Tang, Y., & Jahan, K., & Bielefeldt, T. (2015, June), The Effectiveness of an Adaptive Serious Game for Digital Logic Design  Paper presented at 2015 ASEE Annual Conference & Exposition, Seattle, Washington. 10.18260/p.24861

\bibitem{asee_peer_24861}
  Tang, Y., \& Jahan, K., \& Bielefeldt, T. (2015, June), \emph{The Effectiveness of an Adaptive Serious Game for Digital Logic Design}
  Paper presented at 2015 ASEE Annual Conference \& Exposition, Seattle, Washington.
  10.18260/p.24861

Ying Tang, Kauser Jahan, and Talbot Bielefeldt.  "The Effectiveness of an Adaptive Serious Game for Digital Logic Design".  2015 ASEE Annual Conference & Exposition, Seattle, Washington, 2015, June.  ASEE Conferences, 2015.  https://peer.asee.org/24861 Internet.  24 Apr, 2024

\bibitem{asee_peer_24861}
  Ying Tang, Kauser Jahan, and Talbot Bielefeldt.
  "The Effectiveness of an Adaptive Serious Game for Digital Logic Design".
  \emph{2015 ASEE Annual Conference \& Exposition, Seattle, Washington, 2015, June}.
  ASEE Conferences, 2015.
  https://peer.asee.org/24861 Internet.  24 Apr, 2024

@INPROCEEDINGS{asee_peer_24861
author = "Ying Tang, Kauser Jahan, and Talbot Bielefeldt"
title = "The Effectiveness of an Adaptive Serious Game for Digital Logic Design"
booktitle = "2015 ASEE Annual Conference \& Exposition"
year = "2015"
month = "June"
address = "Seattle, Washington"
publisher = "ASEE Conferences"
note = {https://peer.asee.org/24861}
number = {10.18260/p.24861}
}

TY  - CPAPER
AB  -                             The Effectiveness of An Adaptive Serious GameABSTRACTMost students benefit more deeply from guided learning than discovery learning. Even so, not all students are alike.Some of them are highly self-motivated, and they benefit most from the approach that gives them the freedom andopportunities to discovering a realm of knowledge on their own. Others prefer coaching and a more structuredapproach. For students who lack motivation and prior knowledge in particular, practice alone in problem solving withpoor strategies, unproductive ideas and very little guidance will just get them nowhere. Instead, they will end upfrustrated and give up on the task. Nonetheless, instructors are often faced with limited resources to provide “just-in-time” instructional support that is tailored to individual and small group needs.This need of “just-in-time” support becomes even prominent in Gridlock, a domain-specific game that engagesstudents to apply their digital logic design skill for a solutions that fixes the malfunctioning traffic light controller fora 4-way intersection. The Gridlock was piloted in fall 2011 and fall 2012. In fall 2011, A pre- and post- conceptualsurvey was given to 18 students in the experimental group who use the game system and to 21 students in the controlgroup who use the traditional labs to rate their proficiency on 19 core concepts of Digital I (scores ranged from zero tofive (unfamiliar to proficient)). An independent sample t-test revealed no difference between experimental and controlgroups at pre-test (p=0.774) but a significant difference between these groups at post-test (p=0.028). The resultsdemonstrated that students in the game learning had greater proficiency with the course concepts. However, as far asthe learning support is concerned, students voiced their different views of what the game system should provide.Some felt that the current roadmap was just right to provide necessary assistance in identifying domain knowledge. Ina survey given to 20 students in the experimental group about their game behaviors, 10 of them chose road map as themost helpful tool in navigating the game labs, since road map, in their words, “gives an idea of what information touse when I am lost”, “hints necessary concepts”, and “provides formulas to carry out whatever tasks I must”. Othersconsidered the expert guidance is too shallow, and should be more detailed with additional question prompts closelytied to each milestone. Without a direct purpose for and required responses from students to the interventions,particularly KWS, students were not focused on the learning modeled in such support as deeply as we had hoped. Outof the 20 students, 15 chose KWS as the least useful tool since “I don’t usually track my progress in such a way as theone (KWS) presented in the games”, “I like to write the problems down in my notebook”, and “typing about circuits(in KWS) is too difficult”. The similar responses were received from the fall 2012 implementation. Those responsesclearly ratify that every pupil has their own needs and their own particular way of learning. If there were a way inwhich our game system can understand such differences and provide support accordingly, the resulting system couldbe more efficient and effective in promoting successful learning.In the past year, we have been exploring solution feasibility to this critically important challenge, and improveGridlock with an adaptive learning engine that assess what really happens when a student’s capacity is sabotaged inproblem solving and to provide the help that is tailored to his/her needs. However, the data that is necessary for themapping of the student’s knowledge level to differentiated coaching is often uncertain and only available throughobservation of the learning process itself. Therefore, any attempt towards developing an accurate mapping solution,which is part of our work focus, must involve some algorithmic components that will allow the decision makingprocess to (1) accumulate its past experience to a pertinently defined set of data structures, and at the same time, (2)exploit the “knowledge” captured in the data set towards improving the overall system performance. The idea isimplemented using a k-nearest neighbor (KNN)-based close-loop control as depicted in Fig. 1. The proposed systemincorporates classification and feedback into the existing Gridlock game. Each player is evaluated at individual gamestages on their understanding of the material. The results of the evaluation as well as other factors that representhis/her behavior and understanding are classified to determine if the player masters the required material beforeproceeding to the next section of the game. Such classification is cascaded from one game stage to another,reinforcing the system understanding of student domain knowledge. If a student is found lacking in some area of theinformation, he/she is presented immediate feedback and detailed help as for how to acquire the knowledge.                                                                                                  KNN-based VR Game System                                                                                                            Instruction                                                                            Expert Module                                                                                                             Database                                                                             Expert Model                                                                                                           Pedagogical                                                                              Knowledge                      Model                                    Past “knowledge”                           Database                                                       Student Model                Observed evidence                                                                                                        Pedagogy Module                                                        Other action                                                         measure                                                                                                                KNN Classifier                                                       Prompt-based                                 Differentiated instructions                                                                             Gridlock                                                           KWS                            (i.e., a set of progressive prompts or cues)                                                           TA2S                                                                                              Roadmap                                                        Student Actions   General guideline                                                       Student Module                                                                                                  Existing VR Game System                                       Fig . 1: The system architecture of the KNN-based adaptive serious game systemTogether with the implementation, a thorough evaluation plan is conducted as well to help investigatorsanswer three important questions as listed below. This paper reports the findings of this assessment. To what extent is the adaptive features in the VR games useful to student learning To what extent does the VR games with adaptive features play in fostering student interests in engineering  problem-solving To what extent is the student learning improved by the VR game experience in general
AU  - Ying Tang
AU  - Kauser Jahan
AU  - Talbot Bielefeldt
CY  - Seattle, Washington
DA  - 2015/06/14
PB  - ASEE Conferences
TI  - The Effectiveness of an Adaptive Serious Game for Digital Logic Design
UR  - https://peer.asee.org/24861
DO  - 10.18260/p.24861
ER  -