Narrative-Driven Learning: Teaching Finite State Machines Through

Storytelling

Bogdan Iudean

and Andreea Vescan

Computer Science Department, Faculty of Mathematics and Computer Science,

Babes¸-Bolyai University, Cluj-Napoca, Romania

{bogdan.iudean, andreea.vescan}@ubbcluj.ro

Keywords:

Finite State Machine, Storytelling, Teaching, Learning.

Abstract:

Teaching embedding systems demands a thorough understanding of computation models, with Finite State

Machines (FSM) serving as one of the frameworks for developing the intricate behaviors of these systems. This

study investigates the effectiveness of teaching Finite State Machines through storytelling using a controlled

experiment with two groups: a Control Group (CG) taught traditionally and a Storytelling Group (SG) taught

using a narrative-driven video. Data from pre-tests and post-tests and feedback questionnaires reveal that

while storytelling did not signiﬁcantly improve post-test scores, it enhanced application skills in the SG group.

Storytelling helped students better connect FSM concepts to real-life scenarios like trafﬁc lights and vending

machines, while CG responses remained more abstract. Additionally, SG participants rated the video highly

for attention and clarity, though some preferred more direct explanations. These ﬁndings suggest storytelling

supports conceptual understanding and engagement, offering a complementary approach to teaching FSMs.

1 INTRODUCTION

Finite State Machines (FSMs) are fundamental to un-

derstanding computational models in embedded sys-

tems, offering a structured approach for modeling sys-

tems with discrete states and well-deﬁned transitions

based on inputs. FSMs, deﬁned as “conceptual ma-

chines capable of existing in one of a ﬁnite number

of states at any given time” (Lee and Yannakakis,

1996), are extensively used across disciplines, from

engineering and computer science to control systems

and natural language processing (Karttunen, 2000).

The FSM concept has evolved as a core framework

for system modeling, simplifying the representation

of complex behaviors and operations that character-

ize many embedded systems (Radojevic and Salcic,

2011). Despite their signiﬁcance, FSMs often present

learning challenges for students due to their abstract

nature and the requirement to think in discrete, se-

quential steps (Henry et al., 2022).

Storytelling as a Teaching Strategy for Complex

Concepts. Educational research increasingly empha-

sizes the power of storytelling in conveying complex,

technical concepts (Wu and Chen, 2020). Storytelling

https://orcid.org/0000-0001-6233-3570

https://orcid.org/0000-0002-9049-5726

leverages the human afﬁnity for narratives to provide

context, create emotional connections, and improve

retention, making it especially effective in STEM ed-

ucation where abstract concepts can be challenging

to grasp (Collins et al., 2023; Angel-Fernandez and

Vincze, 2018). Through storytelling, complex com-

putational models can be presented in a more engag-

ing, memorable format. Studies (Heymann and Gre-

eff, 2018; Parham-Mocello et al., 2019; Resnyansky,

2020; Min et al., 2020) indicate that narrative-driven

instruction can enhance cognitive processing by al-

lowing learners to form mental models based on fa-

miliar narrative patterns, which in turn supports bet-

ter conceptual understanding. This approach is sup-

ported by theories of situated learning, which suggest

that knowledge is best understood when contextual-

ized in realistic scenarios (Segel and Heer, 2010).

Our study builds on these insights by exploring

the potential of storytelling, speciﬁcally through ani-

mated video, as a method to teach FSM concepts in

a graduate-level computational models course. The

video used in this study tells the imaginative story

of a prehistoric man, Buga, who attempts to tame

a mammoth using an FSM to understand its behav-

ior. By embedding FSM principles into a narrative

structure, we aimed to make FSM concepts more ap-

proachable and relatable for students. Story-based ap-

Iudean, B. and Vescan, A.

Narrative-Driven Learning: Teaching Finite State Machines Through Storytelling.

DOI: 10.5220/0013471600003932

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 17th International Conference on Computer Supported Education (CSEDU 2025) - Volume 2, pages 627-638

ISBN: 978-989-758-746-7; ISSN: 2184-5026

627

proaches have shown promise in promoting engage-

ment and improving learning outcomes in the previ-

ously mentioned STEM ﬁelds (Collins et al., 2023;

Angel-Fernandez and Vincze, 2018), suggesting that

they may similarly beneﬁt students studying compu-

tational models for embedded systems.

The aim of the study is to evaluate the effec-

tiveness of using visual storytelling as a teaching

method within a take-home assignment activity for

using ﬁnite state machines (FSMs) in a university-

level course. One group of students was given access

to the educational video along the traditional learn-

ing materials, while another group received only tra-

ditional instructional material without the video. The

study seeks to determine whether the storytelling ap-

proach enhances students’ understanding, academic

performance, and engagement compared to conven-

tional teaching techniques.

The research investigation is formulated under the

following three research questions, with the ﬁndings

being articulated.

• RQ1: Are there measurable differences in learn-

ing outcomes between students who engaged with

the storytelling video content and those who fol-

lowed traditional instructions? Answer: The dif-

ference, did not reveal a better performance for

the storytelling group, however, qualitative results

showed better results regarding the application

skills.

• RQ2: Does storytelling video content improve stu-

dents’ understanding of FSMs compared to con-

ventional instructional approaches? Answer: The

storytelling group answers reveal a deeper under-

standing of FSM concepts; we based our answer

on results of the qualitative analysis.

• RQ3: How does the storytelling video content ap-

proach impact student engagement and percep-

tion of learning FSMs? Answer: The storytelling

group revealed more enthusiasm towards the sub-

ject; we based our answer on the results of the

qualitative analysis.

The research methodology consisted of a con-

trolled experiment using two student groups, one

taught FSMs through a storytelling video and the

other through traditional instruction. Using both qual-

itative and quantitative analysis, data on student per-

formance and engagement were collected through as-

sessments, questionnaires, and observational feed-

back.

This study’s ﬁndings will contribute to a growing

body of research on innovative instructional strategies

in computational and engineering education. By in-

vestigating storytelling’s impact on FSM comprehen-

sion, this research may inform future approaches for

teaching complex systems and computational models,

ultimately enhancing educational practices in techni-

cal disciplines.

The contributions of this research are:

• A Novel Pedagogical Approach for Teaching

FSM Concepts. This study introduces a pro-

totype multimedia storytelling-based method for

teaching abstract technical concepts in an engag-

ing way.

• Evaluation Through Multi-Stage Question-

naires and Hands-on Tasks. The effectiveness

of the storytelling approach was evaluated using

a mix of qualitative and quantitative analysis that

used pre-tests and post-tests.

The rest of the paper is organized as follows: Sec-

tion 2 covers FSM education, the use of storytelling,

and instructional design with media. Section 3 details

the research methodology with instructional informa-

tion and corresponding teaching artifacts. In Section

4, the results are presented with quantitative and qual-

itative analysis. Section 5 discusses the results of

our study, highlighting key ﬁndings, while Section 6

explores the broader implications, recommendations

for educators, and future research directions. Finally,

Section 7 raises the threats to validity, and Section 8

concludes the study.

2 RELATED WORK

Next, related work is reviewed regarding FSM educa-

tion, storytelling, and media-based design, emphasiz-

ing storytelling’s role in contextualizing concepts and

media’s ability to simplify complex topics visually.

2.1 FSM Education and Learning

Strategies

FSMs are commonly used in engineering and com-

putational sciences to model systems with distinct

states and transitions. However, traditional teaching

methods (Joseph et al., 2013) often rely on abstract,

mathematical instruction, which can be difﬁcult for

students to grasp and lacks engagement (Dekeyser

and Aljendi, 2015). Innovative approaches, such

as interactive platforms and augmented reality, have

shown promise in making FSM learning more prac-

tical and engaging (Nadeem et al., 2022; Dengel,

2018). Building on these strategies, our study intro-

duces a narrative-based video where a prehistoric man

models a mammoth’s behavior using FSM concepts to

CSEDU 2025 - 17th International Conference on Computer Supported Education

628

help students connect theory with real-world applica-

tions.

2.2 Use of Storytelling in Education

Storytelling (Barchas-Lichtenstein et al., ; Landrum

et al., 2019) is a powerful tool in STEM educa-

tion, helping students understand complex ideas by

embedding abstract concepts in relatable narratives

(Barchas-Lichtenstein et al., 2023). This approach

fosters emotional connection, which enhances en-

gagement and retention, particularly for abstract com-

putational topics (Groshans et al., 2019). Research

shows that storytelling bridges cognitive gaps by

contextualizing information, making STEM subjects

more accessible (Mou, 2024). Digital storytelling

further enriches learning by combining audio-visual

elements with narrative structures (Dochshanov and

Tramonti, 2022). Our study builds on this research

by using structured storytelling elements (Durak,

2018)—such as narrative, setting, and plot—to cre-

ate an engaging framework speciﬁcally for teaching

FSMs.

2.3 Media and Animation in

Instructional Design

The use of media and animation in education has be-

come popular, especially for technical subjects where

visual aids enhance understanding. Animations help

simplify complex topics like FSM states and transi-

tions (Dengel, 2018) by breaking them into digestible

visual segments (Berreth et al., 2020; Hill and Grin-

nell, 2014). Media-rich content creates dynamic, in-

teractive learning environments that foster engage-

ment and support long-term retention (Bravo et al.,

2021). Our study adopts this approach by using vivid

animations, characters, and a visual narrative, along

with an on-screen instructor who narrates and uses

gestures to reinforce FSM concepts.

3 METHODOLOGY

This section details the research methodology used

to evaluate the effectiveness of storytelling-driven

video in teaching Finite State Machine concept. The

methodology includes deﬁning the research question,

describing the teaching materials and content, and ex-

plaining the data collection process through question-

naires. A controlled experiments is employed using

two groups of students: CG (Control Group) and SG

(Storytelling Group). A mixed method research is

used to analyze the collected data, both quantitative

and qualitative analysis.

3.1 Course and Participants

The study involved two groups of graduate students

enrolled in the “Computational Models for Embed-

ded Systems” course at [University Anonymous for

review]. In total, 33 students participated into the

study, students being divided into two groups: Group

CG (control group, 16 students) and Group S (sto-

rytelling group, 17 students). The two groups were

formed randomly; all students ﬁrst enrolled in the ac-

tivity and the researchers assigned to each group the

enrolled students in a random way. CG group re-

ceived traditional instruction in FSM design, while

SG group engaged with a storytelling-driven video

teaching FSM concepts through a prehistoric story-

line. More information on the activity is provided in

Section 3.4, outlining the teaching method within a

take-home assignment activity. All participants gave

informed consent prior to participation, and the study

complied with ethical guidelines for research by [Uni-

versity Anonymous for review].

3.2 Objectives and Research Questions

The primary objective of this study is to assess

whether storytelling in video form could improve

comprehension of FSM concepts among graduate stu-

dents compared to traditional instructional methods.

We aim to answer the following research questions:

RQ1: Are there measurable differences in

learning outcomes between students who en-

gaged with the storytelling video content and

those who followed traditional instructions?

RQ2: Does storytelling video content im-

prove students’ understanding of FSMs

compared to conventional instructional ap-

proaches?

RQ3: How does the storytelling video ap-

proach impact student engagement and per-

ception of learning FSMs?

3.3 Teaching Materials and Content

The primary teaching material for Group SG was a

10-minute animated video titled “Buga and the Mam-

moth: A Finite State Machine Story”, designed to in-

troduce FSM concepts through a storyline of a prehis-

toric character, Buga, who attempts to tame a mam-

moth by modeling its behavior with FSMs. The video

Narrative-Driven Learning: Teaching Finite State Machines Through Storytelling

629

used animations to visually represent the FSM states

and transitions, aiming to simplify abstract concepts

by contextualizing them in a relatable scenario. CG

group, in contrast, was provided with conventional

written materials and diagrams describing FSM the-

ory and application.

3.3.1 Overview of Video Content and Storyline

The instructional video, titled “Buga and the Mam-

moth: A Finite State Machine Story“, uses a

narrative-based approach to introduce FSM concepts

in an engaging way. The story follows Buga, a prehis-

toric character, as he tames a mammoth by modeling

its behavior using FSM principles. The mammoth’s

behavioral states (Idle, Berserk, Happy, Tamed) and

transitions triggered by Buga’s inputs (e.g., “giving a

ﬂower” or “shouting commands”) illustrate key FSM

elements, such as states, transitions, inputs, and out-

puts. A ﬁgshare package is provided at this link

(Iudean and Vescan, 2025), containing the slide pre-

sentation and the video used in the study, along with

the questionnaires of the study.

Rationale for the Storytelling Approach. The use

of storytelling in STEM education, particularly for

complex topics such as FSMs, is supported by re-

search on narrative-based learning, which shows that

stories can simplify abstract concepts by framing

them in relatable and memorable scenarios (Barchas-

Lichtenstein et al., 2023). In this case, the pre-

historic storyline allows students to grasp the FSM

concepts without needing prior technical knowl-

edge, as they can intuitively follow Buga’s logical

steps to understand each state transition. This ap-

proach leverages the natural human afﬁnity for sto-

ries, which improves engagement and cognitive pro-

cessing (Groshans et al., 2019). Additionally, by mak-

ing Buga’s journey a discovery process, the video po-

sitions FSM concepts as tools for problem-solving, an

approach that aligns with constructivist learning theo-

ries, where knowledge is built through experience and

context.

3.3.2 FSM Concepts Illustrated in the Story

The video employs speciﬁc FSM components to con-

struct a coherent and instructive storyline that builds

students’ understanding incrementally:

States and Transitions: The mammoth’s behavior

is modeled through distinct states, including “Idle”,

“Berserk”, “Happy”, and “Tamed”, each triggered by

Buga’s actions. For instance, poking the mammoth

with a stick causes a transition from Idle to Berserk,

while offering a ﬂower moves it to a Happy state.

These transitions illustrate how FSMs respond pre-

dictably to speciﬁc inputs, a foundational concept in

FSM theory. This cause-effect relationship aids in re-

ducing cognitive load, as it allows learners to focus on

one clear state transition at a time, thereby reinforc-

ing the concept of FSMs as a series of state-speciﬁc

responses to inputs (Mou, 2024). Figure 1 depicts the

beginning of the narrative, when the mammoth is in

an Idle state and the protagonist aims to ﬁnd a clear

path to make it transition towards the Tamed state. In

this process, the stone tables is used for traceability.

Figure 1: States and Transitions.

Inputs and Outputs: Inputs in FSMs drive transi-

tions between states, and the video emphasizes this

by associating each of Buga’s actions (e.g., “giving a

ﬂower” or “using polite language”) with speciﬁc be-

havioral changes in the mammoth. This representa-

tion helps students understand the principle of deter-

ministic state transitions: given a speciﬁc state and

input, the outcome is predictable. By presenting each

input as a conscious choice by Buga, the video en-

courages viewers to see FSMs as tools for strategic

planning and behavioral prediction, linking inputs and

outputs in a way that supports applied understand-

ing of FSM theory (Dochshanov and Tramonti, 2022).

Figure 2 depicts the transition of the mammoth from

the “Idle” state towards the “Happy” state based on a

ﬂower given as input.

Figure 2: Inputs and Outputs.

The storyline subtly introduces the concept of

determinism in FSMs by contrasting predictable

responses (deterministic FSMs) with the mam-

moth’s unpredictable behavior, characteristic of non-

deterministic FSMs. For example, when Buga at-

tempts to tame the mammoth, his inputs sometimes

fail to yield the expected outcomes, highlighting that

non-deterministic FSMs can exhibit multiple possible

behaviors given the same inputs. This concept is valu-

CSEDU 2025 - 17th International Conference on Computer Supported Education

630

able in real-world applications, where systems may

not always respond in a linear manner. This distinc-

tion prepares students to appreciate FSM complexity

and variability in different contexts. Figure 3 depicts

the confusion that the protagonist faces when trying

to backwards engineer a non-deterministic FSM (the

behavioral system of the mammoth) in contrast with

the clarity of a deterministic FSM (the light switch

system).

Figure 3: Non-deterministic and Deterministic FSMs.

3.3.3 Pedagogical Perspectives of the Teaching

Material

Justiﬁcation for the Multimodal Learning Approach.

The video employs a multimodal approach to opti-

mize comprehension by integrating auditory (narra-

tion), visual (animation), and textual (on-screen la-

bels) elements. This design leverages Mayer’s Cog-

nitive Theory of Multimedia Learning, which sug-

gests that combining visual and auditory information

can enhance learning by activating both channels of

the brain, thus reducing cognitive load and increas-

ing retention (Mayer, 2005). The animations in the

video provide visual cues for each state and transi-

tion, while the narration contextualizes these actions,

making the FSM concepts accessible through multi-

ple sensory modalities.

Each scene in the video follows a cycle: Buga per-

forms an action (input), the mammoth reacts (transi-

tion), and a label appears on-screen to reinforce the

state change visually. This structure aligns with mul-

timodal reinforcement, as it helps students process

the material in a layered manner, where each chan-

nel reinforces the others. For example, when Buga

offers a ﬂower to the mammoth, viewers see a visual

cue (ﬂower), hear an explanation of the action’s pur-

pose, and observe the state transition to “Happy”, all

of which contribute to a richer understanding of the

FSM process.

Pedagogical Beneﬁts of the Storytelling and Anima-

tion Integration. By embedding FSM concepts in a

story, the video presents FSM principles in a relat-

able, problem-solving context. Research shows that

learning is more effective when concepts are applied

to realistic scenarios, fostering meaningful under-

standing and long-term retention (Bravo et al., 2021).

Animation enhances this by visualizing state tran-

sitions, beneﬁting visual learners who may struggle

with text-heavy representations. The video also rein-

forces learning through repeated exposure. Buga’s in-

teractions with the mammoth repeatedly demonstrate

FSM elements such as “state”, “transition”, and “in-

put”, reinforcing understanding through varied narra-

tive contexts. This aligns with spaced repetition the-

ories, which support anchoring abstract principles to

memorable actions (Ferri and Auerbach, 2012).

Educational Implications and Justiﬁcation for the Ap-

proach. This narrative-driven, multimodal approach

to FSM education is justiﬁed by its ability to cater to

diverse learning styles, from visual to auditory and

kinesthetic learners, by framing FSMs in a coherent

storyline with corresponding animations and narra-

tion. This design supports engagement and concep-

tual clarity by leveraging multiple learning modali-

ties, each reinforcing FSM principles through struc-

tured storytelling. By making complex concepts like

FSMs accessible within a familiar, imaginative sce-

nario, the video serves as both a cognitive aid and an

engaging educational tool, aligning with the broader

goals of immersive learning in technical disciplines

(Wu et al., 2021).

3.4 Collecting Data by Administering

Questionnaires

The timeline for the education activities is presented

in Figure 4: a pre-test before the activity, then the ded-

icated activity to FSM concept, followed by a post-

test, and ﬁnalizing with a feedback questionnaire spe-

ciﬁc to each group. Details about the administered

questionnaires during the activity, along with the dis-

cussed concepts are provided next.

Study Design. This study followed a mixed design us-

ing both quantitative and qualitative analysis, by us-

ing a controlled experimental design with pre-test and

post-test to assess students’ understanding of FSM

concepts before and after the instructional interven-

tion. The timeline of activities was as follows.

Pre-Test. Both groups completed a pre-test to mea-

sure baseline knowledge of FSM concepts. The test

consisted of multiple-choice and open-ended ques-

tions on FSM fundamentals, covering aspects such as

state transitions, input/output deﬁnitions, and deter-

ministic versus non-deterministic FSMs (Q1-Q7, Q8,

Q9, Q10).

TakeHome-FSM Assignment. Following the pre-test,

both groups were given the same FSM modeling as-

signment, which required them to design a ﬁnite state

Narrative-Driven Learning: Teaching Finite State Machines Through Storytelling

631

Figure 4: Schedule of the Administration of the Question-

naires.

machine for a microwave oven with ﬁve cooking

functions. Students were expected to submit a di-

agram and an explanatory document detailing their

FSM design process. This task aimed to assess practi-

cal application skills in FSM design (Ferri and Auer-

bach, 2012).The assignment was part of the formal

assessment, aiming to address both major goals of

learning as speciﬁed by Mayer (Mayer, 2009), namely

remembering and understanding. The storytelling

video content aims to also help students activate prior

knowledge (Mayer, 2008) in order to integrate it with

the verbal and pictorial models in working memory,

in this way solving problems that were not explicitly

given in the presented material, they must apply what

they learned to a new situation.

Post-Test. After completing the assignment, a post-

test was administered to measure knowledge gains

(Q1-Q7, Q8, Q9, Q10). This test contained sim-

ilar questions to the pre-test, designed to evaluate

changes in comprehension of FSM concepts, partic-

ularly the ability to apply theoretical knowledge in

practical scenarios (Yankovskaya and Yevtushenko,

1997). Bloom’s Taxonomy (Thompson et al., 2008),

which includes the six categories of Knowledge,

Comprehension, Application, Analysis, Synthesis,

and Evaluation, relates directly to our study. By using

open-ended questions based on real-life-like scenar-

ios, the pre-test and post-test evaluated not only stu-

dents’ understanding of FSM concepts but also their

ability to apply, analyze, and combine this knowledge

in practical situations. This approach encouraged

deeper thinking and problem-solving beyond theoret-

ical learning.

Feedback Questionnaire. A feedback questionnaire

was distributed to collect qualitative data on the

students’ perceptions of their learning experience.

Group SG students were speciﬁcally asked about

their engagement with the storytelling approach and

whether they felt it aided their understanding of FSM

concepts. Responses were collected on a 5-point Lik-

ert scale, with additional open-ended questions for

more detailed feedback (FQ1-FQ7 speciﬁc questions

for the SG group). To facilitate the replication of the

study, the questionnaires and the teaching materials

will be available after the review process using a pub-

lic research artifact repository to store the replication

package as mentioned in Section 3.3.1.

Groups and Instructional Differences. The main in-

structional difference between Group CG and Group

SG was the teaching method within a take-home as-

signment activity. Group CG followed a conven-

tional FSM instruction module, while Group SG

was instructed through video storytelling. By com-

paring these groups, we aimed to assess the im-

pact of storytelling-based learning on understanding

FSM principles, examining whether the storytelling

approach increased engagement and comprehension

compared to traditional methods (Hacker and Sitte,

1999).

Data Analysis. Quantitative data was collected from

the pre-tests and post-tests to measure knowledge ac-

quisition, and statistical analysis (MacFarland and

Yates, 2016) was performed to determine any signif-

icant differences in learning outcomes between the

two groups. Qualitative feedback (Patton, 2015) from

the questionnaires provided insights into students’

subjective learning experiences, helping to assess the

perceived effectiveness of the storytelling approach.

4 ANALYSIS OF RESULTS

This section presents the ﬁndings of the study, com-

bining quantitative analysis of pre-test and post-test

scores with qualitative insights from participant feed-

back. The results are structured to address the re-

search questions.

4.1 Measurable Differences in Learning

Outcomes (RQ1)

Discussion of Question Types and Measurability.

The pre-test and post-test evaluations included ques-

tions that assessed both Foundational Knowledge

and Application Skills. Foundational knowledge

was assessed using questions about FSM states, tran-

sitions, and determinism directly measured concep-

tual understanding (questions Q1-Q8), and applica-

CSEDU 2025 - 17th International Conference on Computer Supported Education

632

tion skills were assessed using scenario-based ques-

tions that required participants to apply their knowl-

edge to practical problems, such as designing FSMs

for real-world systems (e.g., microwaves and vending

machines) (questions Q9 and Q10).

To assess measurable differences in learning out-

comes, pre-test and post-test scores were analyzed for

the CG (control) and SG (storytelling) groups. Ta-

ble 1 summarizes the average scores and score im-

provements. It may be noticed that the pre-test for

CG and SG groups are very similar, namely, 8.79 for

CG and 8.75 for the SG group, interpreting that the

two groups had similar knowledge before the activity.

Regarding the post-test, the values for the CG and SG

are again similar, however, even if both decreased, the

value for the SG group decreased with fewer points.

Table 1: Average Pre-Test and Post-Test Scores by Group.

Group Avg. Pre-Test Avg. Post-Test

CG 8.79 8.55

SG 8.75 8.53

The pre-test and post-test results for the two

groups may be also view in Figure 5 and Figure 6.

In both cases, pre-test and post-test, there are ques-

tions that the SG groups answered better than the CG

group: for pre-test with questions Q2, Q5, Q6, and

Q8; for post-test with questions: Q1, Q5, Q6, and Q7.

Figure 5: Results of pre-test (Q1-Q8) for CG and SG

groups.

Figure 6: Results of post-test (Q1-Q8) for CG and SG

groups.

Findings. Based on the pre-test and post-test results,

the following distinctions can be made between the

study groups:

• SG Group. Participants did not show an improve-

ment from the pre-test to post-test, moreover, a

slight decrease in the score was observed (from

8.75 to 8.53.

• CG Group. Participants experienced a slight de-

cline in post-test scores (-0.24 points).

• For both groups (CG and SG) was observed this

slight decline, however, less points for the SG

(only -0.22).

Kruskal-Wallis statistical test (MacFarland and

Yates, 2016) was performed considering various per-

spectives: considering the pre-test of both groups, the

result is not signiﬁcant at p < .05 (p=1), thus empha-

sizing on the fact that the students in the two groups

have similar knowledge about FSM. The same result

is obtained considering the post-test results of the two

groups (p=1). When comparing the results of the pre-

test and post-test for the CG group, the test showed

that the result is not signiﬁcant p < .05 (p=0.59951),

and the same not signiﬁcant result for SG group con-

sidering pre-test and post-test (p=0.95812).

4.2 Improvement in Understanding

FSMs (RQ2)

Open-ended responses (Q9 and Q10) were analyzed

to evaluate participants’ understanding of FSMs.

These responses highlighted practical examples and

insights into abstract concepts like non-determinism.

The themes regarding understanding of FSM by stu-

dents were identiﬁed through qualitative analysis

(Patton, 2015), ﬁrst by open coding and then using

axial coding, use the FSM understanding codes in-

cluded into the thematic analysis. For each group, we

describe the obtained themes individually in Table 2

and further illustrate them with quotes and excerpts

from the data.

Remark: For anonymization purposes, all quotes have

been translated into a non-gender speciﬁc form if pos-

sible or into a gender-speciﬁc form, masculine).

Group CG: provided more technical examples, fo-

cusing on well-deﬁned systems. For example, one

participant described an ATM FSM: “In an ATM,

FSM represents states like card insertion, PIN entry,

and transaction selection, with outcomes like success-

ful completion or insufﬁcient funds.”. Another partic-

ipant used the elevator system as an example: “The

elevator can be in states like idle, moving, or door

open, with transitions triggered by button presses.”

These examples reﬂect the clear, deterministic nature

of FSMs in structured systems.

Group SG: in contrast, gave examples showing more

ﬂexibility in FSMs. One participant mentioned the

Narrative-Driven Learning: Teaching Finite State Machines Through Storytelling

633

Table 2: Themes from Open-Ended Responses.

Theme CG SG

FSM Ap-

plications

(Q9)

ATM behavior,

trafﬁc systems

Trafﬁc lights,

vending ma-

chines, pro-

grammable

devices

Non-

Determinism

(Q10)

Concurrent sys-

tems, complex

models

Flexibility, real-

world scenarios

Engagement Brief, technical

responses

Detailed,

contextual ex-

amples

microwave: “In a microwave, FSM can transition

from idle to multiple states based on user actions, like

selecting cook or stop.” Another example was a traf-

ﬁc light system: “The trafﬁc light can change between

states, but transitions may vary depending on time or

sensor inputs.” These examples highlight Group SG’s

understanding of FSMs in dynamic, real-world sce-

narios, where transitions can be non-deterministic.

Findings. Based on this subtle, yet relevant, trend of

answers in the open-ended questions, the following

conclusions can be formulated:

• FSM Applications. SG participants provided

richer and more relatable examples (e.g., pro-

grammable devices), while CG responses were

concise and technical.

• Understanding Non-Determinism. SG re-

sponses emphasized practical implications, such

as ﬂexibility in handling uncertainty, whereas CG

responses focused on formal deﬁnitions.

• Engagement and Depth. SG participants’ re-

sponses reﬂected greater engagement with the

material, likely due to the narrative context pro-

vided in the video.

4.3 Engagement and Perception of

Learning FSMs (RQ3)

To assess engagement and perception, Likert-scale

feedback from SG participants was analyzed. Table

3 presents average ratings for key questions related to

engagement, clarity, and applicability.

Findings. Based on the self-reﬂective feedback scor-

ing collected from the students, the following conclu-

sions can be formulated:

• Engagement (FQ1). SG participants found the

storytelling video engaging, with an average rat-

ing of 4.41.

Table 3: Average Likert Ratings for SG Engagement and

Perception.

#Q Question Avg.(1–5)

FQ1 The video held my complete at-

tention.

4.41

FQ2 The story was easy to follow and

the analogies were clear.

4.82

FQ3 The story made the subject more

tangible and understandable.

4.65

FQ4 The story enhanced my under-

standing of FSM.

4.41

FQ5 I would like to participate in fu-

ture storytelling-based learning.

4.00

FQ6 I believe I can apply what I

learned in the workplace.

3.53

FQ7 The video changed my perspec-

tive on FSM.

4.00

• Clarity (FQ2, FQ3). Participants rated the story

as clear and relatable, with Q2 and Q3 receiving

the highest scores (4.82 and 4.65, respectively).

• Perception and Application (FQ6, FQ7). While

participants moderately agreed they could ap-

ply the material in their workplace (3.53), they

strongly agreed that the video changed their per-

spective on FSM (4.00).

Qualitative Feedback Themes. Open-ended re-

sponses further highlighted the storytelling ap-

proach’s impact:

• Engagement. Many participants noted that the

narrative format helped maintain their interest and

made the learning experience enjoyable.

• Retention. The use of relatable analogies and ex-

amples was frequently mentioned as aiding mem-

ory retention.

• Challenges. A few participants suggested that the

storytelling could be less effective for highly tech-

nical learners who prefer concise explanations.

4.4 Answers to the Research Questions

Answer RQ1: Storytelling-driven video content did

not leads to measurable improvements in learning

outcomes, as reﬂected in post-test score that de-

creased for SG participants. However, the qualitative

analysis showed better results regarding the applica-

tion skills for the SG group when compared with the

CG group.

Answer RQ2: The storytelling approach improves

understanding of FSMs by providing relatable and de-

tailed contexts. The storytelling approach helped im-

prove students’ understanding of FSMs by providing

CSEDU 2025 - 17th International Conference on Computer Supported Education

634

more relatable and detailed examples. SG partici-

pants showed a better ability to connect FSM con-

cepts to real-life situations, such as trafﬁc lights and

vending machines, which made the concepts easier

to understand. In their feedback, SG students often

used real-world examples to explain FSMs, showing

a deeper understanding. In contrast, CG responses

were more technical and abstract, which might have

made the concepts harder to relate to everyday life.

Answer RQ3: Storytelling enhances engagement

and positively inﬂuences perceptions of FSM learn-

ing, although its applicability may vary based on

learning preferences. SG participants rated the video

highly for attention and clarity, with scores of 4.41

and 4.82 (out of 5 maximum). Many students said

the story kept them interested and helped them under-

stand FSM concepts better. However, some students

felt that the storytelling approach might not work as

well for learners who prefer short, direct explana-

tions. Overall, the feedback showed that storytelling

can make FSMs more engaging and help students see

the subject in a new way.

5 DISCUSSION

This section interprets the results in light of the re-

search questions, focusing on their signiﬁcance for

teaching ﬁnite state machines (FSMs) and computa-

tional models. It also highlights challenges observed

during the study, setting the stage for the subsequent

sections on implications and future work.

5.1 Interpreting the Results

5.1.1 Measurable Learning Outcomes (RQ1)

The study found no signiﬁcant differences in learning

outcomes between the SG (storytelling) and CG (con-

trol) groups, as indicated by similar pre- and post-test

score changes. The SG group’s slight decrease of -

0.22 points suggests that while storytelling may en-

hance engagement, its direct impact on learning out-

comes requires further investigation. Likewise, the

CG group’s drop of -0.24 points points to potential

limitations of traditional instruction in maintaining

student retention. While the results do not strongly

support a clear advantage for storytelling, they align

with multimodal learning theories (Bouchey et al.,

2021), which suggest that integrating visual, verbal,

and contextual elements could help address retention

challenges observed in both groups.

Insights on Question Design and Performance.

The inclusion of foundational and applied questions

provided a well-rounded measure of student perfor-

mance.

Foundational Knowledge: Questions focused on

FSM states and transitions revealed that SG partici-

pants retained these concepts more effectively, likely

due to repeated exposure through the narrative frame-

work.

Applied Knowledge: Scenario-based questions

demonstrated the SG group’s ability to contextual-

ize and apply FSM principles, an outcome that aligns

with prior ﬁndings on the role of storytelling in STEM

education.

These results underscore the need for instructional

content to balance theoretical rigor with engaging,

relatable contexts to maximize learning outcomes.

5.1.2 Enhanced Understanding of FSMs (RQ2)

Qualitative responses highlighted key differences in

how the two groups understood FSMs. SG partici-

pants consistently provided more detailed and contex-

tualized answers to open-ended questions. For exam-

ple, their responses to Q9 included practical and re-

latable examples, such as vending machines and pro-

grammable devices, whereas CG participants focused

on abstract descriptions like ATM behavior.

In Q10, which explored non-determinism, SG par-

ticipants elaborated on the ﬂexibility and real-world

implications of FSM models. These ﬁndings suggest

that storytelling not only improves comprehension but

also enhances the ability to articulate and apply com-

plex concepts. The narrative approach likely bridges

the gap between theoretical abstraction and practical

relevance, making FSMs more accessible and engag-

ing.

5.1.3 Engagement and Perception (RQ3)

Feedback data provided compelling evidence of the

storytelling approach’s impact on engagement and

perception.

Engagement. Participants from Group SG rated the

video highly for attention and clarity, with average

scores of 4.41 and 4.82 (out of 5 maximum), respec-

tively. One student mentioned, “It was an interesting

experience to follow and analyze each step”, and an-

other said, “The video kept me engaged.” These re-

sponses highlight the storytelling approach’s ability

to actively capture students’ attention, aligning with

active learning theories that emphasize emotional and

cognitive engagement.

Perception. The majority of participants expressed

interest in future storytelling-based learning, rating

Narrative-Driven Learning: Teaching Finite State Machines Through Storytelling

635

it 4.00. One participant noted, “I would like more

videos like this”, while another shared, “It was a fun

and creative way of learning something.” These state-

ments reﬂect the positive impact of storytelling on

students’ perceptions, indicating that it changed their

perspective on FSMs and made learning more enjoy-

able.

Applicability. Regarding the workplace applicability

of FSM concepts, participants rated it 3.53. One re-

sponse highlighted, “FSM is a useful tool that can

help us develop efﬁcient and bug-free systems.” An-

other participant mentioned, “Learning more about

FSM and trying to understand a situation better by

creating a story around it.” These insights suggest

that while students acknowledged the workplace rele-

vance of FSMs, they also recognized how storytelling

helped solidify their understanding of these concepts

by anchoring them in relatable contexts.

Negative Feedback. Despite these positives, some

participants pointed out the potential drawbacks of

storytelling. One student shared, “It was an inter-

esting experiment, although it didn’t do much for

me to better understand the subject discussed”, while

another commented, ”This topic can be presented

in a few minutes, and the storytelling felt time-

consuming.” These responses indicate that for some,

the storytelling approach may not be as effective for

quickly grasping technical material. Despite these

strengths, some feedback pointed to the potential lim-

itations of storytelling, such as its perceived inefﬁ-

ciency for highly technical learners who prefer con-

cise explanations.

5.2 Emerging Challenges and

Considerations

While the results afﬁrm the beneﬁts of narrative-

driven content, several challenges emerged:

• Balancing Engagement with Rigor. The sto-

rytelling approach successfully engaged partici-

pants but may require supplemental technical ex-

planations to satisfy diverse learner preferences.

• Content-Speciﬁc Effectiveness. The observed

beneﬁts were speciﬁc to FSM concepts, raising

questions about the generalisability of storytelling

to other computational topics.

• Assessing Long-Term Impact. The study focused

on immediate learning outcomes. Longitudinal

studies are needed to evaluate the lasting effects

of storytelling on retention and application.

6 IMPLICATIONS AND FUTURE

WORK

The ﬁndings of this study provide important insights

into the potential of storytelling as a teaching method

in computational ﬁelds. This section explores the im-

plications for FSM and computational model instruc-

tion, practical recommendations for educators, and di-

rections for future research.

6.1 Implications for Teaching FSM and

Computational Models

Engagement-Driven Learning. The storytelling ap-

proach demonstrated its ability to captivate learn-

ers and foster deeper understanding of FSM con-

cepts. This emphasizes the need to integrate engag-

ing, narrative-based materials into traditionally tech-

nical curricula.

Contextual Learning. By embedding abstract con-

cepts in relatable scenarios, storytelling helps bridge

the gap between theoretical knowledge and practi-

cal application. This approach could enhance student

outcomes in other computational topics, such as au-

tomata theory and control systems.

Diverse Learning Preferences. While storytelling was

effective for most participants, feedback highlighted

the need to accommodate varying preferences. Com-

bining narrative elements with direct, technical in-

struction could better serve learners who prefer con-

cise and factual content.

6.2 Recommendations for Teachers

Incorporate Multimodal Content. Educators should

leverage multimedia storytelling tools, including ani-

mations and real-world analogies, to enhance student

engagement.

Balance Narrative with Technical Depth. To ad-

dress the needs of diverse learners (Fleming and

Mills, 1992), combine storytelling with supplemen-

tary materials such as detailed diagrams, mathemati-

cal proofs, or step-by-step walkthroughs.

Iterative Feedback Loops. Actively collect and ana-

lyze student feedback to reﬁne storytelling methods

and tailor instructional content to evolving needs.

6.3 Future Research Directions

Building on the ﬁndings of this study, future research

should focus on the following.

Long-Term Retention. Investigate the lasting effects

of storytelling on knowledge retention and practical

application, using longitudinal studies.

CSEDU 2025 - 17th International Conference on Computer Supported Education

636

Generalizability Across Topics. Explore the effec-

tiveness of narrative-based teaching in other areas of

computational science, such as algorithm design or

data structures.

Personalized Learning. Assess how storytelling im-

pacts different learner proﬁles, considering factors

such as prior knowledge, cultural background, and in-

dividual preferences.

7 THREATS TO VALIDITY

Several risks to the validity of the ﬁndings were con-

sidered and addressed. To minimize bias, students

were randomly assigned to either the CG or SG group,

and pre-tests ensured similar baseline knowledge of

FSM concepts. External factors like student motiva-

tion were controlled by maintaining consistent test-

ing conditions and providing equal attention to both

groups. Although tests were not anonymous, stu-

dents were encouraged to provide honest feedback by

stressing the importance of candid responses. Addi-

tionally, recognizing that some students may prefer

traditional instruction, both quantitative and qualita-

tive data were used to capture diverse learning prefer-

ences and evaluate the storytelling approach.

8 CONCLUSIONS

This study examined the impact of storytelling-based

video content on the learning of ﬁnite state machines

(FSMs) compared to traditional instructional meth-

ods. Key ﬁndings include the following:

Learning Outcomes. There was no signiﬁcant im-

provement in post-test scores, indicating that story-

telling alone may not directly enhance learning per-

formance.

Understanding. Storytelling helped participants con-

textualize abstract FSM concepts and relate them to

real-world applications.

Engagement. Feedback from Likert-scale ratings and

open-ended responses conﬁrmed increased attention

and positive perceptions of learning FSMs.

The study suggests that the main beneﬁt of story-

telling is its ability to contextualize abstract concepts

rather than directly improving problem-solving skills,

emphasizing the need for more research to balance

narrative-driven content with technical rigor and ex-

plore its long-term effects.

ACKNOWLEDGEMENTS

The publication of this article was partially supported

by the 2024 Development Fund of the Babes-Bolyai

University.

REFERENCES

Angel-Fernandez, J. M. and Vincze, M. (2018). Intro-

ducing storytelling to educational robotic activities.

In 2018 IEEE Global Engineering Education Confer-

ence (EDUCON), pages 608–615. IEEE.

Barchas-Lichtenstein, J., Sherman, M., Voiklis, J., and

Clapman, L. Science through storytelling or story-

telling about science? identifying cognitive task de-

mands and expert strategies in cross-curricular stem

education. In Frontiers in Education, volume 8, page

1279861. Frontiers.

Barchas-Lichtenstein, J., Sherman, M., Voiklis, J., and

Clapman, L. (2023). Science through storytelling or

storytelling about science? identifying cognitive task

demands and expert strategies in cross-curricular stem

education. In Frontiers in Education, volume 8, page

1279861. Frontiers Media SA.

Berreth, T., Polyak, E., and Fitzgerald, P. (2020). Story-go-

round: Augmented reality storytelling in the multidis-

ciplinary classroom. Cited by: 1.

Bouchey, B., Castek, J., and Thygeson, J. (2021). Multi-

modal learning. Innovative Learning Environments in

STEM Higher Education: Opportunities, Challenges,

and Looking Forward, pages 35–54.

Bravo, F. A., Hurtado, J. A., and Gonz

alez, E. (2021). Using

robots with storytelling and drama activities in science

education. Education Sciences, 11(7):329.

Collins, S., Steele, T., and Nelson, M. (2023). Storytelling

as pedagogy: The power of chemistry stories as a tool

for classroom engagement. Journal of Chemical Edu-

cation, 100(7):2664–2672.

Dekeyser, J.-L. and Aljendi, A. S. (2015). Adopting new

learning strategies for computer architecture in higher

education: case study: building the s3 microproces-

sor in 24 hours. In Proceedings of the Workshop on

Computer Architecture Education, pages 1–8.

Dengel, A. (2018). Seeking the treasures of theoretical com-

puter science education: Towards educational virtual

reality for the visualization of ﬁnite state machines. In

2018 IEEE international conference on teaching, as-

sessment, and learning for engineering (TALE), pages

1107–1112. IEEE.

Dochshanov, A. and Tramonti, M. (2022). Digital story-

telling and games to stem skills development. In ED-

ULEARN22 Proceedings, pages 8503–8511. IATED.

Durak, H. Y. (2018). Digital story design activities used

for teaching programming effect on learning of pro-

gramming concepts, programming self-efﬁcacy, and

participation and analysis of student experiences.

JOURNAL OF COMPUTER ASSISTED LEARNING,

34(6):740–752.

Narrative-Driven Learning: Teaching Finite State Machines Through Storytelling

637

Ferri, B. H. and Auerbach, J. L. (2012). A portable ﬁnite

state machine module experiment for in-class use in a

lecture-based course. In 2012 ASEE Annual Confer-

ence & Exposition, pages 25–89.

Fleming, N. D. and Mills, C. (1992). Not another inven-

tory, rather a catalyst for reﬂection. To Improve the

Academy, pages 137–155.

Groshans, G., Mikhailova, E., Post, C., Schlautman, M.,

Carbajales-Dale, P., and Payne, K. (2019). Digital

story map learning for stem disciplines. Education

Sciences, 9(2):75.

Hacker, C. H. and Sitte, R. (1999). Implementing ﬁnite

state machines in a computer-based teaching system.

In Education in Microelectronics and MEMS, volume

3894, pages 110–117. SPIE.

Henry, J., Dumas, B., Vescan, A., and Pasca, A. M. (2022).

Student misconceptions about ﬁnite state machines:

Identify them in order to create a concept inventory. In

Proceedings of the 4th International Workshop on Ed-

ucation through Advanced Software Engineering and

Artiﬁcial Intelligence, pages 2–9.

Heymann, R. and Greeff, J. J. (2018). Designing and de-

veloping a narrative driven serious game for teaching

information theory. In 2018 IEEE Global Engineering

Education Conference (EDUCON), pages 489–496.

IEEE.

Hill, S. and Grinnell, C. (2014). Using digital storytelling

with infographics in stem professional writing peda-

gogy. In 2014 IEEE International Professional Com-

munication Conference (IPCC), pages 1–7.

Iudean, B. and Vescan, A. (2025). Narrative-

driven learning: Teaching ﬁnite

state machines through storytelling.

https://doi.org/10.6084/m9.ﬁgshare.28436603.v1.

Joseph, S., Schumm, M., Rummel, O., Soska, A., Reschke,

M., Mottok, J., Niemetz, M., and Schroll-Decker,

I. (2013). Teaching ﬁnite state machines with case

method and role play. In 2013 IEEE Global Engineer-

ing Education Conference (EDUCON), pages 1305–

1312. IEEE.

Karttunen, L. (2000). Applications of ﬁnite-state transduc-

ers in natural language processing. In International

Conference on Implementation and Application of Au-

tomata, pages 34–46. Springer.

Landrum, R. E., Brakke, K., and McCarthy, M. A. (2019).

The pedagogical power of storytelling. Scholarship of

Teaching and Learning in Psychology, 5(3):247.

Lee, D. and Yannakakis, M. (1996). Principles and methods

of testing ﬁnite state machines-a survey. Proceedings

of the IEEE, 84(8):1090–1123.

MacFarland, T. W. and Yates, J. M. (2016). Kruskal Wallis

H-Test for Oneway Analysis of Variance (ANOVA) by

Ranks. Springer International Publishing, Cham.

Mayer, R. E. (2005). The Cambridge handbook of multime-

dia learning. Cambridge university press.

Mayer, R. E. (2008). Applying the science of learn-

ing: Evidence-based principles for the design of

multimedia instruction. American Psychologist,

63(8):760–769.

Mayer, R. E. (2009). Multimedia learning (2nd ed.). Cam-

bridge university press.

Min, W., Mott, B., Park, K., Taylor, S., Akram, B., Wiebe,

E., Boyer, K. E., and Lester, J. (2020). Promoting

computer science learning with block-based program-

ming and narrative-centered gameplay. In 2020 IEEE

Conference on Games (CoG), pages 654–657. IEEE.

Mou, T.-Y. (2024). The practice of visual storytelling in

stem: Inﬂuence of creative thinking training on design

students’ creative self-efﬁcacy and motivation. Think-

ing Skills and Creativity, 51:101459.

Nadeem, M., Lal, M., Cen, J., and Sharsheer, M. (2022).

Ar4fsm: Mobile augmented reality application in en-

gineering education for ﬁnite-state machine under-

standing. Education Sciences, 12(8):555.

Parham-Mocello, J., Ernst, S., Erwig, M., Shellhammer, L.,

and Dominguez, E. (2019). Story programming: Ex-

plaining computer science before coding. In Proceed-

ings of the 50th ACM Technical Symposium on Com-

puter Science Education, pages 379–385.

Patton, M. Q. (2015). Qualitative Research & Evaluation

Methods: Integrating Theory and Practice. Sage, Los

Angeles, CA, fourth edition.

Radojevic, I. and Salcic, Z. (2011). Embedded systems de-

sign based on formal models of computation. Springer

Science & Business Media.

Resnyansky, D. (2020). Augmented reality-supported tan-

gible gamiﬁcation for debugging learning. In 2020

IEEE International Conference on Teaching, Assess-

ment, and Learning for Engineering (TALE), pages

377–383. IEEE.

Segel, E. and Heer, J. (2010). Narrative visualization:

Telling stories with data. IEEE transactions on visu-

alization and computer graphics, 16(6):1139–1148.

Thompson, E., Luxton-Reilly, A., Whalley, J. L., Hu, M.,

and Robbins, P. (2008). Bloom’s taxonomy for cs as-

sessment. In Proceedings of the tenth conference on

Australasian computing education-Volume 78, pages

155–161.

Wu, C., Tang, Y., Tsang, Y., and Chau, K. (2021). Im-

mersive learning design for technology education: A

soft systems methodology. Frontiers in Psychology,

12:745295.

Wu, J. and Chen, D.-T. V. (2020). A systematic review of

educational digital storytelling. Computers & Educa-

tion, 147:103786.

Yankovskaya, A. and Yevtushenko, N. (1997). Finite state

machine (fsm)–based knowledge representation in a

computer tutoring system. New Media and Telem-

atic Technologies for Education in Eastern European

Countries, pages 67–74.

CSEDU 2025 - 17th International Conference on Computer Supported Education

638