By Mohamed Musa, Steven Le, Krystal Li, Bessie Zhang

CS 247G — P2-4

Game Link: https://cs247g-scavenger-hunt.vercel.app/

Google Drive Link to Playtest Images & Videos

 

Artist Statement:

 

“Unveiling Souls” is an immersive scavenger hunt designed to challenge and intrigue participants with a narrative-driven exploration of Stanford Arts. Our intention is to create an experience that surpasses the difficulty of traditional scavenger hunts by incorporating unique and original clues that demand keen observation and critical thinking.

 

As players navigate this adventure, they will delve into the hidden corners of Stanford Arts that many students neglect to visit over the four years on campus, uncovering secrets and stories embedded within its spaces. Each clue is crafted to reveal a piece of a larger, captivating story, blending historical intrigue with mythical elements. Participants will assume the role of a historical figure trapped in Hell, tasked with solving puzzles to unveil their past sins and ultimately reclaim their souls.

 

A key narrative game mechanism we employed is branching storylines, where the decisions players make can alter the course of the narrative, leading to multiple immersive art experiences. This adds personal investment in the story as players’ choices directly impact their play journey. Different players are able to take away their version of Stanford Art home post playing the game.

 

Our game aims not only to entertain but also to share interesting stories of different ancient civilizations. Through the journey, players will encounter bits of historical and cultural narratives, enriching their understanding and appreciation of the art and history intertwined within Stanford’s landscape. As Dante wrote in the Inferno, “In the middle of the journey of our life, I found myself within a dark woods where the straight way was lost.” “Unveiling Souls” is more than a game; it is a quest for redemption, a test of wit and resolve, and an invitation to see familiar surroundings through a new, mysterious lens. We hope to inspire curiosity and sparks in our players’ minds through the journey in Hell.

 

Initial Decisions

 

In designing “Unveiling Souls,” our initial decision centered on creating an engaging and immersive experience through strategic game design mechanisms and formal elements. We integrated puzzles, riddles, and interactive challenges requiring keen observation and problem-solving skills, balanced to maintain player engagement and motivation. Narrative structure, role-playing, and a cohesive thematic backdrop further enrich the experience, connecting players with the story and their surroundings.

 

To deepen the experience, we incorporated resource management, where players use gathered clues wisely, and time constraints to add urgency. Hidden object tasks and interactive dialogues with historical characters enhance immersion and narrative flow. By focusing the game within Stanford’s often-overlooked areas, we aimed to spotlight its artistic and historical treasures, providing players with a fresh perspective.

 

We’ve also wanted to ensure that all in-game locations are fully wheelchair accessible, providing an inclusive experience for all players. While “Unveiling Souls” is best enjoyed on-location, we’ve also designed it so that every puzzle can be solved digitally. Whether you’re decoding clues, searching for information, or piecing together the story, the entire adventure can be navigated through a player’s device so that anyone anywhere can experience and play.

 

Our goals for “Unveiling Souls” extend beyond entertainment. We envisioned the game as a conduit for sharing fascinating stories from ancient cultures, enriching participants’ knowledge and understanding. To further enhance the experience, we want to add sensory features such as background music to help immerse players in the game’s atmosphere, making the historical and mythical narratives more vivid and engaging. Each Mechanism such as puzzles and challenges then creates dynamics of emergent behavior, resulting in Aesthetics that evoke curiosity, intrigue, and accomplishment. Ultimately, we aspire for players to emerge from the game both entertained and enlightened, with a newfound appreciation for the art and history surrounding them.

 

Concept Map:

Playtest 1:

 

For our first playtest, we were only able to secure one playtester, but despite this, we were able to conduct an informative playtesting session. The playtest was facilitated through a prototype designed on Google Forms, which enabled us to gather immediate feedback on our narrative, clues, and answer-checking features. The puzzle tested in this playtest was originally designed using a library search code, which would have required participants to engage with an extensive database or the physical library to solve the puzzle. However, since we were not at the location, we altered the puzzle to feature a riddle. This riddle was themed around the Genesis period, which also fit into the narrative structure that we were testing.

 

Overall, the playtest went relatively smoothly with the player engaging with the puzzle through Google Forms. The initial introduction paragraph, which set the stage for the puzzle, was particularly effective. According to the player feedback, this introduction not only captured their interest but also provided a clear thematic grounding that helped them immerse themselves in the task. This effective scene-setting was crucial, as it ensured the player was adequately prepared and motivated to tackle the riddle.

 

Playtest 1 Questions:

• Is the game’s narrative fun?
• Are the clues & how puzzles are phrased understandable?
• Do people respond in expected ways for proper answer checking?

 

Feedback:

• “Hell” Narrative
  • Successes: Firstly, the initial introduction paragraph, which set the stage for the puzzle, was a particularly effective exposition. According to the player feedback, this introduction not only first “grabbed [their] attention” but also helped them to understand why they were playing this character and “why I should care” about getting out of Hell. The implementation of the “Hell” narrative throughout the rest of the game was also successful and our playtester found it enjoyable and that it fit well within the context of the gameplay. Specifically, the player appreciated the “immersive setup” and believed it created a “strong foundation for the game’s narrative.” This solidified that the design of the Google form and the thematic clues were effective at tying together the biblical clue and answer for the first clue. This encouraged us to continue with the game aesthetic and theme and validated that it enhanced the storytelling and overall environment of the game. 
  • Failures: Despite the successes, there were some areas where the narrative could be improved. The primary issue was the narrative’s length, which seemed to initially overwhelm the playtester, leading them to skim through the content rather than engage deeply. It seemed like the narrative’s importance for solving the puzzle was not immediately apparent to the playtester.
• Clues
  • Successes: The clue was considered interesting by the player and understandable to the playtester and they were able to solve it through thinking about each of the words and doing some internet research. 
  • Failures: While the player found the clues and narrative interesting, they expressed a desire for more challenging puzzles, specifically stating “I think the riddle is pretty easy for me, especially since I know the religious figures quite well.” This response indicated to us that we needed to better understand how different player backgrounds could affect their understanding of the game and the puzzle difficulty. Based on this feedback, we will focus on creating more intricate puzzles in future iterations that increase in difficulty as the game progresses while maintaining a cohesive narrative. We will also consider how players coming in with different understandings will interact with our puzzles in different ways. 

 

Going Forward: ​​Overall, we found in this playtest that our narrative was engaging and our clue was easily understood by the player. However, in our next iteration, we will ensure that the narratives are shorter in length so that players only get the information that is necessary to progress through the game and understand the story, without the burden of reading unnecessary prose. Moreover, we plan to employ a gradual increase in difficulty for the puzzles within our scavenger hunt. This strategy will facilitate a smoother initial engagement by allowing players to solve the first “level” with relative ease, which will help to bolster their confidence and commitment to the game. As players advance, the increasing complexity of the puzzles will draw on their sense of achievement and create greater challenges for their problem-solving skills, which addresses the failures of the clues in this iteration.

 

Playtest 2: 

 

In our second playtest, we playtested the first and second clue of our game using the same Google Forms platform as our last iteration, with changes to the narrative and added difficulty for the second clue. This playtest was also conducted not at the location. The first puzzle was similar to the first with a few wording changes to reflect the need for a shorter narrative. The second clue added leveraged the online Cantor collection portal, providing greater player autonomy, particularly for players who are unable to visit the physical location. The clue required the player to navigate through the digital archives of the artwork, which had an increased difficulty from the first puzzle.

 

Playtest 2 Questions:

• Do people respond in expected ways for proper answer checking?
• Do the clues feel at a more appropriate difficulty?
• How do people like having hints after getting something wrong?

 

Feedback:

• Auto Answer Checking

• • • Successes: This feature was successful as we were able to utilize Google Forms capabilities to effectively handle the different ways players submitted their answers to puzzles. By implementing regular expressions for answer checking, the form could accommodate various answer formats, including capital letters and the use of words like “the” and “a.” This setup allowed players more autonomy since it enabled them to interact with the game more freely without strict constraints on answer formats such as “one word answers using all lowercase.” 

• Hint Reveal after Incorrect Answer:

• • • Successes: The introduction of hints when players entered incorrect answers proved effective for infavoidance as it redirected the play tester when they were following incorrect assumptions or stuck in unproductive lines of thought. This feature was instrumental in guiding them back onto the correct path and ensuring they remained engaged while also preventing unnecessary frustration from feeling hopelessly blocked.
    • Failures: Despite the success, the player feedback indicated that hints often felt as if they were “provided too soon,” which reduced the satisfaction they were getting from solving the puzzles independently. The timing of hints being directly after getting something wrong led to a diminished sense of accomplishment when the player eventually got the question correctly. The player also expressed that they wanted guidance that “aligned more with the narrative,” instead of directly telling them what to consider or explore next. 

• Unexpected Issue: Gameplay Feel

• In this playtest, the player expressed that the gameplay mechanics leaned heavily towards research, which suggests that the game could potentially be played without any engagement with the narrative. This aspect was notable in Clue 1 and 3, which required looking up information and where the narrative seemed secondary to the process of gathering facts. Particularly, the player described feeling like they were “doing research on art,” which, while educational, felt stressful and disconnected from the game’s storyline. This mismatch suggests a need for clues that are better integrated with the narrative to enhance coherence and player engagement. Despite these issues, the player appreciated the freedom to use the internet to assist in solving puzzles and also added that they felt like they were “learning about Stanford’s art collection” more. While this is a positive aspect, we also want to adjust the clues to feel more story-based than research-based.

 

Going Forward: In future iterations of the web app, we will work on refining the hint reveal mechanism so that players have the option to get a hint or also choose not to use one. This will allow for them to develop a greater sense of competence from solving each puzzle entirely on their own if they choose to. We will also incorporate the hints into the narrative more and adjust the amount of help the hint gives so they enhance the player’s experience and add to the immersion of the environment.

 

Moreover, we will adjust our Clues 1 & 3 to better balance the gameplay mechanics and narrative. We will ensure that our narrative elements are not only present but integral to the gameplay and that they help shift the experience from a research task to a more engaging story-based game.

 

Playtest 3:

 

This playtest involved a single player and was our final playtest session using the Google Forms prototype and without an on-location experience. For this iteration, we adjusted our narrative elements reflecting feedback from playtest 2 and also changed our clue wording to involve more narrative for understanding the answer. 

 

Playtest 3 Questions:

• Does the narrative feel more consistent with the puzzle’s questions?
• Do people understand the different puzzle locations (without being on location)?
• Are the clues challenging but fun?

 

Features:

• Adjusted Narrative:

• • • Successes: The adjusted narrative was successful and the player mentioned that they liked how the “religious aspect is consistent” and how it gave the game a sense of “continuity and a better level of immersion.” 

• Integrated Narrative in Clues:

• • Successes: the player said that the clues were difficult to understand without reading the narrative more carefully, which is what we hoped for by integrating the clues into the narrative. They also said that it was fun to search for the answer and even complimented the first puzzle saying that it was the “perfect difficulty.” 
  • Failures:  After successfully solving the sin puzzle, the subsequent “Mark of Cain” riddle confused the player. Specifically, the player thought it was out of place as it seemed unrelated to the central goal of discovering their identity and escaping hell. The player said that they “expected the answer to be a name or a group,” given the narrative, rather than something so specific. 

 

Going Forward: Using the feedback and observations from this playtest, going forward we will continue to refine the clues to better align with the overarching story and gameplay goals of the player uncovering their mystery identity. We will review all of the puzzles, including the “Mark of Cain” riddle, to make sure they contribute meaningfully to the narrative. Going forward, we will also use the web app and test on location to get a better sense of the player experience at physical locations.

 

Playtest 4:

 

For our 4th iteration, we conducted our first on-location playtest using our newly developed web app to guide participants through the scavenger hunt. We started by leading players to the “Gates of Hell” sculpture in the Rodin Sculpture Garden and then provided them with the game link, allowing them to independently navigate the puzzles and clues using the app. 

 

Playtest 4 Questions:

• Does the web app function well during play?
• Can people understand and decipher the three clues?
• Are hints useful to players when they get stuck?

 

Feedback:

• • Web App
    • Successes: The newly created web app worked in a much better format than the Google Form as we were able to share each narrative in pieces and take a variety of possible answer submissions, which Google Forms was not able to accommodate. This allowed the players to better understand how to play the game and they finished the playtest at the location within 40 minutes.
  • 3 Clues with Narrative
    • Successes: With this iteration, we included 3 fully fleshed puzzles with clues. The first clue code was understood by the players perfectly, and they felt it had perfect difficulty. This solidified that we had achieved an optimal level of challenge in our clues. The players also appreciated that there was a narrative as well as a direct question leading them to what kind of answer they should expect to give.
    • Failures: On the other hand, for clues 2 and 3, players did not fully understand what the clues were asking for, leading to a lot of confusion and guessing. The players seemed to be getting distracted by the exact wording of the clues, which often misled them. For example, players said that to “pinpoint time period” was too specific wording and made them believe they were looking for a specific time date. The actual answer was “Genesis,” so going forward we will use a word like “era” so they better understand what kind of answer to look for. 

• Multiple Hint Reveal

• Successes: The implementation of multiple hints was successful in aiding participants who struggled to find specific locations to maintain a sense of fun. For example, from the beginning location, they needed to understand that the clue was leading them to McMurtry Library but they had trouble figuring out “what world of knowledge meant.” Using the clue of “Stacks” they understand that they had to go to some sort of book stacks, leading them to the library. By providing these minor location hints, it created opportunities for action and the players were less likely to get stuck and could maintain their playing momentum throughout the game.

• Failures: Because of fidelity issues with the web app, there was no timer restricting when people received hints. Because of this, after trying for a minute, people would get a hint without continuing to think about the puzzles individually on a deeper level. Towards the end of the game, they were looking at the hints immediately because they were available. Going forward, we will implement a timer to prevent abuse of the hint function. Having them readily available made the players feel like they should use it even though they had not spent a lot of time with the puzzle yet. 

• Unexpected Issue: An unexpected issue that came up was that players found it unclear which buildings were part of the puzzle. For example, this caused them to waste time and effort exploring irrelevant areas. Since this was our first playtest at the location, we did not realize how important an understanding of the game area would be for knowing what location to go to. 
• Unexpected Issue: During the playtest, it became apparent that the narrative elements of the scavenger hunt were not essential for solving the puzzles once the players were on location. Participants often ignored or overlooked the story elements, focusing solely on the hints and clues to progress through the game. For example, players said they “didn’t understand the narrative” until the debrief after they had finished the game. This revealed a need for better integration of the narrative to make it a more integral and engaging part of solving each puzzle. 

 

Going Forward: To address the challenges we found in this playtest, we introduce a timer system that makes hints available after a certain period of time, encouraging players to engage more deeply with each puzzle before accessing help. This will help to create a more rewarding problem-solving experience since players will have the opportunity to solve the problems on their own. For the clues that caused confusion, such as the term “pinpoint time period,” we will revise the wording to more intuitive terms like “era,” enhancing clarity and preventing misinterpretation. To address issues with players being unclear about relevant game areas, we will clearly define the playable areas and include a map to better focus player efforts and prevent unnecessary exploration. Lastly, to make the narrative a central component of gameplay, we will introduce more narrative-driven hints and puzzles that require an understanding of the story to solve even when players have the environment around them to help navigate the puzzles. 

 

Playtest 5:

 

In our fifth playtest with two players on location, we introduced several new features: a map of the playing area, timers for hints, and a recap section. The map aimed to help players better understand and navigate the game area, adding to their sense of autonomy as well as competence. Moreover, the timers were designed to control the availability of hints which reflects the feedback that players felt if hints were available they would use them too readily, and the recap section allowed players to revisit their answers to past puzzles.

 

Playtest 5 Questions:

• How is the timing of our game?
• Is the map of the playing area effective in giving players a better understanding of the scope of the game?
• Are timed hints still able to be used by players? 
• Do players understand the recap section and use it?

 

Feedback:

• Map of Playing Area

• • • Successes – This feature was appreciated by the players since it  helped them better understand the playing area and identify which locations were relevant to the puzzle. Previously, players occasionally visited locations that were not part of the designated playing area, leading to wasted time and confusion. With the introduction of the player map, players immediately knew where to go, making their gameplay experience more fluid and enjoyable. 
    • Failures – Despite the success of the player map, some players struggled to realize that there were different, distinct locations within the buildings marked on the map that could be in play. This caused them to overlook critical areas and clues, as they assumed the map only highlighted the overall buildings without detailing specific internal locations. 

• Timers for Hints

• Successes – The timer was effective in preventing players from overusing hints. By making hints available after a delay, players felt that they were more encouraged to try solving puzzles on their own first.

• • • Failures – Since the hints were timed, they often became unhelpful for the player at the point they became available. For example, one hint in the second puzzle was meant to help the player find a specific artifact. However, by the time it became available, the player had already found the artifact and needed help with the next step. This became discouraging since they had to wait even longer for the next hint and they felt stuck. Looking forward, the hints need to be better catered towards where the player should be at that point in time in order to best help them progress. 

• Recap Section

• Successes – The recap section was highly appreciated by players as it allowed them to easily revisit their answers to past puzzles. This feature helped players use previous information to solve subsequent puzzles. Many players found this useful as it also helped them “understand the narrative better” and enabled them to piece together information for the final puzzle more effectively.

• • • Failures – The recap section included a bit of extra contextual information “You killed your brother Abel” that made the next puzzle’s answer too obvious. It gave away the final character to the person paying attention to the narrative because of the wording. 

• Unexpected Issue:  While the 30-minute timing for the game was perfect in terms of pacing, we encountered an unexpected issue: players wanted more puzzles. They found the three provided puzzles enjoyable and were sad that they only got to solve a few puzzles the whole time. To better meet player expectations and maintain engagement, we will incorporate mini-puzzles into the guiding location narrative to give some additional challenges and also make the location aspect more engaging. 

 

Going Forward: To address this issue of the playing map area, we altered our hints to better guide the players to the location of the next clue. We also restructured the hints so that they were more useful to players at different places within the puzzle. For example, if a player had already found the subject of the clue, the later hints should more directly help them figure out the answer. Moreover, we revised the content in the recap section to provide enough information to remind players of their past actions and clues without making future puzzle answers obvious.

 

Playtest 6:

 

In our sixth playtest, we introduced all-new transition narratives and integrated location puzzles to guide players through their journey and help them solve the main puzzles. Each location puzzle included an answer submission feature, which addressed two significant issues from previous playtests:

• Players’ uncertainty about the next steps, which previously led to confusion and wild goose chases.
• The desire for more mini-puzzles to enhance engagement and enjoyment.

This playtest went very smoothly, with no clues found to be overly difficult, and players were able to complete the game within 30 minutes, marking our fastest completion time yet. “Honestly, I think this game is super playable, like if you have been at Stanford for a few years. It’s really not that hard to know where you need to go. Plus you have the play area map, so it’s pretty easy to do the multiple choice in your head. So honestly great for visiting students.” Additionally, we not only confirmed that all game locations are accessible, but that the game is fully playable online, expanding its accessibility. 

 

Feedback:

• Web App was completely built and includes the following:
  • Recap Section
    • Successes: The revamped recap section achieved its intended purpose and was popular with the players. Building on feedback from the previous playtest and removing unnecessary contextual information that unintentionally gave away the puzzle answer led to a much more engaging game experience for the players. We saw increased communication between team members on the second and third stages of the puzzle, and more effort on their part to engage with the narrative to look for direction.
    • Failures: A few design issues arose that led to communication mishaps. The recap section occupied a large portion of the screen, leading players to forget to scroll through the entire app and interact with the clue and hint sections. 
  • Instruction Icon
    • Successes: The instruction icon worked incredibly well to combat the issue of players forgetting to scroll, or not being totally aware of all game elements that they can leverage.
    • Failures: At this point in its design, it crowded the screen too much due to its size and alignment, especially on mobile.
  • Map of Playing Area
    • Successes: The map proved to be crucial during this playtest, and allowed the team to navigate through the location puzzles rather quickly (as intended). 
  • Timers for Hints
    • Successes: Building on the feedback from the last playtest, we redesigned the hint timer system to make sure the hints appeared quicker in order for them to still remain relevant and helpful. This way, even if they appear earlier than intended, the option is still there and the players aren’t blocked on progress to complete the current stage.
  • Submission Timers
    • Successes: The current mechanic is a freeze of sorts on the submission functionality in order to prevent random guessing. During this playtest, players immediately resorted to guessing as the first step, and they encountered this mechanic. The timer appeared, and the submission button was disabled, forcing them to actually attempt solving the puzzle in order to not waste time.
    • Other considerations: The more players on the team, the more tries they get if the app is open on different devices. Although this may seem like a loophole, it is intentional to allow all group members to have the option to make their voices heard. 
  • Transition Narratives
    • Successes: The effectiveness of transition narratives was easily observable as players picked apart at the story. One player, for example, spotlighted key phrases, such as “World of Knowledge” (which referred to the McMurtry library), which allowed the team to quickly navigate to the locations of the main stages.
  • Location Puzzles
    • Successes: Location puzzles paired perfectly with the addition of transition narratives and provided a logical context for their existence. One team specifically said that they really liked doing these in order to zero in on the specific game space that they needed to focus on.
  • Background Music
    • Failures: The background music, although on-theme, lacked a mute button, which made it rather inconvenient for players and some found it a distraction.
    • Successes: Certain teams appreciated the way it set the mood for the game.
• The instruction icon was great for players to understand the premise 
• Transition narratives
  • Resolve the previous issue of players not knowing where to go at first
  • But need to be more clear on the answer submission here – need to clearly state “Enter in Location” 
• Unexpected Issue: It was unclear what answers the transition narratives required. 

 

Going Forward: 

To address the issue of the background music constantly playing, we plan on introducing a toggle mute button to allow players to turn it off if desired. To deal with the size of the recap section, we will introduce a scroll bar as well as reduce the height of the box. In addition, we are also adding a small message to the top that reminds players to scroll during the main stages. Regarding the issue of crowding near the top right of the screen due to the information button, we will reduce it to a simple “info” icon. 

 

Playtest 7:

 

This was the final playtest of our fully developed web app, which now includes sound effects, instructions, narratives, and other enhancements from six previous iterations. Players had 30 minutes to solve all the puzzles. Due to the 90+ degree heat, halfway through the game, they decided to play entirely on their phones—a feature that, while possible, wasn’t explicitly stated in the clues. But learned from the last playtest that this was entirely possible! 

 

They also attempted to take screenshots of clues and use GPT-4, but our game proved to be GPT-proof, as they couldn’t simply generate the correct answers. Although they came very close, they were unable to complete the game within the time limit. One player commented, “I was really excited when I was dropped off at Rodin, because I thought, Omg, there must be so many riddles to solve with all the statues I’ve never paid attention to before.’” Another player acclaimed that we should introduce this concept to the Computer Science department’s annual puzzle hunt, believing that students and faculty would really enjoy it.

 

Feedback:

• Wish the play area was smaller because it was really hot out today to be running around
• Wanted more puzzles throughout the play area and thought it was really fun!
• Unexpected Issue: None!

 

Going Forward:

 

For future work, the addition of more puzzles throughout could boost engagement even further. These can have small rewards or convert to some sort of game currency that helps with the main objective.

 

Accessibility

 

Even though “Unveiling Souls” is a scavenger hunt designed to explore the hidden corners of Stanford Arts, we discovered during playtest 6 that the game can be experienced entirely digitally using Stanford SearchWorks and the Cantor Collections online. This digital adaptation allows players to immerse themselves in the narrative and solve intricate puzzles remotely, making the game highly accessible to those who cannot physically visit the campus or have disabilities preventing them from accessing certain areas. By leveraging these online resources, we ensure that everyone, regardless of their ability to be on-site, can enjoy the rich stories and historical insights embedded within the game.

 

Game Answer Key

 

1. McMurtry
2. Genesis
3. Cantor
4. Murder
5. Rodin
6. Cain