Human-Computer Interaction

Immersion through Storytelling - An Eldritch Horror Board Game App

This project is already completed.


In recent time, more and more board games are supplemented with companion apps giving these games multimodal interaction Alchemists, new game features One Night Ultimate Werwolves, or a way to easily feature otherwise unhandy parts of the game XCom. They also improve gameplay aspects like hidden information Rising 5, story telling Mansion of Madness, or player management Leaders.

The board game Eldritch Horror comprises complex gameplay aspects that have to potential to be improved my means of a companion app. Similar games, like Mansion of Madness, already use such an integration to successfully deliver its story. In a standard game of Eldritch Horror, Game Immersion Experience (Cheng, M. T. et al., 2015) is mostly delivered via random card events. Moreover, keeping track of all the different stats, possibilities, and actions can be overwhelming.

This HCI project focuses on the improvement of Eldritch Horror with respect to game immersion experience by supporting the various player actions and encounters in it using a companion app.


Eldritch Horror is a cooperative board game focusing on storytelling and saving the world. It is the successor of Arkham Horror, however it does not take place in Arkham anymore but on the whole world. Players become investigators and must travel different places, fight monsters, close gates and solve mysteries. They do all of this under a time clock and must fulfill their goals before the final boss (The Ancient One) takes over the world. Main mechanics in this game include a lot of (story) cards and dice rolls which determine tests’ success or failure.

Related Work

Commercial Games

During the last decade, digital helpers started to supplement board games in various ways. Different aspects of games have already been improved using apps: XCom uses the app to create a game that runs in real time, while Rising 5 or Alchemists use technology to create logic puzzles the need of an extra player doing that. Other games, like Werwords or One Night Ultimate Werwolves, remove the need of a “game master” so everyone can play. Next, apps are already used for storytelling and random events in Mansion of Madness. On top of that, games for kids like Scotland Yard Master (AR), Ravensburger smartPlay or Ravensburger tiptoi integrate technology in order to explain games interactively and provide multimodal feedback. Finally, also the “heavier” side of games like Leaders or Golem Arcana use apps in which a lot of calculation or help is done for the players.


The HCI research group used the XRoads project to investigate multimodal interaction, storytelling and artificial intelligence in a mixed reality board game with fantasy setting. Their project showed positive ratings when using mixed-reality approaches in board games (Anke Giebler-Schubert et al., 2013).

Andrea Bartel (2017) wrote about “Game Experience und Spatial Presence in Tabletop Rollenspielen” and found that digital supported versions of games can improve immersion but sound did not add anything to that.

Finally, Cheng, M. T. et al. (2015) have taken a look at “game immersion experience as a precondition to flow” and developed the “Game Immersion Questionnaire”.

Taken all that into account, it is promising to add mixed-reality elements or digital support to boardgames which than can then be evaluated using the Game Immersion Questionnaire.


This HCI project will create a (mobil) app that should support the base game, making it more immersive and easier to use. Depending on the results of game mechanic analysis, it could also add new game elements in order to create a richer experience.

All in all, the user experience should improve if the app is used, leading to the general hypothesis that the app supported version receives higher user experience ratings than the base game.



A game mechanics analysis will be conducted. It comprises the extraction of main mechanics from the game’s rules. The extraction focuses on all the different cards and possible actions in the game. First results are shown in the next chapter.


The app will be created using the game engine Unity. This allows for multi platform builds. There is no plan to include other frameworks or libraries. Altogether, the code and system architecture should especially consider extensibility to ease future research and a potential implementation of game features not covered by this project.

###Evaluation A final study will be conducted after the implementation. Since a normal game of Eldritch Horror takes up to four hours, a special, shorter version like one or two rounds must be used. The study will compare the original version with the augmented one and will focus on the app’s and gameplay’s ease of use and the story’s immersion. It will focus on the Game Immersion Experience (Cheng, M. T. et al., 2015). A between design is favored because of number of participants and to allow them to compare the version directly.

First Results (Game Mechanics Analysis)


The game is split in rounds which contain three different stages. They are shortly described here. The letters code the implementation possibilities that fit into this stage.

  1. Action phase
    • Players take two actions like moving, resting, acquiring cards (CE3) or activating abilities.
    • This might include rolling dice (DC) or flipping cards (CE4, RC)
  2. Encounter phase
    • Each player has an encounter (CE1, OS), fights monsters or tries to solve mysteries by collecting clues (UE, OS).
    • All these actions mostly require dice rolls (DC)
  3. Mythos phase
    • One mythos card is resolved (CE2). This might also chain other global effects like flipping cards (RC).

Tutorial mode (TM) would create a special round and guide players through those three phases.

Six different implementation possibilities are described here. However, implementing all of them are out of the scope of this project. Each options also contains the “contextual knowledge” which indicates how much the app needs to keep track of the game’s state (and therefore how difficult it will be to program).

Card replacement (CE)

Encounter, Expedition, Clue and Research Cards (CE1)

All these cards contain story text and a test (resolved with dice) for an investigators. The app will have the texts recorded to improve immersion into the story. Therefore, the real cards a not needed any more.

Contextual knowledge: None

Mythos Cards (CE2)

These cards contain story as well as global effects. The story will be recorded, too. The app will then step the players through the global effects which avoids accidentally overlooking one of them. Therefore, the real cards a not needed any more.

Contextual knowledge: None

Support and Artifact Cards (CE3)

These cards mostly help the investigators by giving them more dice, chances to re-roll them or more. Replacing them allows the app to control the random reserve (which is “a shop” players can buy cards) and is mandatory for the dice calculator (DC).

Contextual knowledge: Medium (Which player has which cards)

Spells and Conditions (CE4)

These cards have two sides which offers the possibilities for random cards (RC) when replacing them.

Contextual knowledge: Low (Which player has which spell / condition)

Dice calculator (DC)

Most of the tests are resolved with dice. However, there are a lot of cards that give or take dice which results in mental effort for players to get the number right. The app could always calculate the amount correctly.

Contextual knowledge: High (Needs to know all cards and effects in any possession)

Random Cards (RC)

Some cards have two sides and the game forces the players from time to time to turn them around. However, once turned, there is no mystery any more. The app can therefore bring up different “back sides” each time which will keep up the feeling of uncertainty and lead to more immersion.

Contextual knowledge: None

Unknown Enemy (UE)

Within the story of the game, the investigators must solve mysteries to figure out what is going on. However, the players themselves exactly know it since the final boss is drawn at the beginning. In order to create a more immersive experience, the app can choose the boss in secret and add a new game feature: The players themselves have to figure out whom they are facing by collection clues.

Contextual knowledge: None

Ongoing Story (OS)

Even though the game is focused around story, the single encounters are not linked to each other. An app integration could keep in memory what already happened to the investigators and bring that up during other events (e.g. mentioning an injury again).

Contextual knowledge: Medium (What already happened to which player)

Tutorial Mode (TM)

This is a completely different idea: While the game offers a lot of story and possibilities, it is also complex. An app might offer a tutorial mode and step the players through the actions one by one. Such an interactive way of learning the game cannot be done without digital help.

Conclusion and Outline

This exposé shows different options for the HCI project. Since implementing all those ideas is out of scope, the focus in this work is on story (CE1, CE2) and immersion (RE, UE, OS). It might be noted, that not all of the story cards must be recorded since there are a lot (122 Encounters, 51 Mythos and 16 Mysteries) and for proving the concept (and the study), this is not required.

The following diagram shows the research plan for the project.


Bartl, A. (2017). Game Experience und Spatial Presence in Tabletop Rollenspielen: Vergleich eines analogen Spiels mit dessen Mixed-Reality Umsetzung

Giebler-Schubert, A., Zimmerer, C., Wedler, T., Fischbach, M., & Latoschik, M. E. (2013). Ein digitales Tabletop-Rollenspiel für Mixed-Reality-Interaktionstechniken. Virtuelle und Erweiterte Realität, 10, 181-184.

Cheng, M. T., She, H. C., & Annetta, L. A. (2015). Game immersion experience: its hierarchical structure and impact on game‐based science learning. Journal of Computer Assisted Learning, 31(3), 232-253.

Contact Persons at the University Würzburg

Dr. Martin Fischbach (Primary Contact Person)
Mensch-Computer-Interaktion, Universität Würzburg

Legal Information