation and learning engaged exploration vs. “performance mode” Before topic covered in classactively interested and engaged in figuring out answer After covering in class (same students, same topic!) struggle to remember what told or read, not use sim to explore and figure out, even with repeated encouragement! Frustrated and unsuccessful! (ed. research microcosm cont. --matches perfectly with results of Dweck psychology studies- “performance mode” )

General results from student interviews* 1. Think aloud interviews (~200 hours) Explore with guiding question a. Surprisingly consistent responses, particularly on interface. b. Vocabulary very serious hindrance to learning and discussion-- see because simulation removes c. Animation attention, but not thinking. Interactivity thinking & learning. W. K. Adams, et al. , A Study of Educational Simulations Part I - Engagement and Learning. , A Study of Educational Simulations Part II - Interface Design. ,

Ph. ET’s Checklist for creating Effective Sims and Tutorials/Activities Is Sim or Activity …. 1) Aligned with learning goals? 2) Aligned with current research? 3) User-tested? • • Ease of use Correct Interpretation Engagement Achieves learning goals

Interesting results from interview studies* (cont. ) d. the good, the bad, and the evil sim Good sim is extremely effective for wide range of students: understand difficult concepts, can explain & apply to real world situations. Bad sim- very little learned. Awkward distracting interface, boring, confusing. evil sim--effective at teaching wrong things! e. Student testing critical! Interviews always reveal undesired perceptions or distractions in first versions!

many other examples of power of visual models all of quantum! (S. Mc. Kagan) quantum wave interference lasers Stern-Gerlach MRI major impact on student thinking tunneling … Rethinking how intro quantum mechanics should be taught. Key missing element.

Some general principles revealed for how people learn physics • Cognitive load important--too much stuff overwhelming, Build up slowly works well. (Radio waves full view) • Need to connect to prior thinking-- start using familiar elements (tire pump in ideal gas), build understanding • Visual models vital (balloon and sweater, CCK, QM, …) • Absorb and make sense only when ask questions and then seek answers. Animation without interactivity draws attention, but not exploration and understanding. • Text and explanations usually ignored or misinterpreted unless seeking that particular point.

Integrating a sim on a topic (Lect. & HW) (Photoelectric Effect in Modern Physics) (S. Mc. Kagan, to be pub. ) Univ. of Wash. : • Student learning of photoelectric effect deficient • Developed & used Photoelectric Tutor (PT) Exam Q: What would happen to current reading if you: Q 1: Changed metal. Why? Q 2: Double intensity of light. Why? Q 3: Increased DV across electrodes. Why? % Correct show photoelectric effect CU: Incorporated sim Course Q 1 Q 2 Q 3 N UW w/o PT 65 40 20 26 UW w/ PT 75 85 40 36 CU Fa 05 91 87 85 189 CU Sp 06 86 88 84 182

Visual Models

Development Process Team for each sim faculty content expert(s) sci. ed. /user expert(s) software engineer Learning Goals Initial Design ~Final Design Research Base Classroom Use Interviews Redesign b Interviews

Initial Design & General Approach Research base: • Ed. Psych / Cog. Sci: How people learn • Educational Software Design • Student Conceptions in Physics • Ph. ET research findings

Assessment of Design: • Usability – easy/intuitive • Interpretation – correct/productive • Engaged exploration General Design Guidelines • Can students construct understanding of main ideas? Achieve learning goals?

+ 0 time Velocity Demo 4: Sketch position vs time and velocity vs time graphs for when Moving Man: walks steadily towards the tree for 6 seconds, then stands still for 6 seconds, and then towards the house twice as fast as before for 6 seconds. Position Lecture – Interactive Lecture Demos Thorton and Sokoloff, 1997 + 0 - time

0 time Position + A Position time 5 s 10 s 15 s 20 s + C 0 Velocity 0 time Position Velocity + time + 0 - time 5 s 10 s 15 s 20 s + D 0 time - + 0 time 5 s 10 s 15 s Velocity - - 0 - - - + B 20 s + 0 - time 5 s 10 s 15 s 20 s

Lecture – Concept tests Electromagnetic waves: Radio Waves sim Concept Tests and Peer Instruction The speed of the wave (signal) is measured as… a. how fast this peak moves to the right. b. how fast this peak moves up and down. c. could be a or b

Incorporating a suite of sims (Modern Physics for Engineers) ~ 10 interactive simulations in lecture and homework (clickers, real life applications, conceptual homeworks) Quantum Mechanics Conceptual Survey normalized Post gain Course Reformed with sims Traditional Pre N Eng. Sp 06 Eng. Fa 05 Eng. Sp 05 Phys. Sp 06 Phys. Fa 05 30 32 (30) 44 40 65 69 51 64 53 0. 50 0. 55 0. 30 0. 36 0. 21 156 162 68 23 54 Phys. Sp 05 (44) 63 0. 33 64

Interactive Recitation Study Reformed large-scale introductory calculus-based physics course with Tutorials vs. “CCK” (N=180) “Real” (N=185) Keller, C. J. et al. Assessing the effectiveness of a computer simulation in conjunction with Tutorials in Introductory Physics in undergraduate physics recitations, ", PERC Proceedings 2005

Interactive Recitation Study p=0. 01 but end of semester exam, no observable difference