CS Connections: Adapting a Popular Puzzle Game for Computing
Concepts
Ethan Fong

Michelle Craig

Jonathan Calver

University of Toronto
Toronto, Ontario, Canada
ethan.fong@mail.utoronto.ca

University of Toronto
Toronto, Ontario, Canada
mcraig@cs.toronto.edu

University of Toronto
Toronto, Ontario, Canada
calver@cs.toronto.edu

Abstract

1 Introduction

Students often find introductory computing courses challenging,
possibly due to their difficulties with motivation and engagement.
One established strategy instructors use to overcome these hurdles
is to introduce games and puzzles. We developed CS Connections, a
web-based application inspired by the New York Times Connections
game. CS Connections presents students with a grid of elements to
categorize based on shared characteristics. These elements can be
text or code snippets, and their common links may involve runtime
behavior, output, or conceptual relationships. By solving puzzles
through correct grouping, students can review recently-learned
material or actively explore deeper programming concepts.
We deployed the tool in two offerings of our university’s CS1
course, and present our initial results on student retention, perceptions and ease of use. Our initial results suggest that CS Connections
is favorably perceived and students find it useful, but student interest declines over the weeks as topics become more complex and
the novelty of the tool wears off. Instructors found the interface
intuitive, but had difficulty creating novel and relevant puzzles.
However, by encouraging the exploration of programming concepts in an interactive format, the tool shows significant promise
as part of the active learning paradigm.
Open-sourced for broader access, CS Connections is available
online for educators in both CS and other subjects to integrate into
their classrooms. This paper includes instructions for educators
interested in using the tool.

A major challenge that has persisted in introductory Computer
Science (CS1) courses over many years is that of high dropout
and failure rates [3]. While this can be attributed to many factors
including teaching approaches or student learning preferences,
one key reason is student motivation and engagement [8].
Researchers have demonstrated that incorporating games into
the classroom enhances the learning experience by making
abstract concepts more interactive and fun [8, 15]. Thus, in
recent years, there has been growing interest in using games to
boost student motivation in educational settings, including in
Computer Science (CS).
With the intention to incorporate a popular puzzle game format
in our CS1 course, we turned our attention to the New York Times
(NYT) Connections puzzle [12]. Our goal was to explore whether
this format could engage CS1 students and support their
conceptual learning in a optional, supplemental activity that
students could access without any grading or performance
pressure. However, in order to create puzzles with this format, we
found that we needed to create a web-based tool to do so, which
we named “CS Connections”.
Evaluation of educational tools presents numerous challenges
and requires a nuanced approach [19]. Due to the constraints in
which we deployed the tool, we were unable to measure the effect of
this tool on student learning outcomes and understanding. Instead,
we focused this initial evaluation of our tool on three areas that
serve as practical proxies for the value of educational tools:

CCS Concepts
• Applied computing → Interactive learning environments; •
Software and its engineering → Software creation and management; • Human-centered computing → Interaction design.

Keywords
Game-based learning, Educational puzzles, Code classification, CS1,
Student engagement, Web-based tools
ACM Reference Format:
Ethan Fong, Michelle Craig, and Jonathan Calver. 2025. CS Connections:
Adapting a Popular Puzzle Game for Computing Concepts. In The 27th
Western Canadian Conference on Computing Education (WCCCE’25), April
28–29, 2025, Calgary, AB, Canada. 7 pages. https://doi.org/10.60770/3jsj-m574

This work is licensed under a Creative Commons Attribution 4.0 International (CCBY 4.0) license. For all other uses, contact the authors.
WCCCE ’25, April 28–29, 2025, Calgary, AB
© 2025 Copyright held by the authors
https://doi.org/10.60770/3jsj-m574

• Student Retention
• Student Perceptions
• Ease of Use
While we used our tool with CS1 concepts, CS Connections can
also be used in other CS courses or even other disciplines. This
paper presents our reasoning for developing the tool, the initial
evaluation of the tool being used in our classrooms, and instructions
for future instructors looking to use the tool.

2 Background
2.1 Relevant Literature
Games can be an effective tool for learning computing concepts
[13]. A 2016 study found that incorporating challenging games at
the edge of students’ skill level significantly boosts their motivation
to continue learning and overcome challenges [7]. Puzzle games
specifically, are a great way to build intrinsic motivation, as they
provide a sense of satisfaction from the "Eureka moment" when the
solution is discovered [5]. Kasmarik has shown that incorporating

WCCCE ’25, April 28–April 29, 2025, Calgary, Alberta

puzzles, and by extension puzzle games, into a CS1 course increases
students’ interest and active participation [11].
One type of puzzle seen in CS is the code classification puzzle.
Code classification puzzles are a type of challenge where students
identify and organize code segments according to their function.
These puzzles require learners to analyze, classify and match code
snippets to corresponding categories or structures. In the BRACElet
project [17], novice programmers were asked to classify and connect code segments to understand their relationships as part of a
larger program. This significantly improved students’ comprehension of the code and retention of programming concepts. Current
studies on code classification tend to focus on traditional assessment
formats [17] and lack investigation into game-based approaches.
This context motivated us to develop a tool that enables instructors
to transform code classification into an interactive, game-based
activity.
To guide the evaluation of evaluation our tool, we draw on SelfDetermination Theory (SDT), a psychological framework that emphasizes autonomy, competence, and relatedness as essential for
motivation [14]. We choose this framework as it has been commonly
applied to the design and evaluation of games in HCI research [18].
In educational contexts, tools that allow students to make their
own choices (autonomy), build skills through manageable challenges (competence), and feel connected to others or to the learning
environment (relatedness) are more likely to foster intrinsic motivation and sustained engagement [14]. Therefore, we selected three
evaluation criteria that we could both measure and reflect these
motivational principles:
• Student Retention can provide insight into whether the
tool helps support autonomy by observing whether students
choose to re-engage with it voluntarily. Although other factors may influence students’ repeated use of the tool, the
fact that some students chose to engage with it voluntarily
over multiple weeks suggests they saw value in the activity.
This kind of self-directed engagement aligns with conditions
that support autonomous motivation, where students participate more because they find the experience personally
meaningful or enjoyable.
• Student Perceptions can reveal whether the tool supports
competence and relatedness by indicating if students found
the puzzles valuable, appropriately challenging, and relevant to their learning experience. Perceptions of usefulness
suggest support for competence, while mentions of collaboration or shared experience can reflect relatedness.
• Ease of Use supports both competence and autonomy by
minimizing technical barriers that might otherwise lead to
frustration or disengagement. When students can easily navigate and use the tool, they are more likely to feel capable
of succeeding (competence) and in control of their learning experience (autonomy), both of which are essential for
fostering intrinsic motivation according to SDT.

2.2 NYT Connections
The NYT Connections game [12] challenges players to organize
a grid of 16 words or phrases into four groups of four based on
shared themes or categories. Players must find the common link

Ethan Fong, Michelle Craig, and Jonathan Calver

between items in each group such as synonyms, colors, or pop culture references while avoiding red herrings designed to fit multiple
categories. Once all the items in a group are correctly identified,
the game reveals the category. Figure 1 presents a sample game.
One category, called "Slimy Animals", has already been found and
grouped. The puzzle is solved once all four categories are revealed,
but only if the player does not make four mistakes while trying to
group the categories.

Figure 1: Sample Game of NYT Connections (link)
Despite Connections’ widespread appeal as a recreational puzzle,
its use in educational contexts is largely unexplored. A recent study
adapted the format in the context of medical education [16], suggesting broader educational potential. To our knowledge, Connectionsstyle puzzles have not been used to support CS1 concept learning.
This represents a valuable opportunity to explore how this familiar,
classification-based puzzle format might foster engagement and
conceptual thinking in introductory programming courses. Using
a familiar format like Connections could make learning feel less
intimidating to students and help present or review content in a
more accessible way.

3

Adapting NYT Connections for CSEd

Tools to create and share NYT-inspired Connections puzzles can be
found on the internet [2, 4]. However, through our own usage, we
found that the existing tools lacked key features needed to use the
puzzles in a CS1 context. Critically, they did not support using code
snippets as game elements. Table 1 summarizes the main features
that we found were lacking and the associated rationale for needing
these features.
Therefore, recognizing available Connections creation tools did
not meet our needs, we developed our own web-based application
that has built-in support for code snippets as game elements. This
allowed us to implement the features that we could not find in
other Connections tools and gave us control over the data that is
collected from these games.

CS Connections: Adapting a Popular Puzzle Game for Computing Concepts

Required Feature

Reasoning

The ability to cus- Allows instructors to fit larger inditomize grid size (M vidual elements in a game and supcategories of N)
ports more flexible game creation
Support for lowercase
letters

Coding languages like Python are
case-sensitive, so capitalization is
critical to represent code clearly

Support for special
characters such as
commas

These are essential for representing
valid code syntax as elements

Use of characters such
as newlines and tabs

These are essential for allowing
longer code snippets as elements

Prerequisite informa- Helps present variable initializations
tion field
or assumptions concisely
Data collection on
guesses and categories

Provides insights into student performance and common challenges

Table 1: Feature Requirements and their Rationale

WCCCE ’25, April 28–April 29, 2025, Calgary, Alberta

Students can also create their own Connections games using a
game creation form on the CS Connections webpage. All puzzles
are organized on a centralized landing page, grouped by course
for easy navigation. Instructor-generated puzzles are distinguished
from student-generated puzzles, allowing students to quickly find
instructor-generated puzzles while still being able to browse student
puzzles.

3.2

3.3
3.1

User Interface and Game Example

The interface of the game is based on an open source rendition
of the NYT Connections game written in React.js and available
on Github [1]. To tailor it specifically for CSEd, we made several
enhancements to the original interface. Features such as syntax
highlighting allow players to easily distinguish between code snippets, while responsive scaling ensures the game works seamlessly
across devices. See Figure 2 for an example CS Connections game,
with one category revealed.

Instructor Interface of the Tool

The instructor interface of CS Connections includes authentication
and a dashboard for managing games in their courses. After logging
into the platform, instructors can create new puzzles through the
game creation form. The form includes a built-in preview feature,
allowing instructors to visualize and test how the puzzle will appear
and behave for students before publishing it. Each puzzle has a
JSON representation which allows for the storing and sharing of
puzzles. The JSON representation of any puzzle can be downloaded
from the homepage of the app and used in game creation allowing
instructors to reuse and edit past puzzles.

Data Collection and Statistics View

One advantage of using our CS Connections tool over other existing
Connections tools is the data collection that is integrated within
the app.

Figure 3: In-game Statistics View

Figure 2: Example Game on List Indexing

Our CS Connections app collects gameplay data for each game
played, capturing details about the cumulative time it took to make
each guess, and the top 10 most common groupings made by the
students as shown in Figure 3.

WCCCE’25,April28–April29,2025,Calgary,Alberta

4 Results
Inthissection,wepresentourresultsfromtheuseofCSConnectionsintwoofferingsofouruniversity’sCS1course,whichis
taughtinPython.InboththeFall2024andWinter2025semesters,
wepresentedweeklypuzzlesinthe10minutesbeforelecture.The
CS1coursemeetsforthreehoursoflectureseachweekandfollowsanactivelearningapproachthatincludesactivitiessuchas
worksheetsduringlectures,makingtheinclusionofthesepuzzles
anaturalextensionofthecourse’sinteractiveformat.Someofthe
CS1topicsanddesignchoiceshavealsobeendocumented[6].

4.1 EvaluationCriteria
Becauseadoptionofthistoolisstillinitsearlystages,wedidnotaim
todirectlymeasurestudentlearningoutcomes,whichwouldrequire
morecontrolledconditionsandlong-termassessment.Instead,we
selectedthreeevaluationcriteriabasedonourinterpretationof
SDTandusedthefollowingdatasourcestoevaluateourtool:
• Studentretentionviagameplaystatistics(𝑛=674)
• Studentperceptionsviaend-of-termsurveyresponses(𝑛=
22)
• Easeofuseviastudentfeedback(𝑛 = 22)andinstructor
reflections(𝑛=2)

hasbeenshowntoimproveengagementwithothereducational
tools[10].
Number of Plays by Week (n=674)


200





Number of Plays

Whenthegameiscompletedorthestudentpressesthe“Give
Up”button,thesolutiontothepuzzleisrevealed,andthegameplaydataissenttothebackend.Thecollecteddataincludesguess
distributions,timetakenforeachgrouping,andwhetherthegame
wassuccessfullycompleted.
Gameplayinformationisautomaticallyvisualizedinadedicated
statisticsviewforeachgame.Thisenablesinstructorstoidentify
trendsinstudentperformance,suchascommonlymissedcategories
orcategorieswhichstudentstooklongertocategorize.Theseinsightshelpinstructorsaddressareaswherestudentsarestruggling
andcreategamesattherightdifficultylevel.

EthanFong,MichelleCraig,andJonathanCalver

150



100



50

0





2



3



4



5



7

Week





8



9



10



11



12





Figure4:GameSubmissionsbyWeek

4.3 StudentPerceptions
Inordertoassessstudentopinionsonthegame,weappliedforand
receivedethicsapprovalfromtheUniversityEthicsBoard(Protocol
Reference#:00047439)toconductasurveyonstudentperceptions
ofthegame.ThissurveywasdistributedtotheFall2024CS1cohort
ofover400students.Therewereatotalof22responsesandthe
surveywasstructuredintothreesections;thefirstaimedtogauge
studentprofiles,thesecondmeasuredperceptionsofthegameand
thethirdcollectedopen-endedfeedback.
4.3.1 StudentProfiles. Tobetterunderstandthecontextofthe
surveyresponses,wefirstgatheredinformationaboutthestudents
whoresponded.SeeTable2forasummaryoftheresults.
SurveyCategory

Option

IntendtostudyCS

Definitely
Mostlikely
Mostlikelynot
Definitelynot
Undecided

2(9.1)
5(22.7)
7(31.8)
3(13.6)
5(22.7)

Gender

Man
Woman
Gendervariant/non-conforming

12(54.5)
9(40.9)
1(4.5)

PlayedNYTVersionBefore

Yes
No

16(76.2)
5(23.8)

IndividualorGroupPlay

Alwaysindividually
Mostlyindividually
Sometimesinagroup
Mostlyasagroup
Alwaysinagroup

6(28.6)
9(42.9)
2(9.5)
3(14.3)
1(4.8)

PlayAttheBeginningofLectures

Frequently(6-10times)
Occasionally(3-5times)
Rarely(1-2times)

6(28.6)
10(47.6)
5(23.8)

PlayOutsideofLecture

Veryfrequently(morethan10times)
Rarely(1-2times)
Never

1(4.8)
5(23.8)
15(71.4)

4.2 StudentRetention
FortheFall2024offeringofourCS1course,therewere674games
playedacross11weeklypuzzles.Thedatashowsthattheinitial
gameswereplayedmuchmorethanthosefromsubsequentweeks.
Thefirstweektheappwasintroducedwasthemostpopular,with
220gamesplayedthatweek.ThenumberofplaysisvisuallyrepresentedinFigure4.Notethatthegraphdoesnotincludestatistics
fromthefirstweekaswewerestillusingotheronlinetoolsto
createthepuzzlesthuswecouldnotcollectanygameplaydata.
Studentengagementinitiallystartsoutstrongbuttrendsdownwardsovertime.Thissuggeststhatinitialinterestinthegamemay
havebeendrivenbyitsnovelty,whilesustainedengagementlikely
requiredmoreincentivesorintegrationintothecoursestructure.
Thissuggeststhatthetoolinthewaythatwepresenteditmay
nothavebeenconducivetoautonomousmotivation.However,itis
importanttonotethatarangeofexternalfactors(suchasmidterms,
decliningattendance,andexampreparation)mayhavealsocontributedtothedecliningparticipation.Futureiterationsofthegame
couldincorporatedirectincentives,suchashavingasimilargroupingproblemonatestorbyusing"marketing"techniquessuchas
weeklyemailremindersdesignedtoadvertisethepuzzleswhich

6

OpenedtheInstructorExplanations Everytime
Mostofthetime
Halfthetime
Lessthanhalfthetime
Never

Table2:SurveyResultsforStudentProfile

Count(%)

5(23.8)
6(28.6)
3(14.3)
4(19.0)
3(14.3)

CS Connections: Adapting a Popular Puzzle Game for Computing Concepts

From the survey, we see that a large majority of students never
or rarely play the games outside of lecture time, suggesting that
the in-class engagement with the game may be sufficient for its
intended purpose. Alternatively, this could also indicate that students view the game as primarily a classroom activity rather than
an independent learning tool.
Another insight is that students tend to complete the game individually, consistent with findings from the use of a Connections
game in a medical context [16]. Although our tool was designed
to support relatedness by encouraging social interaction, this outcome suggests that students may naturally default to individual
play. To better support relatedness and foster a sense of community as outlined in our SDT-based evaluation criteria, instructors
might consider explicitly encouraging collaboration to promote
more social engagement with the puzzles.
4.3.2 Likert Data. The second part of the survey used a Likert
scale which provided students with options ranging from strongly
disagree to strongly agree. The results are shown in Figure 5.

Figure 5: Likert Chart of Survey Responses
Students expressed a positive sentiment towards the use of the
tool with 53% of students respectively choosing "Agree" or "Strongly
agree" when asked if they enjoyed using the tool. Similarly, students largely agreed that the games helped to review knowledge
with 76% of students choosing "Agree" or "Strongly agree" options.
This suggests the tool supports students’ sense of competence by
providing meaningful practice that challenges students at an appropriate level. However, students were notably less positive when
asked whether the tool helped them feel ready for lecture. This contrast may indicate that while the tool reinforced past content, this
was not enough to make students feel comfortable with learning
upcoming material.
It is also worth noting that students who reported enjoying the
tool may already have a predisposition towards puzzle-solving,
which could influence their positive perceptions. Overall, we see
that the tool was able to generate positive perceptions related to
students’ sense of competence, but it did not appear to foster relatedness, which might have emerged more strongly if the games
were played in a social context.

4.4

Student Ease of Use

In addition to questions about learning value and enjoyment, the
survey also asked students to evaluate the tool’s ease of use. As

WCCCE ’25, April 28–April 29, 2025, Calgary, Alberta

shown in the corresponding row of Figure 5, the responses suggest
that while many students found the interface accessible, there is
still room for improvement. Specifically, 66% of students agreed or
strongly agreed that the tool was easy to use, whereas 10% reported
difficulty and 24% remained neutral.
4.4.1 Vignettes of Student Feedback. The survey also collected
student feedback on the games. Table 3 summarizes some relevant
feedback points and takeaways, which may serve as guidance for
future directions of the app.
Table 3: Student Feedback and Possible Adjustment
Feedback

Adjustment

I didn’t have time to play the game
before each class as my phone wasn’t
able to read the QR codes and it took
too long to access the games. I played
the games at the end of the course
while studying for the final exam and
found them extremely educational.

Allow more time for the activity. In the 2024 Fall CS1
course, the QR code for the
game was only visible for
10 minutes before lectures
start.

I might have missed the goal of your
project at the beginning of the course.
It would be helpful for the professor
to go over it with us during the first
class.

Since this project was developed as the course was
progressing, there could
have been clearer communication on the goals of the
project.

It would be helpful to provide a short
description of the game objective for
each week.

We could add an explicit
game objective when presenting the puzzles.

(Instructor explanations are) mostly
helpful but sometimes not detailed
enough; could use more step-by-step
breakdown of the solutions.

In addition to stating the
category, the instructor explanation should explain
the process to arrive at the
correct grouping.

Overall, the results of the survey indicate that students enjoy
using the tool but there is room for improvement in making the
games understandable for all students. Ease of use was also a challenge seen in another use of Connections games in an educational
context [16] and is a common challenge in adopting game-based
learning [9]. Therefore, in future iterations of the game, it would
be beneficial for instructors to explain the goals of the puzzle, and
possibly provide an example of how the game works to help address
students’ concerns over ease of use.

4.5

Instructor Ease of Use

We gathered feedback from two instructors who used the CS Connections tool for two offerings our University’s CS1 course. Here
we present a summary of their experiences.
4.5.1 2024 Fall CS1. At first, the instructor found it challenging
to create good puzzles. She was making the puzzles too hard by
assuming students already deeply understood the concepts (rather
than including easier categories that could be selected without deep

WCCCE ’25, April 28–April 29, 2025, Calgary, Alberta

understanding). After a few weeks, she requested the instructorexplanation feature to address this issue. She reflects however, that
by this time many students had stopped participating because of
the initial challenging puzzles. Nevertheless, there was still a small
group of students that continued to engage with the puzzles and
had very positive attitudes towards the games.
On the topic of creating puzzles, this instructor was surprised at
the variety of concepts she could explore with the puzzles. During
the term she felt that it was important to keep consistent with the
presentation of these puzzles to cater to the students that already
have built a habit of playing them. Note that she only presented
puzzles in a passive way on the pre-class lecture announcements.
In post-term reflection she wondered if instead of being presented
weekly, one or two puzzles could be used judiciously to replace a
few of the in-class active-learning worksheets.
4.5.2 2025 Winter CS1. For the Winter semester, a different instructor was able to re-use puzzles from the previous offering of
the course which allowed the instructor to focus on student engagement rather than content creation. The instructor reflected
on the need to be mindful when creating puzzles, noting that it is
easy to design puzzles that lack a "fun" factor. Maintaining variety across category types was identified as a key design principle.
For example, it would be tempting to create categories that are all
variations on "outputs [BLANK] in the terminal." Once students
encounter a few similar categories, the puzzle becomes repetitive,
detracting from the puzzle. Being intentional about the inclusion
of red herrings (elements designed to fit into multiple categories)
and incorporating overlapping categories also helped to create opportunities for eureka moments, which are critical for an effective
puzzle experience.
Both instructors emphasized that the most significant challenge
in using the platform was not the interface itself, but the effort
required to create original, high-quality puzzles. Designing puzzles
that are both effective and engaging takes time and creativity, especially without prior examples to draw from. To improve ease of use,
we aim to continue expanding the shared repository of puzzles. As
more instructors adopt the tool and contribute their puzzles, it will
become easier to create new games by drawing inspiration from
existing ones. This growing knowledge base can lower the barrier
to entry and make the tool more approachable to other instructors.

5

Adopting CS Connections

Since CS Connections is a web-based tool, instructors do not need
to download or run any additional software to use the tool. Instructors can choose to use our online version of the app or to utilize
the source code and host the app themselves. While self-hosting requires additional setup, it offers full control over the data collected.
The source code of the app is available at https://github.com/ethanfong/cs-connections, and our online version can be accessed at
https://cs-connections.app. Regardless of the hosting method, follow these steps to create the first game.
1. Create a Course on the Platform. Instructors interested in using
the platform can request a course to be created on the platform
using the provided contact information on the homepage of CS
Connections.

Ethan Fong, Michelle Craig, and Jonathan Calver

2. Receive Credentials to Upload Games. After having a course created, they will also be provided with credentials to log into the
instructor interface to upload and manage games.
3. Create a Puzzle and Upload it. Instructors can create a game using
the game creation form and once the form is submitted, the app
will generate a four letter game code and the game will be visible
for students to play online.
4. Present Puzzles to Class. Present the link to the class to allow students to play the games. One way to share the link is by displaying
a QR code linking to the puzzle so students can access the games
from their mobile devices.

6

Future Directions

CS Connections has been designed as an open-source tool to ensure
accessibility for educators and students for use in the classroom
or for further research. Over time, CS Connections continues to
be iteratively improved, based on needs expressed by instructors.
The CS Connections tool is being tested in a CS2 course and a Data
Structures and Algorithms course, which has expanded the tool to
be used with Java and a language-independent theory course on
advanced data structures.
The code classification format of the NYT Connections puzzle
was selected because prior research has shown that these types
of exercises improve beginners’ understanding of code. However,
while this pedagogical foundation informed the tool’s design, further research is needed to assess whether the specific implementation of the game leads to measurable improvements in students’
learning and conceptual understanding. Future studies could explore whether playing the games before lectures helps students
better grasp key programming concepts. Evaluating these outcomes
would help determine the educational value of the tool beyond engagement and motivation.

7

Conclusion

CS Connections introduces an interactive, game-based learning
activity that instructors can easily integrate into their classrooms.
By adapting the familiar structure of the NYT Connections game,
the platform adds an interactive element to the classification puzzle. Results from its use in a CS1 course show promise in initial
student engagement, although sustained interest retention remains
a challenge. Survey results indicate that students are enjoying the
game and find it useful as a review tool, but more can be done to
encourage the students to treat this as a social activity. Instructors’
experiences emphasize the difficulty in creating original and relevant puzzles, a challenge that should be alleviated as the knowledge
base of puzzles expand through continued instructor use. Another
insight was to intentionally thinking of a "fun" factor for each puzzle such as by using red herring elements and varying category
types.
Ultimately, CS Connections represents one component of the active learning toolkit, offering educators a convenient and enjoyable
addition to their teaching arsenal. In this way, the platform can
serve as a robust tool for fostering both engagement and deeper
learning in CS and other subjects.

CS Connections: Adapting a Popular Puzzle Game for Computing Concepts

References
[1] AndComputers. 2024. React Connections Game.
https://github.com/andcomputers/react-connections-game
[2] Anthony Kenzo Salazar. 2024. Swellgarfo Connections. https://connections.
swellgarfo.com/
[3] Jens Bennedsen and Michael E. Caspersen. 2007. Failure rates in introductory
programming. SIGCSE bulletin 39, 2 (2007), 32–36. https://doi.org/10.1145/
1272848.1272879
[4] Claire Wang. 2024. Custom Connections. https://custom-connections-game.
vercel.app/
[5] Karolína Dočkalová Burská, Vít Rusňák, and Radek Ošlejšek. 2022. Data-driven
insight into the puzzle-based cybersecurity training. Computers Graphics 102
(2022), 441–451. https://doi.org/10.1016/j.cag.2021.09.011
[6] Ethan Fong, Michelle Craig, and Jonathan Calver. 2025. Crafting Interesting
Puzzles with CS Connections. In Proceedings of the 2025 Conference on Innovation
and Technology in Computer Science Education V. 2 (Nijmegen, Netherlands)
(ITiCSE 2025). Association for Computing Machinery, New York, NY, USA, 2 pages.
https://doi.org/10.1145/3724389.3731261
[7] Juho Hamari, David Shernoff, Brianno Coller, Jodi Asbell-Clarke, and Teon Edwards. 2016. Challenging games help students learn: An empirical study on
engagement, flow and immersion in game-based learning. Computers in Human
Behavior (08 2016). https://doi.org/10.1016/j.chb.2015.07.045
[8] María-Blanca Ibáñez, Ángela Di-Serio, and Carlos Delgado-Kloos. 2014. Gamification for Engaging Computer Science Students in Learning Activities: A
Case Study. IEEE Transactions on Learning Technologies 7, 3 (2014), 291–301.
https://doi.org/10.1109/TLT.2014.2329293
[9] Elina Jääskä and Kirsi Aaltonen. 2022. Teachers’ experiences of using game-based
learning methods in project management higher education. Project Leadership
and Society 3 (2022), 100041. https://doi.org/10.1016/j.plas.2022.100041
[10] Abdullah Karaksha, Gary Grant, Shailendra Anoopkumar-Dukie, S. Niru
Nirthanan, and Andrew K. Davey. 2013. Student Engagement in Pharmacology Courses Using Online Learning Tools. American Journal of Pharmaceutical
Education 77, 6 (2013), 125. https://doi.org/10.5688/ajpe776125

WCCCE ’25, April 28–April 29, 2025, Calgary, Alberta

[11] Kathryn Kasmarik. 2010. An Empirical Evaluation of Puzzle-Based Learning as an
Interest Approach for Teaching Introductory Computer Science. Education, IEEE
Transactions on 53 (12 2010), 677 – 680. https://doi.org/10.1109/TE.2009.2039217
[12] New York Times. 2023. New York Times Connections. https://www.nytimes.com/
games/connections
[13] Marina Papastergiou. 2009. Digital Game-Based Learning in high school Computer Science education: Impact on educational effectiveness and student motivation. Computers Education 52, 1 (2009), 1–12. https://doi.org/10.1016/j.compedu.
2008.06.004
[14] R M Ryan and E L Deci. 2000. Self-determination theory and the facilitation of
intrinsic motivation, social development, and well-being. Am. Psychol. 55, 1 (Jan.
2000), 68–78.
[15] Bianca L. Santana and Roberto A. Bittencourt. 2018. Increasing Motivation of
CS1 Non-Majors through an Approach Contextualized by Games and Media.
In 2018 IEEE Frontiers in Education Conference (FIE). IEEE Press, 1–9. https:
//doi.org/10.1109/FIE.2018.8659011
[16] Shane N. Stone and Steven Kirshblum. 2024. PMR connections: A pilot study evaluating the adoptability and impact of a serious educational game on trainees at
a Physical Medicine and Rehabilitation Program. American Journal of Physical Medicine amp; Rehabilitation (Sep 2024). https://doi.org/10.1097/phm.
0000000000002620
[17] Errol Thompson, Jacqueline Whalley, Raymond Lister, and Beth Simon. 2006.
Code Classification as a Learning and Assessment Exercise for Novice Programmers.
[18] April Tyack and Elisa D. Mekler. 2024. Self-Determination Theory and HCI
Games Research: Unfulfilled Promises and Unquestioned Paradigms. ACM Trans.
Comput.-Hum. Interact. 31, 3, Article 40 (Aug. 2024), 74 pages. https://doi.org/10.
1145/3673230
[19] Megan Venn-Wycherley, Ahmed Kharrufa, Susan Lechelt, Rebecca Nicholson,
Kate Howland, Abrar Almjally, Anthony Trory, and Vidya Sarangapani. 2024.
The Realities of Evaluating Educational Technology in School Settings. ACM
Trans. Comput.-Hum. Interact. 31, 2, Article 26 (Feb. 2024), 33 pages. https:
//doi.org/10.1145/3635146