https://upload.wikimedia.org/wikipedia/commons/e/e6/Flickr_-_Official_U.S._Navy_Imagery_-_Doctors_perform_surgery_together..jpg
MASTER THESIS

  • MHF1 - USING EYE-MOVEMENT DATA TO DETECT LEARNERS’ PYCHOLOGICAL STATE(S)

    SUPERVISORS: PROF. DR. WILLEM VERWEY, SHARANYA LA, MSC

    35EC

    Several studies have used eye tracking measures such as gaze (for instance, looking at or away from the learning environment) and blink rate as indicators of attention to identify states of disengagement or mind-wandering behaviour (when thoughts are diverted away from the task). Such measures have also been used to make inferences about learner’s state of confusion (for instance detection of repeated eye movements around the same area of the screen) and confidence at varying levels of accuracy. This is very useful since it is common knowledge that learners who pay less attention to the study material show lower learning outcomes. While eye tracking is usually done with sophisticated hardware, there is growing interest in being able to collect reliable eye-tracking data through a computer’s webcam (despite its limitations), so that such measures may be used more widely. In our ongoing project, we aim to create a frugal design that uses multiple kinds of physiological data-eye movements (from a webcam based sensor) being one. The prerequisites to this are a) finding a web-cam based eye-tracking software that fits the needs of the project, b) conducting a (tech) feasibility study with the tool, c) doing usability studies to understand in what conditions the tool can be used (eg. room lighting, distance from screen etc.) and b) validating the efficacy and reliability of the eye-tracking software in determining learner attention/mind-wandering - this could be done in relation to learner-outcomes and in comparison to a sophisticated device such as the Tobii eye tracker.

    In summary, this is a multi-tiered project at BMS and the direction you take will be based on the results of the initial tests you conduct. We invite 1 MSc student with excellent statistical skills to join this project.

    Hutt, S., Mills, C., Bosch, N., Krasich, K., Brockmole, J., & Mello, S. D. ’. (2017). Out of the Fr-"Eye"-ing Pan Towards Gaze-Based Models of Attention during Learning with Technology in the Classroom. https://doi.org/10.1145/3079628.3079669

    Xu, L., Wang, N., & Lewis Johnson, W. (2005). LNAI 3538 - Using Learner Focus of Attention to Detect Learner Motivation Factors. In LNAI (Vol. 3538). 

    Smallwood, J., McSpadden, M., & Schooler, J. W. (2008). When attention matters: The curious incident of the wandering mind. Memory and Cognition36(6), 1144–1150. https://doi.org/10.3758/MC.36.6.1144

  • MHF2 - EARLY-STAGE ASSESSMENT OF INFOTAINMENT INTERFACE FOR LUXURY CARS (IN COLLABORATION WITH A MARKET LEADER)

    SUPERVISORS: DR. SIMONE BORSCI

    35EC

    Type of thesis:

    35 EC. Please express interest to Dr. Borsci by email (s.borsci@utwente.nl) as soon as possible (deadline before the end of February) so you can have a preliminary chat with him about the project. Only one candidate will be selected for this work.

    Non-disclosure agreement (NDA)

    You will be asked to sign an NDA form as your work will be in collaboration with an industrial partner.

    Introduction

    Infotainment systems (collecting both “information” and “entertainment”-related contents) are the main point of contact between users and car functionalities. This is where all core electronic functions, like stereo, navigation, HVAC, etc., are controlled. Infotainment functionalities are getting more complex over time; also, they are distributed over multiple in-vehicle displays and can be accessed through a variety of control interfaces (touch, in-air gestures, speech…).

    In collaboration with a market leader, you will work on a new methodological approach to define and implement key indicators for infotainment UX/UI intuitiveness and effectiveness; such methodology will be used both for benchmarking existing systems, and to drive the early design of new systems.

    Overall goal

    Identify aspects that capture automotive infotainment UX/UI complexity and find a way to measure them, by defining relevant metrics and possible equipment to perform the measurement. Define an overall methodology and procedure for assessment of infotainment concepts and prototype review at early stages

    Deliverables and key tasks

    • Systematic literature review (Using Prisma Methodology).
    • Definition of methods and principles used in literature for assessment of Car Infotainment
      interfaces, and possible ways to objectively measure relevant key indicators
    • Focus groups and Delphi methodology to define a procedure for UX/UI assessment
    • Scenario testing of the final list of principles and metrics with participants

    Previous knowledge

    You should have all the necessary knowledge to perform the tasks. Nevertheless, familiarity with the systematic review of literature and knowledge about focus group and Delphi methodology, as well as good capabilities of management of applied projects and systematic writing are considered relevant expertise to achieve the goal of this work.

  • MHF3 - EXPLORATION OF NOVEL APPROACHES TO CONVEY 3D EFFECTS IN AUTOMOTIVE USER INTERFACES (IN COLLABORATION WITH A MARKET LEADER)

    SUPERVISORS: DR. SIMONE BORSCI

    35EC

    Type of thesis:

    35 EC. Please express interest to Dr. Borsci by email (s.borsci@utwente.nl) as soon as possible (deadline before the end of February) so you can have a preliminary chat with him about the project. Only one candidate will be selected for this work.

    Non-disclosure agreement (NDA)

    You will be asked to sign an NDA form as your work will be in collaboration with an industrial partner.

    Introduction

    Digital instrument clusters – which show vehicle gauges and information on a display - have almost completely replaced analog instrumentation in car dashboards. On top of that, the increased attention to safety will lead to the integration, over the next few years, of camera-based systems for driver monitoring.

    In collaboration with a market leader, you will explore the possibility to use real-time head pose information, provided from driver monitoring systems, to convey the illusion of 3D depth on instrument cluster user interface.

    Overall goal

    Study the cognitive aspects that induce the human brain to perceive a flat image as a tridimensional object, by exploiting a combination of parallax, head pose tracking, virtual lighting and image framing. Implement an experimental setup to test the effectiveness of such approach.

    Deliverables and key tasks

    • Systematic literature review
    • Definition of key cognitive aspects involved in conveying depth illusion on bidimensional surfaces
    • Define potential solutions to convey tridimensional images
    • Implementation of a software prototype on a PC environment (with the support of the company)
    • Perform scenario-based testing to assess the effectiveness of the solution(s) and user preferences 

    Previous knowledge

    Knowledge of Python and/or C/C++ programming languages are required to implement the system prototype on a PC environment. Good capabilities of management of applied projects and systematic writing are considered relevant expertise to achieve the goal of this work.

  • MHF4 - ARTIFICIAL INTELLIGENCE CONVERSATIONAL AGENTS: A MEASURE OF SATISFACTION IN USE

    SUPERVISORS: DR. SIMONE BORSCI

    35EC

    Background

    Conversational agents, such as chatbots and voice interfaces, can be used for multiple purposes e.g., support customer experience with services etc. These new tools are growing and more and more integrated into systems such as websites, social networks, cars. Smart and AI-based conversational agents are shaping the future of human-computer interaction however little is known about how to assess people reaction and satisfaction after the use of these systems.

    Goals

    Advance previous work done on a new scale to assess satisfaction with chatbots. Your experimental work will focus on the evaluation of conversational agents to further streamline the reliability and validity of the scale.

    Your work will consist of testing with a remote usability test different chatbots with a set of tools, including the new scale to perform a confirmatory factorial analysis. You should be aware of statistical methods regarding factorial analysis and be able to use R.

    This assignment is suited for one master student.

    Key references

    • Coperich, K., Cudney, E., & Nembhard, H. Continuous Improvement Study of Chatbot Technologies using a Human Factors Methodology.
    • Duijst, D. (2017). Can we Improve the User Experience of Chatbots with Personalisation? MSc Information Studie, Amsterdam. 
    • Følstad, A., & Brandtzæg, P. B. (2017). Chatbots and the new world of HCI. interactions, 24(4), 38-42.
    • Hill, J., Ford, W. R., & Farreras, I. G. (2015). Real conversations with artificial intelligence: A comparison between human–human online conversations and human–chatbot conversations. Computers in Human Behavior, 49, 245-250.
  • MHF5 - LEARNING TO DRIVE WITH ONLINE DRIVING LESSONS

    SUPERVISORS: DR. MARTIN SCHMETTOW (CPE), JORRIT KUIPERS (GREEN DINO, CPE)

    In this project you will work with Green Dino, a Dutch pioneer in simulated driver training. Green Dino has developed a web-based driving simulator, which has been employed on broader scale during the first lockdown. This project is a Data Science project where you will examine the learning effects of online driving lessons, based on existing data.

  • MHF6 – DRIVER TRAINING FOR PROFESSIONAL CHAUFFEURS

    SUPERVISORS: DR. MARTIN SCHMETTOW (CPE), JORRIT KUIPERS (GREEN DINO, CPE)

    Green Dino, a Dutch pioneer in simulated driver training, has developed a simulator-based training for professional bus drivers. In your research project, you will evaluate the effectiveness of the training, as well as User Experience.

  • MHF7 – IS VIDEO-STIMULATED RECALL AN EFFECTIVE METHOD OF TASK ANALYSIS?

    SUPERVISOR: PROF. DR. JAN MAARTEN SCHRAAGEN

    25/35EC

    Thinking aloud while performing a task is not always possible in real life, for instance when there is time pressure, when there are colleagues around that should not be disturbed, or when safety is at stake and thinking aloud could result in loss of attention or concentration. Previous research has compared concurrent and retrospective thinking aloud, for instance in the context of usability testing (Van den Haak, De Jong, & Schellens, 2003). In this research, the focus was on problem detection in usability research, rather than on strategies. From a different research tradition, video-stimulated recall has been extensively used in teacher training (Gazdag, Navy, & Szivak, 2019) and in medical research (Paskins, McHugh, & Hassell, 2014). Yet another research tradition has employed ‘video-reflexive ethnography’ to optimize health professionals’ work practices (e.g., Carroll & Mesman, 2018). Although this method has generally been viewed positively in terms of yielding many new insights and resulting in positive learning outcomes, it is an intrusive method that may yield ‘a second-order reconstituted account’, that is, a distorted view of the processes actually carried out during task performance. The issue here is whether “recall” or “retrospective thinking aloud” is being confused with a “stimulated-recall interview” or a “probed interview”. Frequently, researchers are unclear in their precise instructions to participants.

    What complicates matters is that many different factors play a role here: novelty of the task (e.g., insight problems versus routine problem solving; Schooler, Ohlsson, & Brooks, 1993), task load, type of task (e.g., design, writing, speaking, interacting with machines or humans), whether retrospection takes place immediately after task performance or after some delay, epistemic self-efficacy in oneself as an evaluator of knowledge (Trevors, Feyzi-Behnagh, Azevedoz, & Bouchet, 2016),  the types of cues used during retrospection (e.g., audio only; video cued, gaze plot, gaze video, verbal cues; Olsen, Smolentzov, & Strandvall, 2010), whether the professionals themselves comment upon their own task performance or whether this is done by colleagues very familiar with the task (Erlandsson & Jansson, 2007). It is of course also important to be explicit about the goals to be achieved with retrospective methods: are they used for learning purposes (to improve performance), for scientific purposes (to gain additional insight in cognitive processes or strategies of information use), or for practical purposes (to obtain more usability problems or to verify or expand upon existing task analyses)?

    The purpose of this master thesis assignment is, first, to systematize the research on concurrent and retrospective thinking aloud by developing an explanatory framework than can shed light on the sometimes disparate results. Second, an experiment will be carried out to specifically focus on the added value of video-stimulated recall for task analysis purposes. The various factors mentioned above will have to be taken into account in this experiment.