Putting Teachers in the Loop : Tools for Creating and Customising Simulations

The complex nature of these factors leads to the conclusion that the notion of a ' d e f i n i t i ve' simulation interface is a myth. Simulation users must be empowe red by tools allowing them to take control of the design process. The range of changes which could be facilitated by giving teachers the tools to alter simulation visualisations are discussed and demonstrated with examples of simulations and online learning materials produced using a suite of tools for creating and customising educational simulations, Je L S I M .


B a c k g r o u n d
When Bill Gates said "Content is king"1 , he understood that simply moving printed information to the web isn't enough.Sa d l y, many producers of online learning don't seem to have listened.Instead of exploiting the real benefits of computer based d e l i ve ry (calculation, sorting, querying, retrieving and presentation), to cre a t e i n t e r a c t i ve environments that engage learners, online learning materials are typified by static content and linear stru c t u res and amount to little more than electro n i c books (e.g.WBEC (2000) 2 ).e-Learning materials are still routinely judged by q u a n t i t a t i ve criteria (how many hours of learning online?) rather than qualitative ones (has the student learning experience been improved?).As the cost of pro d u c i n g 'text and images' (or conve rting from existing materials) is far less than pro d u c i n g i n t e r a c t i ve components, this diet of 'text and images' continues to pre d o m i n a t e .
T h e re is howe ver a specific need for interactive content in an online setting.He l p i n g the learner to engage with the learning material is vital where access to a tutor may be restricted.By creating content that is task based, or designed to challenge existing understanding, the student is encouraged to adopt a more active role in their learning.No matter how flexible and interactive the learning material howe ve r, the teacher cannot simply give the student a re s o u rce and leave them to get on with it.One of the key skills of educators (where ver they fall on the constructivistbehaviourist spectrum) is in tailoring or customising the material that is available to the student to reflect their understanding of the learner's needs.Cu r rent technological advances which standardise the exchange and storage of blocks of learning content and provide effective ways of discovering content3 , empowe r teachers by giving them a greater degree of control over what gets delive red to students.If teachers want to change the material then they may alter the text or replace other components.Howe ver they wouldn't normally be able to alter the i n t e r a c t i ve content such as simulations as its' appearance and functionality is fixe d .This paper examines the possibilities that are created if simulations are made as easy to customise as text or images, allowing simple interactive re s o u rces to be tailored for use in a wide range of teaching situations.The paper considers the needs of the Putting Teachers in the Loop: Tools for Creating and Thomas & Milligan (2004) Customising Si m u l a t i o n s

. The Educational Value of Simulations
The term 'simulation' is being applied in e-Learning in an increasingly broad manner and is sometimes used synonymously with 'animation', howe ve r, for the purposes of this paper a simulation is defined as having the following two key feature s : 1. T h e re is a computer model of a real or theoretical system that contains information on how the system behave s .
2. Experimentation can take place, i.e. changing the input to the model affects the output.
As a numerical model of a system, presented for a learner to manipulate and explore , simulations can provide a rich learning experience for the student.They can be a p owe rful re s o u rce for teaching: providing access to environments which may o t h e rwise be too dangerous, or impractical due to size or time constraints; and facilitating visualisation of dynamic or complex behaviour.
Ed u c a t i o n a l l y, simulations have a unique role in supporting learning as they allow learners to directly manipulate a system and to observe the effect of the change, p roviding a form of intrinsic feedback.This interaction between the learner and learning material allows students to develop a feel for the relationship between the underlying factors governing a system, promotes an appreciation of appro p r i a t e ranges for system parameters, and gives a qualitative feel for the system before the i n t roduction of theory (Thomas and Neilson (1995)).Simulations can be used as c o g n i t i ve tools, allowing students to manipulate the parameters to test hypotheses, t rying out "what if " scenarios without a real consequence or risk and in a time frame that is convenient and manageable to them, they enable the learner to ground their c o g n i t i ve understanding of their action in a situation.(Laurillard (1993)) A constructivist view of learning (Jonassen (1994), Duffy and Cunningham (1996), Wilson (1997)) encourages educators to recognise their students' strongly held p reconceptions and knowledge to provide learners with experiences that will help them revise and build on their current understanding of the world.The student should not be a passive participant, but should actively engage in the experience, which should allow exploration and encourage reflection.Clearly, simulations have the potential to form an important component of a constructivist learning e n v i ronment (Jonassen, et al. (2003) Simulations can be used to provide realistic problems and scenarios.Ad vocates of situated learning (Winn (1993), Brown, Collins & Duguid (1989)) believe it is easier for learners to apply new concepts if they are acquired whilst undertaking authentic tasks in real world situations or re p resentations of them.Schank and Cleary (1995), p roponents of case-based reasoning believe that learning by doing in meaningf u l situations is an important component of education.
Simulations have the advantage over other media that they can bring both reality and interactivity to eLearning.They provide a form of feedback that facilitates exploration in a manner that can mimic scientific method, allowing students to e x p l o re and build their own understanding.

. Factors Affecting Simulation Design
Simulations clearly have educational potential.Howe ve r, re s e a rch shows that the educational benefits of simulations are not automatically gained and that care must be taken in many aspects of simulation design and presentation.It is not sufficient to provide learners with simulations and expect them to engage with the subject matter and build their own understanding by exploring, devising and testing hypotheses.Rieber (1996) cautions designers of interactive learning enviro n m e n t s , "not to assume that explicit understanding will follow even if users are successful at completing a task".Learners must be guided and supported in their use of s i m u l a t i o n s .
Educators may wish to focus the learner's use of the simulation by setting the scene, p roviding objectives, directions, context, tasks and problems.The need for additional s u p p o rt, in the form of guidance, feedback and scaffolding has been recognised for some time (Thomas and Neilson (1995), Pilkington & Grierson 1996)).From within a simulation, it is possible to provide feedback and guidance in the form of hints, c o r re c t i ve feedback, tips on ro l l ove r, highlighting or the addition of elements to augment reality (de Jong and van Joolingen, 1998).Feedback is important, but not Putting Teachers in the Loop: Tools for Creating and Thomas & Milligan (2004) Customising Si m u l a t i o n s all teachers will agree on how much to provide in a given subject, e.g.providing too much feedback wouldn't support constructivist goals.Su p p o rt for simulation use can come from human experts (teachers, coaches, guides) or from peers as well as from electronic help and guidance mechanisms.Ex p e rts can p rovide considerable support during simulation use, but when they are not pre s e n t , s u p p o rt provision can attempt to duplicate their role (e.g.Ac ovelli and Ga m b l e (1997)).Two techniques used in this area are coaching and scaffolding.Coaching i n vo l ves monitoring and regulating learner's performance, provoking reflection and p e rturbing learner's models (Jonassen (1999)), Scaffolding invo l ves manipulating the task to supplant the students ability to perform the task, either by changing the n a t u re or difficulty of the task or imposing the use of cognitive tools.Some support and guidance can also be provided by linking the simulation with other multi-media re s o u rces, which provide answers to frequently asked questions, but within typical simulation software, feedback related to the students' immediate need or curre n t task, or assessing their performance, is often limited or absent.
It has been noted, that learning with simulations in a discove ry learning enviro n m e n t is often not as effective as expected (Lee, 1999;de Jong and van Joolingen, 1998).A p a rticular problem seems to be difficulties students have in generating hypotheses ( Shute and Gl a s e r, 1990; Njoo and de Jong, 1993).Quinn and Alessi (1994) suggest that "Prior understanding of the scientific method may be necessary to learn fro m simulations".Howe and Tolmie (1998) demonstrated the benefits of contingent p rompting in hypothesis formation.
It is important that students engage with the underlying simulation model, not just with the user interface.As Davies (2002) points out, interactivity is not synonymous with engagement.Pilkington and Pa rk e r -Jones (1996) noted the tendency for students to concentrate on manipulating objects without generating a deeper understanding of the model or principles behind the observed behaviour.Laurillard (1993) draws the distinction between qualitative reasoning: incorporating know l e d g e of real world objects and quantitative reasoning: referring only to quantities and p rocesses explicitly presented on the screen and suggested that the more interpre t i ve a p p roach of qualitative reasoning must be encouraged.One method of doing this is to allow students to construct the models themselves, for example using systems dynamics software such as STELLA4 or Powe r Si m5 .Though Alessi (2000) is of the opinion that for most educational purposes, such models should be enhanced with an i n s t ructional ove r l a y.Alessi also points out that depending on the instru c t i o n a l Putting Teachers in the Loop: Tools for Creating and Thomas & Milligan (2004) Customising Si m u l a t i o n s Journal of In t e r a c t i ve Media in Education, 2004 (15) Page 6 paradigm adopted it may be advantageous to expose the model (glass-box), e.g. in e x p o s i t o ry learning or to hide the model (black-box), e.g. in discove ry learning where students must discover it for themselves.Thus, educators may wish to control the d e g ree of access the learner has to the internal model.
Allen, Hays and Bu f f a rdi (1986) suggest that physical fidelity (look and feel) is more i m p o rtant in cases of development of skills which invo l ve little or no cognitive effort in their execution, while functional fidelity (realistic cause-effect relationships) is m o re important in tasks that depend on deeper cognitive processing of task information.Howe ve r, providing an authentic experience by maximising the fidelity with which it duplicates the real world does not always engender learning.Alessi (1988) suggested that increased fidelity could inhibit initial learning for novices by ove rwhelming them with details that they cannot assimilate, but conve r s e l y d e c reasing fidelity may engender learning that is not transferable to the real world.He proposed a solution of dynamic fidelity that is determined by instru c t i o n a l e f f e c t i veness and changes based on learner performance.Examples of functional fidelity that can be introduced to students as they pro g ress are experimental error and malfunctioning equipment, e.g.Magin and Re i zes (1990) and Thomas and Mi l l i g a n (2003).Again, the educator could benefit greatly from having control over the fidelity of the simulation.
C l e a r l y, there is no "right" way to present a topic using a simulation.Learners of d i f f e rent age, ability and level re q u i re different approaches in their use of simulations.Teachers may wish to adopt different instructional strategies.De c i s i o n s about the type of task, the type of support and how to ensure engagement will be reflected in the visualisation presented to the learner.Id e a l l y, these decisions should be made by educators, as it is they who have the subject expertise, teaching experience, and knowledge of typical student misconceptions.If a teacher finds a p a rticular approach does not work, he or she may well wish to modify a simulation to remedy this.Consideration of these points leads to the conclusion that a single f i xed interface designed by a non-teacher is never going to be the optimal solution.

. Modification of Simulations
Simulations generally re q u i re programming skills to produce them and as a result are e x p e n s i ve to create, and can be difficult to modify.In the past, there f o re, teachers h a ve been effectively disenfranchised from invo l vement in simulation design.A teacher may wish to alter any of the features relating to factors identified in the p revious section.The following section looks at the specific reasons why teachers Putting Teachers in the Loop: Tools for Creating and Thomas & Milligan (2004) Customising Si m u l a t i o n s may wish to modify visualisations for a given model.The modifications are c o n s i d e red in three categories, simple presentational issues, the scenario addressed by the visualisation and the instructional overlay applied to that scenario.
• Learning Styles and Preferences: Learners have different learning styles, p re f e rences and abilities.Di f f e rent media, forms of internal feedback in the simulation (Rieber (1996)), graphical user interface metaphors (Cates (2002)) or multiple re p resentations (Ainsworth (1999)) can be utilised to suit specific n e e d s .
• Aesthetics: By altering the look and feel of an interface, changing colour or fonts, adding logos and images, a teacher can personalise an interface and gain a sense of ownership of the re s o u rce.Learners, too, are more at ease with materials with a familiar 'look and feel'.
• Minor Details: Sometimes teachers may wish to make minor changes to the i n t e rface, for example to naming conventions, units and symbols appro p r i a t e to the curriculum they are teaching.Minor changes can counter a 'not i n vented here' attitude, and aid localisation.

Simulation scenario:
• Re-using the model: T h e re are many examples of situations where the same underlying mathematical model can be used to describe different phenomena in different disciplines, (e.g.feedback control theory, exponential grow t h ) .Ta i l o red visualisations of a general purpose model can provide re s o u rces in a range of disciplines.
• Scenario Focus: Teachers may wish to modify a scenario to focus the student on a particular aspect of the model, to provide a different task or problem to teach the same subject in different disciplines (e.g.chemistry for biologists).This can be facilitated by changing the input variables the student is allowe d to modify or providing a different default starting condition for the simulation.They may also wish to set up scenarios to stimulate learners into testing 'what if ' scenarios in an attempt to understand models and pro c e s s e s at a deeper leve l .
Journal of In t e r a c t i ve Media in Education, 2004 (15) • Student expertise: The same model may be of use at different educational l e vels within the same subject.Dynamic fidelity, mentioned in the pre v i o u s section, can be used to change the interface to suit the learner's needs.T h e model can be simplified for novice users allowing them access to a limited number of the model variables or to provide a user interface suitable for their needs and skills.For example, limiting the use of technical terms for the n ovice or avoiding the use of graphs and numbers for the less numerate l e a r n e r.In t roducing the form of a system before discussing the complexity of the model is of general use when introducing students to any new topic.
• Exposure of the model: As highlighted in the previous section, depending on the educational aims, the teacher or designer may wish to va ry the degre e to which the learner is allowed to view and even manipulate the underlying m o d e l .

Instructional Overlay:
• Educational context: By altering the instructional overlay a simulation can be re-used within a variety of contexts, i.e. in lectures, as preparation for a real laboratory exe rcise or as a computer-based laboratory, for tutorials, as a re s o u rce in problem-based learning environment, as a game, for self-study or in assessment.
• Educational Approach: As can be seen from the sections 1.1 and 2, there is a spectrum of educational uses for simulations, from the open ended discove ry based learning through POE to more directed, expository learning, teachers would benefit greatly from the ability to produce their own visualisation of a pedagogically neutral simulation.
• Integration: Teachers may wish to integrate simulations into other learning material, in which case they will need to contextualize it and prov i d e i n s t ructions for the simulation use in the surrounding material.In a c o n s t ructivist or problem based learning scenario, the simulation can be made a vailable as one of a range of re s o u rces re l e vant to the domain being i n ve s t i g a t e d .
• Modifying the learner support: As discussed in the sections 1.1 and 2, d i f f e rent support mechanisms are re q u i red when teachers are present in a c l a s s room, when students are learning with others and have an opport u n i t y Putting Teachers in the Loop: Tools for Creating and Thomas & Milligan (2004) Customising Si m u l a t i o n s Journal of In t e r a c t i ve Media in Education, 2004 (15) to discuss and negotiate their understanding and when students are using simulations for self-study.The ability to change the feedback style, i.e. to what degree it is natural vs. artificial or immediate vs. delayed (Alessi and Trollip (2001) and the actual feedback wording would be of great benefit in tailoring re s o u rc e s .
A system that empowers the teacher to create and modify simulations benefits learners too.By providing learners with multiple re p resentations, different forms of feedback and choice in data presentation, they can be empowe red to choose the style of visualisation that best suits them.Mo re ambitiously learners could take over the role of teachers and construct interfaces to learn by teaching (Pl o e t z n e r, Di l l e n b o u r g , Praier & Traum (1999))

. Tools as an Answer
The JeLSIM toolkit is a suite of software written to support easy creation and customisation of simulations.It provides a toolkit for the rapid creation of Ja va applets through a visual interface not unlike a traditional drawing package.T h e simulations produced by the toolkit are small to medium sized desktop applications usable as components within an eLearning module or problem solving enviro n m e n t .
The underlying model can be any type (e.g.continuous, discrete or logical).The key f e a t u re of the tools is that they separate the model from the visualisation.T h e JeLSIM tools we re originally developed as the Mu l t i Verse toolkit with funding fro m the JISC Technologies Application Program (J TA P ) .
Like Reigeluth and Schwartz (1989), the JeLSIM toolkit makes a distinction betwe e n the computer model and visualisation.The following is a typical scenario for simulation production.A model is written by a Ja va programmer based on a design specification agreed after consultation with a design team (consisting variously of subject domain experts, teachers, educational designers, graphic designers).In p roducing the design specification, the design team broaden the problem out to consider more than just the immediate re q u i rements.The aim is that the model p roduced will be re-usable for a wide range of scenarios and levels, thus even though the immediate need may re q u i re a low fidelity interface where much of the model complexity can be hidden, a model will generally be designed to maximise its future re -u s a b i l i t y.The design of the JeLSIM toolkit means that even novice pro g r a m m e r s can create effective models because the Ja va code describes only the algorithm and not the user interface.The intention is that once written, the model code doesn't change and the programmer's invo l vement ends.The model code is then loaded into the JeLSIM tools and provides a template for all visualisations of the model.
A visualisation is created by a teacher or instructional designer.The tools themselve s p resent a visual environment for the creation of new visualisations, rather like a computer drawing package.A set of common visualisation objects -graphs, tables, digital inputs and outputs etc. are provided -which the visualisation author can use to display the value of any of the input and output pro p e rties for the simulation.A typical interface would consist of a number of input parameters and visualisation objects to illustrate the outputs.Once finished, the simulation is easily deployed (as a Ja va applet embedded in a web page, or even an IMS Content Pa c k a g e ) .
As the simulations are created as Ja va applets, they are easily integrated into existing web based learning materials.The context for the simulation can then be provided by altering the web content in the surrounding pages.The rationale is summarised in Fi g u re 1, below: one model written in java becomes the basis for many visualisations, c reated in the JeLSIM tools and embedded in web pages.

Fi g u re 1: The Se p a ration of Model and Vi s u a l i s a t i o n
O ver time, a library of models will be generated which can be used by others.At p resent around forty models are ava i l a b l e .
Putting Teachers in the Loop: Tools for Creating and Thomas & Milligan (2004) Customising Si m u l a t i o n s Journal of In t e r a c t i ve Media in Education, 2004 (15)

. 1 E x a m p l e s
The separation of model and visualisation outlined above allows the JeLSIM tools to s u p p o rt many of the types of customisation identified in section 4.This section p rovides examples of work where a series of individual models have been customised to produce a diverse range of re s o u rces.T h ree examples are examined in chemical kinetics, solar geometry and entre p re n e u r s h i p.

. 1 . 1 Solar Geometry
The solar geometry model calculates the location of the sun in the sky and the angle it would shine on a building, given a time, date and location on the earth's surf a c e .
A simulation of this model may be re l e vant to young children being taught about the seasons or the solar system, or to architects who need precise calculations to help them predict the incidence of sunlight on buildings.Clearly the same visualisation would not be appropriate for both sets of users but as the examples show, a single model can be used to provide alternate interfaces suited to the needs of these d i f f e rent groups.A number of interfaces have been constructed for this; they are listed below together with some of the customisation features they demonstrate: Calculation interface: For a building engineer, the model can be used purely as a calculation tool; the interface assumes prior knowledge of the subject, uses technical terms and provides numeric output.The calculation interface can be viewed and e x p l o red online6 .A screenshot is provided for illustration (Fi g u re 2).Journal of In t e r a c t i ve Media in Education, 2004 (15) Page 12 Primary school interface: In contrast, in the second example, to cope with differe n t learning styles re q u i red for primary school use, minimal text is employed, numbers a re not used and the sun's location is presented by an image of the sun that move s t h rough the sky over the period of a day.The visualisation is still generic, in that a teacher can use it as a tool for exploration of a number of questions e.g. the va r i a t i o n of sunrise and sunset through the seasons, what happens at the equator or the nort h pole over the ye a r.The default latitude and longitude as well as images used for the s c e n e ry can be changed by teachers to mimic the local area.The primary school i n t e rface can be viewed and explored online 7 .A screenshot is provided for illustration (Fi g u re 3). to all variables has been provided.From this interface, all aspects of the simulation can be explored.No tasks are set or guidance provided in the interface, it is anticipated that these would be provided externally, either by the teacher or in s u r rounding web-based material.The exploratory interface can be viewed and e x p l o red online8 .A screenshot is provided for illustration (Fi g u re 4).

Fi g u re 4: Ex p l o ra t o ry In t e rf a c e
Mo re details of the solar geometry model are ava i l a b l e 9 .Customising Si m u l a t i o n s re q u i red to suit the two sponsors.Localisation issues such as differing conve n t i o n s for chemical naming and units we re easily dealt with using the tools.The generic n a t u re of the model means that a wide range of chemical reactions can be simulated and the model re-used for other experiments.The whole chemistry re s o u rc e d e veloped for SQA and UCLES is ava i l a b l e 1 3 , with documentation.A re p re s e n t a t i ve practical experiment interface can be viewed and explored online 1 4 along with a completed analysis 1 5 .A screenshot is provided for illustration (Fi g u re 6).Journal of In t e r a c t i ve Media in Education, 2004 (15) Journal of In t e r a c t i ve Media in Education, 2004 (15) Page 17

. 1 . 3 Entrepreneurship
The model used in this example was designed to assist those starting a small business in the production of a business plan.A single model covers cash flow pro j e c t i o n s , b re a k e ven analysis and trading forecasts.The same model is utilised to provide all the i n t e r a c t i ve content in the business start up course.The visualisations begin with a simplified view to allow the learner to explore the principles.They then take the learner through to a realistic case study of a cash flow and trading forecast which can be explored either through a number of pre-defined "what if " scenarios or completely o p e n l y.Additional case studies that show the typical difficulties in setting up businesses of other types can be produced from the same model.The course can be v i ewed and individual simulations explored online1 6 .A screenshot is provided for illustration (Fi g u re 7).

Fi g u re 7: En t re p reneurship In t e rf a c e
These examples do not re p resent the definitive answers to teaching a part i c u l a r subject, they are illustrative suggestions for teaching a subject.Any one of them can be tailored to suit individual teaching pre f e re n c e s .

. The Future
It has been shown that the teacher can have considerable control over ways in which they use simulations in their teaching.Howe ve r, further re s e a rch and development is re q u i re d .The system is being actively deve l o p e d : Putting Teachers in the Loop: Tools for Creating and Thomas & Milligan (2004) Customising Si m u l a t i o n s • Prototype systems exist demonstrating simple model building and collabor a t i ve simulation use; • A system is being developed to allow use of simulations within computer aided assessment systems.

. 1 Learner Support
It is one thing to provide support and guidance in anticipation of the learners needs, it is quite another thing to provide it contextualised and on demand.An ultimate aim must be to duplicate the role of teacher or other expert for the remote learner (Thomas and Milligan (2003)).Fu rther re s e a rch is re q u i red into mechanisms for p roviding context sensitive feedback, support, scaffolding and contingent tutoring as the learner uses a simulation.
Another way of providing support to the isolated learner is to facilitate communication with other learners and with experts.In terms of simulations, this doesn't just mean providing text based chat or email facilities to allow learners to ask questions of an expert, but collaborative access to the simulation to allow teachers to understand the context of a student's question or even to take control to demonstrate a point.Such functionality would be valuable to students as well as expert s .An early p rototype of a simple collaborative system exists.

. 2 Learner Control
De velopments that allow learners greater control over their learning are also desirable.Providing learners with a choice of visualisation styles and objects to suit their personal needs and abilities would improve learner's choice and autonomy and could even be tied to a learner profile.Techniques that allow learners to annotate and s a ve simulation states or save a learning history (Pa rush, Hamm & Shtub (2002)) are of benefit.Mechanisms for re c o rding and replaying activities in a simulation for post task analysis or to learn from others are also desirable.IMS specifications curre n t l y in development explore the possibility of standardising the functionality offered by v i rtual learning environments to facilitate such support.Wo rk is ongoing in this are a and a "Use Case" submitted to IMS illustrates a possible solution 1 7 .

Fi g u re 2 :
Calculation In t e rf a c e Putting Teachers in the Loop: Tools for Creating and Thomas & Milligan (2004) Customising Si m u l a t i o n s 7 Solar Ge o m e t ry: Visualisation for the nov i c e .h t t p : / / w w w. s c ro l l a .h w. a c .u k / t a l k s / s h o c k 0 3 / s o l a r / s o l a rn ov i c e .h t m l

Fi g u re 3 :
Pr i m a ry School In t e rf a c eExploratory interface: The third example has been devised for more adva n c e d learners who can cope with graphs.He re an exploratory interface that provides access Putting Teachers in the Loop: Tools for Creating andThomas & Milligan (2004)   Customising Si m u l a t i o n s

Fi g u re
6: Practical Experiment In t e rf a c e 1 3 S QA-UCLES Chemistry Kinetics Project.http://www.s c ro l l a .h w. a c .u k / t a l k s / s h o c k 0 3 / c h e m we b / 1 4 Chemical Kinetics labora t o ry experiment.h t t p : / / w w w. s c ro l l a .h w. a c .u k / t a l k s / s h o c k 0 3 / c h e m we b / s q a -u c l e s / p p a 2 / s q a p p a 2 .h t m l 1 5 Completed Chemical Kinetics experiment.h t t p : / / w w w. s c ro l l a .h w. a c .u k / t a l k s / s h o c k 0 3 / c h e m we b / s q a -u c l e s / p p a 2 / g e n p p a 2 d o n e .h t m l 1 7 CETIS Use Case for IMS In t e ra c t i ve Content SIG.CETIS-005: Managing and An n o t a t i n g Simulation States within LMS, Milligan, C., Cu r r i e r, S. and Cross, R. h t t p : / / w w w.imsglobal.org/usecases/ic/cetisusecase005_v4.pdf