Magnification has become the standard approach to tackle accuracy problems in eye-tracking based interaction systems, but suffers many problems itself. Magnification can be inefficient and oftentimes occludes contents in direct proximity to the magnifier, which is especially problematic when dealing with inaccuracies. While some of these problems can be tackled with special magnifying shapes, like hybrid linear-fisheye lenses, most approaches are rather limited one-size-fits-all approaches agnostic to user characteristics and content semantics. This limits the ability of magnification approaches to help users in scenarios where they encounter a below average accuracy or interact with targets that are especially small or crammed together. This paper presents Adaptive Magnification: a novel approach that is personalized to user-specific parameters and also context-aware by utilizing and mapping the screen content's semantic information to choose a magnification factor and size to optimally support interaction. The approach was evaluated against the use of no magnification and the use of hybrid linearfisheye magnification in a user study ($\mathrm{n}=20$). Task effectivity was evaluated by measuring the misclick count, while task efficiency wasn't measured as the approach is not focused on speed increase but on functionality and usability improvements. The user study revealed a 95.7% lower misclick count when compared to the use of no magnification and 13.1% fewer misclicks compared to hybrid linear-fisheye magnification.

With factors like the COVID-19 pandemic, digitalization and climate change, hybrid work and education models with in-person and remote participation have gained popularity. According to existing studies, the hybrid format comes with a set of challenges in comparison to pure online or in-person lectures. Using a small setup with products from Sennheiser electronic SE & Co. KG as an example, the influence of hybrid lectures on the student experience was measured. The Kruskal-Wallis test was used to evaluate the influence of the format, while the difference between online and in person participation was evaluated using t- and U- tests. In addition, qualitative data was collected in the form of interviews, observations, and brief feedback integrated into the surveys. The data showed that to avoid a disadvantage of the hybrid format, careful preparations with a focus on sound quality of the lecturer is necessary. The experience of online and in-person participants only differed in comprehensibility of the used technology.

During the COVID-19 pandemic, learning had to be shifted from in-person settings to entirely online or hybrid teaching scenarios. Schools and universities in Germany were forced to quickly deploy solutions which often provided lackluster quality. This shift was also implemented indiscriminately across subjects so that screen intensive courses (such as programming) were impacted in the same way as practical courses like hardware engineering. In many cases, this left a negative impression of such hybrid scenarios on participants. However, with steady improvements in technology, hybrid meetings have established themselves in corporate environments and steadily gained popularity in parts of higher education programs. Especially in the case of dual programs, this is also applicable to lectures and seminars. This paper explores using products from Sennheiser electronic SE & Co. KG with microphone arrays and audio-over-IP technology in combination with network cameras as an example, this concept aims to create an environment for seamlessly integrated hybrid lectures with about 30 participants in presence.

The precision of modern commercial off-the-shelf eye trackers has reached a level sufficient for developing gaze-based applications. In many but not all applications, accuracy even allows for replacing a computer mouse with gaze-bazed pointing. The Multi-Modal Interaction Concept for Efficient input (M2ice) tackles accuracy problems with an on-demand hybrid fisheye magnifier. This paper introduces an image-analysis-based approach that identifies areas on the screen where to automatically activate magnification for improved interaction. It combines the separate actions magnifying and clicking into one seamless action. The approach works by combining OpenCV filters for detection of clickable elements on the screen with a local machine learning algorithm predicting whether magnification is needed based on size and position of screen elements. A user study (n = 28) showed a significant speed increase of 18.50 percent (t(27)=-3.95, p=.0002 at α = .05) with automatic magnification compared to separate shortcuts for clicking and magnification.

Spaced repetition is a learning task that helps users repeat learning items at optimally computed intervals. It is best suited for small learning units such as learning cards. In a typical learning scenario these learning cards do not stand by themselves, rather they are the last step in the students’ work process, summarizing the students’ interaction with the learning material. In this paper a framework is developed that allows learning material curators and students to create spaced repetition learning units with back-references to multimedia learning content, i.e., learning videos. With the advent of video learning nuggets, a wealth of content was created that can now be used in spaced repetition learning. The approach also allows for linking to specific parts of those learning videos to enable students to quickly review learning content specific to the learning unit currently worked on.

Eye tracking based interfaces have to solve two major problems: The midas-touch problem and accuracy. While the midas-touch problem can be tackled with innovative interaction concepts, accuracy problems result from human physiology, hardware limitations and increasing screen resolutions. The Multi Modal Interaction Concept for Efficient Input (M2ice) tries to tackle these problems with a hybrid linear-fisheye magnifier. This magnifier works in the center of the screen but is problematic on the borders/corners of the screen, as it only appears as a half or quarter circle, limiting the space of the enlarged area. This paper presents a novel teardrop magnification approach for border and corner cases that solves these problems by using a different shape and by shifting the user's focus, making more of the magnified area available for interaction.

Evaluating the success of learning technologies with respect to improvement in the learners' abilities and knowledge is not an easy task. The problem is formed by the existence of many different definitions with different perspectives like grades on the one hand and workplace performance on the other. This paper reviews definitions from the literature with the aim to find a suitable definition for the evaluation of learning success in spaced repetition based mobile learning for knowledge improvement. It also borrows approaches from learning analytics to tackle the fact that learner groups are heterogeneous which leads to the need of analyzing learning success differently in different groups of learners.

This paper presents an approach on how to retrieve relevant data from a huge set of game-data in order to find evidence for the impact of using spaced-repetition algorithms on the learning success in a mobile learning game. After having collected approximately 12 million sets of playing data, the database needs to be preprocessed before analyzing it in order to filter out any data that is irrelevant or useless for our analysis or may even dilute its results. One structured and established way to do this is to follow the KDD process, which includes several consecutive steps of consolidating the available data, with preprocessing it being one of them. In order to be able to define the data were are looking for, we set up some proposals about how the relevant data should look like and how to retrieve it from our database.

Gaze-based interaction approaches like the Multi Modal Interaction Concept for Efficient Input (M2ice) minimize the need for using pointing devices like a computer mouse in human computer interaction. While most core functionalities of pointing devices can be addressed in a more or less straightforward manner, secondary functions like scrolling remain to be a challenge. This demo paper presents a gaze-based extension for the M2ice interaction concept that is realized by dividing the screen into distinct areas and combining scrolling areas with a reading area in the middle. The approach is compared to other gaze-based solutions.

This paper presents results from an analysis of player behavior in the popular mobile learning game "Where is that". Playing data of nearly 24,000 unique users were gathered over a period of three months and subsequently analyzed in order to get a better insight in how games are played. The results will then further be used to compare learning results with a spaced repetition approach. Our analysis revealed four distinct clusters of learner types that can be categorized as Learners, Confirmers, Leisure Players and Sporadic Players. The data shows the player types' playing patterns and gives indications about what motivates them to play. It can thus give valuable hints for the design of player interaction in learning games as well as content selection.

Eye tracking based interfaces have to solve two major problems: The Midas-touch problem and accuracy. Midas-touch describes the fact that many interfaces use fixation as an interaction signal. Hence, orientation and interaction can hardly be kept apart. While the Midas-touch problem can be tackled with innovative interaction concepts, accuracy problems result from both human physiology and increasing screen resolutions. The Multi Modal Interaction Concept for Efficient Input (M2ice) approach tries to tackle these problems by providing a magnifier. This magnifier can have different shapes and magnification methods that all come with various advantages and disadvantages. This paper discusses the advantages and disadvantages of a number of state-of-the-art magnification approaches in the light of eye tracking based interfaces and presents a new implementation of the M2ice prototype implementation that employs a hybrid fisheye magnification.

Most modern keyboard-based computer systems cannot be operated without pointing devices such as mice, trackballs or touch interfaces. They hence have two input modes: keyboard and pointing. A pointing device's purpose is to convey the user's focus of attention to the computer in order to connect actions like clicking or moving the mousepointer to specific areas of the display such as buttons or input fields. This demo paper introduces a multi modal approach (keyboard and eyetracking) that retains all functionalities of a pointing device while eliminating the need for an actual device. This goal is achieved by combining eyetracking with keyboard interaction. Initial informal user studies have suggested that the proposed interaction paradigm makes interaction that combines textual and pointing input (such as most work related tasks) more intuitive as it eliminates the need to switch input devices during interaction. The demo presents a working prototype implementation of the approach.

Unit testing is based on the idea that units under test behave in a reproducible and deterministic way. If the unit's code is dependent on external context factors like time or location, these factors have to be controlled in order to produce meaningful results. Spaced repetition mobile learning games in which users are reminded to play at time intervals are based on previous user interaction with the system. Hence, time as an external factor as well as user input has to be provided by the unit test. This means that a unit test framework has to not only provide input to the unit under test but also manipulate time as a context factor. This paper introduces such a test framework especially designed for mobile applications.

The computer mouse is the main interaction device for graphical user interfaces. Many attempts have been made to replace it or render it obsolete, ranging from more ergonomically shaped designs to pen-like devices or touch screens. While the latter have opened up the way for completely new interaction designs mainly on mobile devices, PC and laptop users still prefer the computer mouse as a pointing device, especially when used in conjunction with a keyboard. This paper provides an analysis on how users interact with different pointing devices and identifies factors for efficient pointing-based human-computer-interaction in standard working scenarios. Based on this analysis a concept for minimally distracting pointing based human computer interaction is introduced. The concept is based on a multimodal combination of gaze tracking and keyboard, enabling users to interact with a keyboard alone in scenarios that usually require a keyboard in conjunction with a pointing device. The paper also presents a prototype and discusses its conceptual and implementation details.

Especially software running on mobile devices does increasingly rely on contextual information such as time and location. And whenever a software product is affected by context, this context has to be replicated for testing and debugging. This paper introduces an external context manipulation interface for a previously developed learning item scheduler. The scheduler determines when to present a learning item in a learning game based on previous interaction in order to maximize learning efficiency and is based on psychological models. As inter-presentation-intervals can be in the range of days to months, system testing cannot be conducted in a conventional manner. Hence, virtual time hops can be used to fast forward to any specific point in virtual time which would make the software act like it was system time. The approach has shown to be a valuable debugging and testing aid and can be extended for other contextual information sources.

Web lectures can be employed in a variety of didactic scenarios ranging from add-on for a live lecture to stand-alone learning content. In all of these scenarios, though less in the stand-alone one, indexing and navigation are crucial for real world usability. As a consequence, many approaches like slide based indexing, transcript based indexing, collaborative manual indexing as well as individual or social indexing based on viewing behavior have been devised. The approach proposed in this paper takes individual indexing based on viewing behavior two steps further in that (a) indexes the recording at production time in the lecture hall and (b) actively analyzes the students attention focus instead of passively recording viewing time as done in conventional footprinting. In order to track student attention during the lecture, recoding and analyzing the student's behaviour in parallel to the lecture as well as synchronizing both data streams is necessary. This paper discusses the architecture required for personalized attention based indexing, possible problems and strategies to tackle them.

Spaced repetition learning is an approach for choosing the most efficient intervals between rehearsing learning content. Typically used for tasks like learning vocabulary it also offers great potential for content selection in learning games. Learning games do, however differ from classic spaced repetition learning approaches in that content is not only accessed when indicated by a spaced repetition scheduling algorithm but also when the users simply want to play the game or when they decide to play the game multiple times in a row. In these cases, short term memory effects might mask learning effects in user performance, leading to faulty inputs to the calculation of spaced repetition interval lengths. This paper reviews current research literature on the interaction of short term and long term memory in order to determine how short term memory effects can be coped with in the context of spaced repetition based learning games.

When programs get big enough to be called software, they need an architecture. An architecture helps organizing a software's structure so that it is easier to maintain and so that parts of it can be re-used in similar circumstances. When spaced repetition based mobile learning is concerned, re-use is crucial. Spaced repetition learning is an approach that schedules presentation of learning content based on psychological models. Paired with learning games, it can be used to control the flow of the game so that learning content is presented in the most efficient manner. In order to achieve this, the game and the spaced repetition model have to act in a highly interleaved manner. However, the spaced repetition model remains the same, regardless of the game at hand. A useful architecture would design interaction between the spaced repetition model and a game so that developers can focus on the game part and use a minimal interface to turn a simple learning game into a spaced repetition learning game. As space repetition relies on calculated presentation intervals, mobile devices are an optimal platform since they allow for learning anytime. However, mobile devices come with unique constraints that have to be considered when conceiving an architecture. This paper discusses possible software architectures for spaced repetition learning games on the iOS platform as well as architecturally relevant details of spaced repetition learning.

Learning requires repetition. Spaced repetition algorithms are aimed at reducing the number of times a learning item has to be accessed by the learner by scheduling item presentation based on psychological models. These models take into account learner performance on previous interactions with the learning item and the rate at which humans forget what they have learned. In recent years, spaced repetition learning software has become popular for simple learning tasks like flash cards used for learning vocabulary. This paper presents a prototype application that extends the spaced repetition learning approach to more complex content like the kind usually found in learning games. One major difference between this content and flash cards is that learning games usually contain a number of different tasks that convey the same underlying concept categories. To complicate matters, one task might even be classified as belonging to a number of independent or orthogonal categories. This paper explores how these categories can be modeled on the basis of a mobile game designed for training in the field of relational databases. We have chosen a mobile approach to leverage it's anytime/anyplace availability which allows a more precise scheduling by the spaced repetition algorithm.

Time-based slider interfaces haven proven to be an intuitive and effective means for navigation in web lectures and other video content. They allow users to easily navigate to arbitrary positions in the video and clearly visualize the video's linear structure. They do, however, lack any contextual information about the video's content at these positions. Earlier approaches have tackled this shortcoming by visualizing slide titles or slide previews on the timeline. The approach presented in this paper goes one step further in that it links the current slider position to a fisheye-style view of the lecture slide overview. The fisheye shows the slide playing at the current position in full detail and even in the corresponding animation step -- if animated. It also shows the neighboring slides' previews and visually maps them to the timeline, thus providing contextual information on the current slider position while still maintaining a global overview.

Most state-of-the-art interfaces for multimedia browsing come with two inherently different navigation components: a time-based slider interface and a structure based overview. This demonstration introduces the Tempo-Structural-Fisheye-Slider (TSF-Slider). The TSF-Slider combines time- and structure-based navigation in one single navigation component. To merge time- and structure based navigation information, a fisheye based approach is used. The idea of fisheye visualization is extended to rescale time-based information so that it can be merged with structural information while maintaining the general advantages of fisheye visualizations: focusing on one area of interest while still maintaining a general overview. The main motivation for TSF-Slider is, however, that it brings together the advantages of time- and structure based navigation. The TSF-Slider is implemented in the context of the virt Presenter web lecture framework.

User awareness has become a popular feature in many social web applications. In classic text-based web-systems, user awareness features show how many users are online in a web application or how many users are accessing the same web page. When time based media like web lectures are concerned this approach comes to its limits since time-based media are inherently different from classic text-based media. The main difference in these two types of media is that text-based media can easily be skimmed at a glance while video- or audio-objects have to be fully replayed when users want to grasp their content. This paper presents an approach that employs time-based usage statistics for all users in a web lecture system in order to provide users with a means to communicate with other users who are online and have watched those parts of a media object that the user is interested in. The work is implemented in the context of the Opencast Matter horn project - an open source based project for producing, managing and distributing academic video content.

Adaptive hypermedia bring a major benefit of traditional face to face teaching to electronic media in that they allow to custom tailor the way in which one media object is delivered to different users. While the creation of adaptive hypermedia is a highly labor intensive process for text based media, creating adaptive multimedia documents poses nearly insurmountable obstacles. This is mainly due to the fact that creating and editing multimedia documents requires considerably more effort then the creation of text based documents. However, recently more and more universities have begun to make video lectures publicly available. Community efforts like the Opencast project aim at standardizing video lecture formats, making re-use of content from different institutions possible. Involving end users can decrease per capita workload by involving learners in media creation. This paper describes a basic technology framework that empowers end users to create adaptive multimedia documents by merging parts from different sources.

At many universities, Web lectures have become an integral part of the e-learning portfolio over the last few years. While many aspects of the technology involved, like automatic recording techniques or innovative interfaces for replay, have evolved at a rapid pace, Web lecturing has been independent of other important developments such as Web 2.0.Web 2.0 is an emerging trend that among other things aims at integrating user generated data from different systems in order to enhance the user experience. This paper discusses the benefits Web lecturing could gain from a Web 2.0 perspective. It also introduces a prototypical interface for the virtPresenter lecture viewer that enriches Web lectures by Web 2.0 features, such as a specialized player that can be integrated in blog or wiki sites and multi-medial links that can be used to reference arbitrary passages of a Web lecture from an external blog, wiki or forum.

Lecture recordings can be a powerful addition to traditional lectures and they can even serve as a main content source in a number of didactic scenarios. If users can quickly locate relevant passages in a recording, the recording combines the ease of search that comes with electronic text based media with the authenticity and wealth of information that is delivered in a live lecture. Locating relevant passages in a time based media such as a recorded lecture is, however, not as easy as searching an electronic text document. This paper advocates for interactive content overviews as an efficient navigation aid and presents an implementation that is based on SVG. It also discusses the feasibility of likewise approaches in other media formats