Dynamic and Adaptive Playout of Competency-Based Learning Games Based on Data in Learners’ Competency Profile Considering Didactical Structural Templates

Abstract

In this chapter, a competency-based approach, based on the qualifications-based learning model (QBLM), will be presented, which makes it possible to store the acquired competencies and qualifications (CQ) in a so-called CQ-profile (CQP) while playing an educational game. The game and the course are based on a didactical structural template (DST), making it possible to switch between the course and the game. Before the educational game can be played, it must be checked whether a learner has all the required CQs for the course or learning unit in which the educational game is included. Therefore, this chapter shows an approach to how the CQPs can be implemented, and the data from the CQP can be used to check course prerequisites. Finally, a prototypical implementation with its evaluation and the remaining challenges will be presented in this chapter.

1. Introduction

A central content focus of the module AF A, “Industrial and Organizational Psychology,” in the bachelor’s degree program in Psychology at the University of Hagen (FUH) [1] is industrial psychology [2], which deals with the effect of work on the working person. The critical teaching of theoretical basics of psychological work design, which is mainly done by reading and discussing relevant theories and research results, is unfortunately mostly lacking in the experience of actual work design during studies. This can only be achieved by experiencing and trying out different forms of work design. However, confrontation, one’s own experience and trying out, as well as intensive reflection on what is experienced is an essential prerequisite for the acquisition of action competencies [3], as they are also demanded within the framework of the recommendations of the German Psychological Society for the design of psychology studies [4, 5]. According to [6], the main tasks of work psychology consist of analysis, evaluation, and design of work activities and systems according to defined human criteria. Accordingly, theories and models are taught in the study of work psychology that explain and predict the effect of specific characteristics of work (characteristics of work content, work processes, or social interactions, [7]) on people, their work performance, their motivation, and their health (e.g., action regulation theory, job demand-control model, JDR model, effort-reward imbalance, cf. Lehrbrief Modul AF A Grundlagen und Arbeitspsychologie: p.66, p.126f, p.132ff.). The topic has gained relevance due to an increased social focus on psychological stress at work, which has also been reflected in the consideration of the subject in the Occupational Health and Safety Act. A growing field of work for (industrial) psychologists has emerged. Psychologically relevant task features (e.g., time pressure, work interruptions, information overload, social support, feedback, and task variability) can be systematically manipulated from the outside. After the processing, the feedback of the results, the own reflection of individual and condition causes for specific results, and the debriefing with a systematic analysis of the work situation and the independent derivation of solution suggestions for a better work design occurs. The planned didactic innovation’s primary learning objective is to acquire competencies in occupational psychology to analyze, evaluate, and design work tasks according to defined human criteria [6]. In addition, going through the simulation task and the subsequent reflection should lead to a deeper and better understanding of the differentiation between situational and behavioral prevention, which is central in occupational science and condition-related and person-related interventions [8]. Through minor adjustments, other learning objectives can also be focused on (e.g., employees’ leadership, communication organization, and information flow). Methodological competencies are also developed through a systematic work analysis, which the students must carry out following a work task they have experienced themselves. The development of digital technologies in the form of so-called serious/applied gaming (SG/AG) [9] allows the use of computer-based simulations to enable experiences in the completion of work tasks quasi-virtually, which are typically only possible in actual practical activities. These experiences are at least like those in real life and allow the reflection of unexpected or surprising results [10].

According to the results from [10], the learners’ follow-up/simulated training success is to be captured with a final quiz and then measured using learning analytics (LA) [10] and map it to training outcomes and corresponding qualifications in terms of factual knowledge (competencies) and action knowledge (skills and proficiency levels) on the topic area of job design. The factual knowledge and action knowledge refer to declarative and procedural knowledge definitions of the adaptive control of thought-rational (ACT-R) theory [11]. Action knowledge refers to the procedural knowledge that indicates how something should be executed. This is the knowledge about the appropriate execution of actions [12]. Factual knowledge refers to the declarative knowledge of the ACT-R theory [13]. To accomplish a task or problem, an interplay of both bits of knowledge is needed [14].

The PAGEL project has the aim to provide an AG within the so-called knowledge management ecosystem portal (KM-EP) [15] in combination with the learning management system (LMS) Moodle [16]. As outlined in a previous paper [10], currently, the qualifications-based learning model (QBLM) [17, 18] supports the assignment of competencies and qualifications (CQ) [18], which are the achievements of the learning objectives and learning successes of the game/simulation sequences within an integrated applied game (AG) or any other learning unit. The learning management system (LMS) used at the FUH is Moodle [16] and offers digital learning content at the FUH. Therefore, the already existing LMS will be used as a basis in this work [19]. A didactical structural template (DST) [20, 21] supporting QBLM can be used as a starting point to describe the underlying process and provide the measurement criteria for the success of achieving learning objectives regarding CQs. This includes the success of training skills on different proficiency levels in a game-based simulation. DSTs represent the didactical structure of a course.

Furthermore, it can support a hybrid environment existing of a “classical” course in combination with applied gaming content, like a pedagogical structure for AG. For example, the AG can be a web-based computer game or a VR/AR-based game; therefore, one DST can have different modalities [20, 21].

Within the framework of the PAGEL project, there is a “classical” course with study books [10] for teaching theoretical content and factual knowledge. After successfully completing the “classical” course, the knowledge of action is to be imparted with the help of an AG [10]. “The DST, with its possibilities, offers the option to provide the learner with different media types for the same learning content. The same contents can be given for different platforms or in different modalities. The DST can provide the information cross-media. Some courses are more useful in a texture-based form, others as a game or video. The most obvious dimension in choosing a media type is the content. For example, it does make sense to provide learners in a driving school with a game or simulator to teach them driving, more than giving them a text with technical instructions. On the other side, a list of traffic symbols would also be meaningful. A mixture seems to be a good solution. However, other dimensions are also conceivable. These additional dimensions depend on the situation and can vary based on some parameters. Learners have different preferences and strategies for learning. Some learners have already acquired knowledge or CQs, making the course partly a repetition. That should lead to an abbreviated form of that content” [22]. Subsequently, the measured results from the AG are to be explained in a final quiz using the previously imparted theory knowledge [10]. These results are to be evaluated accordingly concerning the achievement of a CQ using LA. The results are stored in a competence and qualifications profile (CQP) [23]. Whether parts of the course have to be attended or not has to be checked in advance. Here, it should be checked whether the learners have already stored the corresponding CQs in their CQP or not [24].

The motivation of the book chapter is to enable the dynamic playout of the different course contents of the PAGEL project based on DSTs and to attest to corresponding CQs in a CQP [23] based on the learning outcomes and check in advance whether the learners have to attend this course at all.

Several problem statements (PS) can be derived from the objectives and motivation mentioned above.

Problem statement 1 (PS1) is that from today’s perspective, for the PAGEL project, the DST is based on the current course structure [10], and to be added, AG does not exist. PS2 is that it is impossible today to store and read out the acquired CQs in a CQP to play out courses and learning units adaptively. PS3 is that from today’s point of view, it is not possible to check, based on the CQs in a CQP, whether learners already have all the prior knowledge to complete a course or whether they first need to acquire further CQs. PS4 is that currently it is impossible to track the progress inside a DST implementation. Progress tracking across DST implementations is required to allow learners to switch the modality inside the learning unit. The possibility of switching the modality while executing an ACT and continuing at the same position in the learning progress inside another DST implementation is desirable for learners. This enables the learners to switch according to their current needs.

The PSs mentioned above result in the following research questions (RQs): RQ1: “What does a DST for the PAGEL project need to look like?”, RQ2: “What components and interfaces are needed to adaptively play out a course based on the data in a CQP?”, RQ3: “How can the progress tracking in a DST based on the CQs of a learner be tracked and verified?”, and RQ4: “How can the progress across DST be tracked?”

Based on the research methodology of [25], the following research objectives (ROs) were derived from the RQs. RO1 is assigned to the observation phase (OP). This phase identifies suitable interoperability standards for interfaces for the CQs exchange. Also, suitable systems and tools are identified. RO2 is assigned to the theory-building phase (TBP). A concept is designed that shows what system components and interfaces are needed. The system development phase (SDP) moves the concept into a prototype and is assigned to RO3. The result of the SDP is evaluated in the evaluation phase (EP) in the context of a cognitive walkthrough (CW) [26]. Finally, the EP is assigned to RO4. In this phase, all RQs are evaluated.

The remainder of this paper is structured according to the ROs. This means that in state-of-the-art section, the OP is described. In conceptual design section, the TBP is described, and the SDP is presented in this paper in proof-of-concept implementation section. In evaluation section, the EP is presented. Finally, the paper concludes with a summary and indications of future developments.

2. State of the art and technology

The previous section has already mentioned some research projects and software systems related to the research goals. In the following, the most important are described in more detail.

2.1 Qualifications-based learning model

To increase the comparability of learning objects based on qualifications, the QBL-approach described by [17, 18] was implemented at the FUH [1]. The QBLM has evolved out of the CQ-based learning approach [18]. The QBLM approach includes a domain model (QDM), an architectural model, and several service distribution models. The QDM introduces the qualifications-relevant learning element (QRLE). A QRLE instance can de facto stand for a personal development plan (e.g., a study program), a unit of learning, a learning activity, or a knowledge resource. In QBL, CQ-based learning goals (LGs) and access requirements (ARs) are represented by CQPs consisting of CQ instances (the actual CQs). CQPs are applied to specify QRLE-related LGs and ARs and describe students’ personal LG (target profiles) and the current state of attested CQs (actual profiles). After successfully completing a QRLE, the students’ actual profiles are updated. The gap between the actual and the target profile must be bridged by appropriate CQ programs, including suitable QRLEs such as courses, learning activities, and knowledge resources [18]. In addition, KM-EP [15] has been developed within the project Realizing an Applied Gaming Ecosystem (RAGE) [17]. In addition, the portal offers various web-based tools for knowledge management and user-friendly course authoring tools (CATs) to create Moodle courses without much previous knowledge [17].

The KM-EP enables QBLM-based work with CQs. For this purpose, a CQ administration enables creating CQPs and assigning LUs (modules, courses, and teaching resources within courses) to these profiles. Furthermore, courses with assigned CQs can be exported to Moodle and executed [17]. Moodle stands for modular object-oriented dynamic learning environment [16] and is a freely available open-source software under the GNU Public License. It is software with which courses can be conducted and developed via the internet.

With the learning tools interoperability (LTI) standard, further applications such as games can be integrated into an LMS like Moodle [16]. To transfer QBLM-based CQs to Moodle, [18] developed the plugin QBL4Moodle. This plugin is the interface between Moodle and KM-EP. QBL4Moodle is used to work with QBLM in Moodle itself and map CQs created with it to the CQy approach of Moodle itself. The plugin also imports QBLM-based CQs, profiles, and frameworks from other systems. Currently, this is realized for the KM-EP [18] only.

The LMS used at FUH is Moodle. This LMS already offers digital learning content at FUH. Therefore, the already existing LMS will be used in this work. Within the KM-EP, it is possible to define CQs as conditional- and goal profiles (CP and GP) for courses [22]. QBL4Moodle also allows the transfer of CPs and GPs for courses to Moodle and directly connects them with the courses [18]. This feature has been extended to make it possible in the KM-EP to define CPs and GPs on every level or element below the course level [27]. In addition, the QBL4Moodle-Plugin and Moodle have been extended in the same way that it is possible to define CPs and GPs on every course element [28].

To support, for example, adaptive learning or the dynamic playout for courses, [23] extended Moodle in the way that explicit CQPs exist for learners. Until then, learners did not have a specific CQP. When learners wanted to know which CQy profile they had, they had to gather every single CQ spread over different Moodle pages. With the implementation of [23], learners can go to a specific page in Moodle to see which CQs they have. At the same time, the CQPs in Moodle have been designed and implemented; [24] designed and implemented a possibility for Moodle to support a dynamic outplay for courses. Dynamic outplays mean that if course A has qualification A’, course B needs qualification B’ and course C needs qualification C’, and the student has already acquired qualification A’ and the student wishes to start course C, then course B must be offered to the student. To realize the dynamic outplay of courses in Moodle, [24] analyzed Moodle’s functionalities and how it would be possible to extend Moodle so that CQs can be supported for the dynamic outplay. Due to the fact that [23, 24] have been designed and implemented at the same time, the CQP did not exist in Moodle, so [24] had to assume that a CQP exists to build his design accordingly. Therefore, the remaining challenge is to bring both parts together so that the result of [24] uses the implemented CQPs [29].

2.2 Didactical structural templates

The so-called didactical structural templates (DSTs) have been introduced in [30] and extended in [20]. As described, the DSTs are based on the IMS learning design (IMS-LD) [31] and represent the didactical structure of a course and cannot only be used as a didactical structure for creating courses. The DSTs can also be used as a didactical structure for a hybrid environment existing of a “classical” course with integrated applied gaming content, just like a pedagogical structure for the applied game, which can be a web-based computer game. Therefore, one DST can have different implementations. The advantage of this approach is that learners will be able to switch between different implementations of one DST whenever they want to. They have got the same learning progress as if they had used only one specific implementation of this DST. This means that if learners like gaming, they can use the applied gaming implementation to work on the learning content. For example, suppose it is easier for the learners to answer the self-test or the final test—to stay in the exemplary stated pedagogical structure of a course—for example, multiple-choice quizzes. In that case, they can switch to a course within an LMS to answer the questions. Therefore, the DSTs have the following hierarchical structure:

Method: There are many different ways a person can learn or teach. Each learning method is a sequence of learning processes.

Play: It is a key part of the learning design, representing a teaching-learning process. Like a theatrical play with a sequence of acts, when an act is completed, the next act begins until the completion condition is met.

Act: An act represents a series of simultaneous activities and activity structures.

Activity: It is one of the core elements of learning design, which relates to many learning environments.

Activity structure: Activities can be combined into an activity structure with sequence mechanisms or freely selectable structuring. To access the DSTs, we have provided a RESTful API described in [32]. With the DSTs, we can provide the didactical structure of learning content in a standard, conform way.

2.3 E-learning interoperability standards

To fulfill the RO for the OP, corresponding interoperability standards were consulted and examined regarding their suitability for the given use case. “In general, the purpose of e-learning interoperability standards is to provide standardized data structures and communications protocols for e-learning objects and cross-system workflows. When these standards are incorporated into vendor products, users of e-learning can purchase content and system components from multiple vendors, based on their quality and appropriateness, with confidence that they will work together effectively” [33]. According to [33], the interoperability standards can be divided into the following five categories:

1. Metadata

2. Content Packaging

3. Learner Profile

4. Learner Registration

5. Content Communication

The “metadata” type [33] is used in the e-learning context to ensure that data is stored, indexed, searched, and retrieved appropriately [33]. For this purpose, corresponding standardized exchange formats have been developed for platform-independent exchange by different organizations [33]. The type “Content packaging” [33] is used to exchange course data between different learning systems [33]. Specification examples according to [33] are IMS Content Packaging specification, IMS Simple Sequencing specification [33], Sharable Content Object Reference Model (SCORM) [34] or Aviation Industry CBT Committee (AICC) guidelines [35]. The “Learner profile standards [33] allow different system components to share information about learners across multiple system components. Learner profile information can include personal data, learning plans, learning history, accessibility requirements, certifications and degrees, knowledge assessments (skills/CQs), and the status of participation in current learning” [33]. According to [33], examples of specifications for the Learner profile standards are IMS Learner Information Package (LIP) [36] or the Personal and Private Information (PAPI) [37] specification. The Learner Registration standard is intended to enable learning management components to know what learning content to deliver to each learner [33]. Examples of this are IMS Enterprise and Schools Interoperability Framework [33]. The Content Communication Standard allows information about various activities to be communicated to individual pieces of learning content [33]. Activities can include but are not limited to the start of the learning content or the completion of a learning content [33]. The completion status ranges from test results to course grades or achieved CQs [33]. Currently, two initiatives create standardized communication protocols and data models for content communication [33]. One is the Computer Managed Instruction (CMI) [38] standard of the AICC [35], and the other is the SCORM standard [39]. An even newer standard in the context of content communication is the Experience Application Programming Interface (xAPI) standard [34, 40]. In addition to the standard scope of SCORM functionalities, the xAPI standard also offers further functionalities such as tracking data in serious games or simulations [34].

To export CQs from the personal CQP, currently, only an interoperability standard of the type “Learner Profile” [33] can be considered. This allows information about the learners to be shared, such as personal data and data about degrees or assessments of knowledge [33]. According to [33], the exchange formats LIP [36] and PAPI learners [37] can be considered. IMS LIP was chosen because it also supports the transfer of CQs. One standard published by IMS is the learner information package specification (LIP) [36]. The purpose of the specification is to define a set of packages that can be used to import data into an IMS LIP-compliant system and extract the data from it [41]. In addition to a textual description of the standard, an extensible markup language (XML) [41, 42] schema is also provided. The LIP is based on a data model that describes the characteristics of learners [43]. A LIP consists of 11 main information pieces considered the basis for the learner information data structures [43]. The learner information [43] element acts as the container [43]. Each of the 11 main structures can occur any number of times within the learner information structure [44]. The content information, the so-called content type [43] element, refers to the data described with the LIP [43]. Each content type consists of a referential [43], temporal [43], and privacy [43]. The 11 main structures can contain content information and sub-content in addition to their content. The content information here allows a specific reference to the data described there, for example, for a CQ , an ID with the help of which one can find the CQ in a table [43]. With the sub-content, recursive substructures can be defined [43].

To import CQs into a personal CQP, the interoperability standard of the type “Content Communication” is used since results of, for example, learning games or a quiz can be transmitted with this. As shown above, the CMI guidelines [38] and xAPI are available from the AICC. Since the CMI [38] is a standard that can only communicate with systems that have also implemented the CMI standard, xAPI was chosen. The experience API (xAPI) is a free specification that enables learning technologies to collect data about a person’s diverse experiences (online and offline) [40]. The collected data are managed consistently about a person or group [40]. Disparate systems can thus communicate securely by capturing and sharing this stream of activities using xAPI’s simple vocabulary [40]. xAPI statements describe an experience (usually a learning experience, but it can be any other activity or state of affairs) that a person has performed [45]. The general structure of a statement is described in English as “I did this“ or formalized as “[actor] [verb] [object]” [45]. An xAPI statement is represented as a JavaScript Object Notation (JSON) document [46]. It contains at least three components, namely actor, verb, and object [46]. An actor is an individual or a group whose activity is recorded with the statement [46]. A special feature is that no ID has to be used for internal identification (in the sense of a record within a database table) [46]. Instead, other data or methods can be used for unique identification, such as e-mail addresses. The verb is the action performed by the actor within the activity [46].

Verbs are represented as Internationalized Resource Identifiers (IRI) and can optionally be extended with a short textual description in different languages. IRIs are an evolution of the Uniform Resource Identifier (URI), using Unicode characters instead of American Standard Code for Information Interchange (ASCII) characters [47]. There are no restrictions on the definition of verbs. It only has to be a valid IRI [46]. However, this circumstance makes it possible for a verb to be defined multiple times with different URIs. To prevent this, there is a central registry for normalizing common verbs [48]. An object represents the experience, activity, or other states of affairs that an actor has performed [46]. A type property can be used to specify what type it is. By default, an activity is assumed. Depending on the type, the further structure of the object is determined. As with verbs, there is also a registration point for normalization [49, 50].

3. Concept

This section of the chapter represents the results of the ROs for the TBP. First, the overall system architecture will be represented. Afterward, the concept for the CQPs and the adaptive playout will be described. Finally, in the last subsection, the concept for the usage of DSTs will be described in more detail. Except for subsection 3.2, which is placed in Moodle, all other components described in the other subsections are within the KM-EP.

3.1 DST-based system architecture with established features

This section of the chapter will present the concept for the overall system using the CQP and DSTs.

The KM-EP holds the modules for the following provided features. As a base for the course, the DST is used. The implementation of a DST is named a didactical application (DA). The DA is registered in the KM-EP as the didactical application configuration (DAC). The DAC contains the reference to the DST identifier, a title, a Uniform Resource Identifier (URI)—the implementation is reachable with—and a Universally Unique Identifier (UUID). The UUID is used to link the external system with the DAC. An extension of the KM-EP is implemented to provide the possibility to add and edit DACs, allowing the simultaneous existence of different DAs for one DST.

The system suggests to the user that data rank the registered DAs out of the related DACs. For the recommendation, the system uses several classification parameters to describe learners and the DA. For the DA, the DAC holds the parameters. The parameters are different devices the media type is introduced, which offers a generic categorization. The DAC contains the flags for accessibility (“readout available,” “audio required,” and “plain language”), information about the possible screen resolutions, an extended reality flag, and a flag to describe the need for user interaction (input device) and the level of interactivity. The concept for the level of interactivity is derived from [51]. It describes six stages of possible levels.

Learners are described by multiple parameters, which describe the learners’ preferences, limitations, and devices. The visual aural read kinesthetic (VARK) profile [52] is used to adjust the preferences. Fleming and Mills introduced the model in 1992 [52]. The model categorizes the learners into four types: visual, aural, read, and kinesthetic. Via a test, learners can identify four dimensions points in four dimensions and push them into a radar diagram. The radar shows afterward the preference meaning a higher swing to one of the four dimensions. The limitations of learners are reflected by three flags, which are “visual impairment,” “hearing impairment,” and “plain language requested.” The device and its attributes are automatically derived from the browsers’ information and contain the resolution, XR device, and interaction possible (input device) [52].

In the DA recommendation, these two classifications are confronted with each other. This is done via a transposition of the DA classification as a point into a two-dimensional graph. This graph contains the centric point of the VARK profile. The distance between both points defines the rating for a recommendation. If the points are close, the learners’ preferences coincide with the DA. The learners’ limitations overrule the recommendation process to offer the learners only the DAs they can consume. For example, the DA recommendation offers blind persons (“visual impairment”) only DAs with the flag “readout available” true. After rating the different DAs, they are listed in a UI in descending order.

DA recommender and learner adaptive flow enrich this system with additional features for tracking the learners’ progress and a CQ-based learning path for every individual learner.

According to the learners’ presets according to their VARK profile and their impairments, the registered DAs, and the media type, the so-called didactical application recommender (DAR) calculates the best-fitting DAs. This calculation shows all eligible DAs in descending ranking order in a table so that the learners can choose on their own the DA they want to continue with learning. For more information on how the ranking is calculated, see [22, 53].

As shown in Figure 1 on the right side, the architecture of Moodle consists of six components [15]. The core, the subsystems, the plugins, the plugin types, the sub-plugins, and the dependencies. In Figure 1, the three components, namely core, plugin, and subsystems, are represented [50]. The Moodle concept is explained in more detail in [50]. The PAGEL learning resources are part of the core of Moodle. The core contains all the basic functions of Moodle [15].

Figure 1.

Overall system architecture.

The course authoring tool (CAT) on the right side offers a user-friendly CAT to create Moodle courses without much previous knowledge [17].

On the left side in Figure 1 are placed all components within the KM-EP, which are used in the context of this paper and described in further subsections.

3.2 Concept for the adaptive playout of CQs with CQPs

As shown in the use cases in [50], a CQP is a profile that is assigned to an individual learner. CQPs describe each student’s personal learning goals (LG) (target profile) and the current state of attested CQs (actual profile) [50].

As shown in Figure 2, the LMS Moodle is required. The architecture of Moodle consists of six components [15], which are the core, the subsystems, the plugins, the plugin types, the sub-plugins, and the dependencies. In Figure 2, the three components, namely core, plugin, and subsystems, are represented [50]. The Moodle concept is explained in more detail in [50].

Figure 2.

System concept for the adaptive playout of CQs with CQPs [29].

The functionality of a subsystem can be partially enabled or disabled by configuration. As an example, it serves the package CQs, which contain the functionality of the CQs. The plugin component [54] contains optional components that extend the basic functionality of Moodle. Plugins such as QBL4Moodle are optional components. The QBL4Moodle plugin contains a bidirectional communication relationship realized via core hacks [50]. To check the course prerequisites, the corresponding interfaces, described in Section 2, query the respective CQs of an individual learner from the CQP [29, 50]. To combine the dynamic outplay of courses in [24] with the learners’ CQP [23], the implementation of [23, 55] has to extend the way that the just described interface is called when the CQ-checks are executed. Additionally, it has to be ensured that the checks and the interface work with the same structure of the CQs—either the one Moodle provides or the one QBL4Moodle provides [29, 50].

To export CQs from the CQP, a representational state transfer (REST) [56] interface needs to be provided. A web service called “QBL personal qualification profiles” must be added to the plugin QBL4Moodle. The web service is named “local_qbl_datastructures_personal_ profiles,” according to Moodle naming conventions for web services. This provides the functions that a client can call, such as an educational game [23, 50]. To export the CQP, an IMS LIP document containing the CQs is generated from a personal CQP. As a container of all competence/qualifications instances (CQIs) [17, 18] in the CQP, it summarizes them. To import a CQ to a CQP, an xAPI statement must contain all the necessary information to generate a CQI. To identify a single learner, a personal e-mail address is used prerequisite; the address is not used in multiple user accounts. The following structure of the IRI [47] is defined for the object: “qblm://CQI/<CQF Unique identifier>>/<CQS Unique identifier>>/<PL>>.” By specifying the unique identifiers for CQF and CQS and a PL, a CQI can be generated and entered in a personal CQP [23, 50].

3.3 Concept for the didactical structural template for the PAGEL project

To be able to use the QBLM and DST software infrastructure inside an actual course at FUH for the PAGEL project, first of all, a DST has to be conceptionally created. Therefore, the course and game flow described in [10] will be used as a basis.

Currently, the course “Industrial and Organizational Psychology” consists of three study books, which give the learners an overview of the theoretical background of industrial and organizational psychology. In addition, an AG should be developed to provide the learners with hands-on experience and action knowledge in the subdomain work task design, as described previously and in [10]. In the AG, the learners should experience the psychological effects of work task design for themselves. To check the learned factual and action CQs afterward, it should be possible that the game results are explained and analyzed using the previously acquired theoretical knowledge. In the end, the learners should receive a detailed overview of their results, including a review [10].

AS described previously, a DST has a hierarchical structure with different hierarchy types. At the top level of the hierarchy for a DST is the DST name. In the given case, this should be the name of the project PAGEL for easier identification. On the level below is the play level. A play element represents a learning process. When one play element is finished, the next one starts. In the PAGEL project, three learning processes could be identified. These would be the preparation, the play, and the course completion. Finally, each play element has the respective act elements on the level below. An act describes activities or activity structures that can run parallel in a play. An activity element is a core element of the learning design. Activity structure elements can be used to combine several activities.

For the PAGEL project and the identified play elements, the structure of the DST results is shown below.

The preparation includes only the theoretical background. This was thought of as an act element because this element and its activities must be completed before the play and evaluation can take place. The activities here are the three study booklets, which the learners can work on in any order.

The play element contains the “work task simulation.” In the PLC, it should be possible for the player to carry out various activities in parallel with so-called tools [10]. The tools are each represented as a separate activity.

After the play element preparation and game are completed, the play element course completion can start. The course completion is divided into three ACTs: final quiz, evaluation, and results, each of which has its own activity (Figure 3).

Figure 3.

Concept for the DST for the PAGEL project.

3.4 Concept for learner adaptive flow based on didactical structural templates

The learner adaptive flow is introduced to support a learning journey, adapting itself according to the preset of already gained CQs. The learner adaptive flow controller controls the learner adaptive flow. It compares the target CQs of a learning objective (act, activity structure, or activity) with the learners’ CQs within the KM-EP. A learning section skip is possible if the learners’ CQs contain all target CQs. This skip is offered to the user via a dialog inside the start DA process. The dialog pops up if a learning section skip is possible and lists all possible skips. As the experts for their skills and their need for repetition, the learners select then the desired skips. A learning journey skip is reflected in the system as a completed activity. This is applied in the DSTS and loaded by the started DA afterward [22, 53].

3.5 Concept for didactical progress tracking

As described earlier, the DAs existence is independent of the KM-EP and the DST. That results in the circumstance that the execution of a DA is detached, and the KM-EP or other DAs do not receive any update about the progress of learners. With the didactical progress tracker (DPT), this gap is closed. The DPT holds for every learner’s journey a didactical structural template session (DSTS). The DSTS maintains information about the successful completion of activities inside the journey. All activities of the learners’ journey DST have an entry inside the DSTS, which is initially “not completed.” While executing a DST-based learning journey in a DA, the DSTS gets updates from the DAs after completing an activity. This feature allows tracking in a central place as well as the sharing of the progress through DAs.

This sharing results in the possibility of switching between different DAs. When a DA is started, it requests the DSTS from the KM-EP to get the learner’s progress. It then has to define the already completed activities as finished inside itself.

To support the switch of a DA inside a journey, a control panel allows the learners to call a different DA at any time. The DA is called with a link, which contains the reference to the DSTS to allow the called DA to load the DSTS, respectively, the already gained progress.

A management dashboard shows all DSTS, including the learners and their progress. In addition, it allows root users to maintain their progress and support the learners in completing their journeys [22, 53].

4. Proof-of-concept realization

This section of the chapter represents the results of the ROs for the SDP. First, the proof-of-concept (PoC) realization for the adaptive playout of CQs with CQPs will be presented, and afterward, the realization of the DSTs components.

4.1 Realization of the adaptive playout of CQs with CQPs

This section will describe the realization of the combination of both concepts of [23, 24]. Because Moodle is the target platform for the described extensions, the implementation will be done with PHP [57] in combination with JavaScript [58], respectively, jQuery [59]. For the prototypical implementation in [24], Moodle had to be extended to different surface components, which are presented with different code snippets and sequence diagrams to show how the prototypical implementation will work. The prototypical implementation in [23] is presented with different code snippets and screenshots to display the layout of the extended pages in Moodle. The conceptual part of this paper shows how the two separate works [50, 55] can be combined. The prototypical implementation of this part is open and will be presented in a separate paper [29].

4.2 Realization of the didactical structural template for the PAGEL project

Based on the concept for the DST, the DSTM was opened in the KM-EP. The corresponding structures from the concept were created in the DSTM. The finished DST is shown in Figure 4.

Figure 4.

Realized DST for the PAGEL project.

4.3 Realization of the learner adaptive flow based on didactical structural templates

For ACTs and elements below that hierarchy, possible learning section skips are provided to offer an adaptive learner flow. The learners’ CP is compared with the goal CQs of the DST learning elements. If the learners’ CP contains the goal CQs, the element is a candidate for a skip. The possible learning section skips are provided to the learners, and they decide whether the learning section skip is executed or not. The skip is reflected as a preset of the learning elements’ ACTIVITIES to “completed” in the DPT [22, 53]. A detailed overview and description of the implementation can be found in [22, 53].

4.4 Realization of the didactical progress tracking

The introduced DPT tracks the progress of the learners based on ACTIVITIES. The DAs have to send a progress update to the KM-EP after finalizing an ACTIVITY. From inside the running DA or with a control panel, the learners can switch to another DA. The learners only lose their progress within an ACTIVITY [22, 53]. A detailed overview and description of the implementation can be found in [22, 53].

5. Evaluation

In the form of a cognitive walkthrough (CW) [26], an initial evaluation of the proof-of-concept implementation was accomplished by domain experts in the field of education in computer science. The evaluation’s primary goal is to estimate the productive capacity of the implementation and orientate future development. The developed DST for the PAGEL project for RQ1 was presented and accepted by all project participants during a CW. The CWs for RQ2, RQ3, and RQ4 have been documented in [22, 23, 24, 29, 50, 53, 55]. The occurred errors and misbehaviors have been defined and fixed accordingly. Based on the evaluation, improvements and renewals were identified. These will be implemented and considered in future work.

6. Conclusion and future work

We have demonstrated in this chapter how it is possible to extend Moodle with CQPs, and a dynamic outplays of courses. In the past, both parts were not combined; we have shown how a combination of both concepts is possible through a conceptual model. The remaining challenge is the prototypical implementation of the combination of the CQPs, the dynamic outplays of courses, and an evaluation of this combination.

The DST is independent of the DAs. Multiple DAs implement the same DST received by the XML export, the endpoint, or the CAT. Afterward, the author can register the DA as a DAC in the KM-EP. In addition, the DAC contains characteristics about the DA as well as a link to redirect learners to the DA.

Before starting into a DA to execute the learning journey, learners have to enter their preferences into a UI. The preferences are the restriction flags (visual impairment, hearing impairment, and plain language requested), the learners’ VARK profile values, and the media type of the learners’ device. These are compared with the characteristics of the available DACs related to the requested DST. The DA recommendation provides a rating for the DACs. The learners can then select from the sorted list a DA to start the journey.

The introduced DPT tracks the progress of the learners based on activities. The DAs have to send a progress update to the KM-EP after finalizing an activity. From inside the running DA or with a control panel, the learners can switch to another DA. The learners only lose their progress within an activity.

For acts and elements below that hierarchy, possible learning section skips are provided to offer a learner adaptive flow. The learners’ CP is compared with the goal CQs of the DST learning elements. If the learners’ CP contains the goal CQs, the element is a candidate for a skip. The possible learning section skips are provided to the learners, and they decide whether the learning section skip is executed or not. The skip is reflected as a preset of the learning elements’ activities to be “completed” in the DPT.

For the PAGEL project, we have shown a DST concept and the DST definition in the DSTM.

The learners’ CQP, used for the adaptive playout, is placed in Moodle and does not interact with the KM-EP. On the other hand, the learner adaptive flow is based on the learners’ CQP placed in the KM-EP. To make the learner adaptive flow possible within the KM-EP, the learners’ CQP has to be synchronized with the learners’ CQP placed in Moodle (and other systems when other systems are involved in the process).