=Paper=
{{Paper
|id=Vol-3879/AIxEDU2024_paper_5
|storemode=property
|title=Enhancing Instructional Design: The Impact of CONALI Ontology and ChatGPT in Primary Education Training
|pdfUrl=https://ceur-ws.org/Vol-3879/AIxEDU2024_paper_5.pdf
|volume=Vol-3879
|authors=Dario Lombardi,Luigi Traetta,Antonio Maffei,Primož Podržaj
|dblpUrl=https://dblp.org/rec/conf/aixedu/LombardiTMP24
}}
==Enhancing Instructional Design: The Impact of CONALI Ontology and ChatGPT in Primary Education Training==
https://ceur-ws.org/Vol-3879/AIxEDU2024_paper_5.pdf
Enhancing Instructional Design: The Impact of CONALI
Ontology and ChatGPT in Primary Education Training
Dario Lombardi1,†, Luigi Traetta1,†, Antonio Maffei2, †, and Primož Podržaj3, †
1
University of Foggia, Arpi Street, 176 Foggia, Italy
2
KTH Royal Institute of Technology, Brinellvägen 8, Stockholm SE-100 44, Sweden
3
Univerza v Ljubljani, Fakulteta za strojništvo, Aškerčeva 6, City, Ljubljana, Slovenija
Abstract
The integration of Artificial Intelligence (AI) in education is becoming increasingly crucial [1], especially
for the preparation of future educators. Despite these requirements, there is a paucity of training courses
for educators in the use of AI systems [2]. AI-driven tools play a crucial role in personalizing, simplifying,
and innovating educational pathways at all levels [3]. his study examines the combined application of the
CONALI Ontology [4][5] and ChatGPT in supporting the instructional design process among 110 students
enrolled in the Primary Education Sciences Laboratory at the University of Foggia. The goal was to evaluate
how AI can enhance the creation of Learning Units (LUs) by streamlining design processes and promoting
educational innovation. Participants received training in using the CONALI framework, which emphasizes
the identification of learning objectives, activities, and assessment methods through constructive alignment
[5]. After this initial instruction, students engaged in the design of their own LUs, integrating ChatGPT as
a support tool. At the conclusion of the exercises, validated questionnaires were administered to assess
participants' perceptions of the implementation of the CONALI ontological framework in combination with
generative AI systems ChatGPT. The goal was to explore the advantages, disadvantages, and emerging
challenges that must be addressed in the training of future educators, specifically regarding the integration
of AI in the design of Learning Units and its implications for both their current and future educational
practices. Results demonstrated that the CONALI Ontology was instrumental in helping students articulate
SMART objectives, resulting in clearer, more focused instructional design. Moreover, the integration of
ChatGPT was perceived as significantly improving the efficiency and creativity of the design process,
enabling students to quickly generate ideas and refine their projects. This study highlights the
transformative potential of AI in conjunction with structured ontological frameworks to enrich
instructional design, ultimately enhancing the skills and competencies of future educators in a technology-
enhanced learning environment.
Keywords
AIEd; Instructional Design; Teacher Training,1
1. Introduction
Artificial Intelligence (AI) systems are becoming increasingly integral to education, especially in
preparing future educators to navigate a rapidly evolving technological landscape. AI refers to
software and hardware designed by humans to act autonomously in the physical or digital domain
by perceiving their environment through data acquisition, interpreting the structured or
unstructured data, reasoning on the knowledge derived, and deciding the best actions to achieve a
specific goal [6].
In the educational sphere, AI offers significant opportunities to enhance teaching methodologies,
streamline administrative tasks, and provide personalized learning experiences [7]. Recent studies
have shown that AI's application in education can improve teachers’ work efficiency, support
professional development, and foster a positive attitude towards technology in teaching contexts
[8][9]. However, despite the clear advantages, there remains a paucity of research and training
∗
Corresponding author.
†
These authors contributed equally.
dario.lombardi@unifg.it (D. Lombardi); luigi.traetta@unifg.it (L. Traetta); maffei@kth.se (A. Maffei);
primoz.podrzaj@fs.uni-lj.si (P. Podržaj)
0000-0002-3405-6016 (D. Lombardi); 0000-0001-7785-9751 (L. Traetta); 0000-0003-4470-6243 (A. Maffei) 0000-0002-9932-
9521 (P. Podržaj)
© 2024 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).
CEUR
ceur-ws.org
Workshop ISSN 1613-0073
Proceedings
initiatives aimed at equipping future educators with the skills necessary to integrate AI systems into
their pedagogical practices effectively [10].
This study aims to address this gap by examining the combined application of the CONALI Ontology
and AI-driven tools, such as ChatGPT, in supporting the instructional design process. The CONALI
Ontology, based on constructive alignment theory, ensures coherence between learning objectives,
activities, and assessments, making it a valuable tool in educational contexts [11][4][5].
When integrated with AI systems like ChatGPT, which assist in generating, refining, and
personalizing content, these tools hold significant potential for transforming how educators design
and implement instructional units. The importance of such AI-driven tools lies not only in their
ability to streamline complex tasks but also in their capacity to innovate educational pathways at all
levels [12].
In this study, we examine how AI, particularly ChatGPT, can enhance the creation of Learning Units
(LUs) by simplifying the design process and promoting educational innovation. A total of 110
students enrolled in the Primary Education Sciences Laboratory at the University of Foggia
participated in this research, receiving training on how to use the CONALI framework in conjunction
with AI tools.
This training focused on identifying learning objectives (LO) and Educational Goal Verb (EGV),
designing appropriate Teaching and Learning Activities (TLA), and determining assessment methods
that align with the principles of Constructive Alignment (CA) [13].
Upon completing the training, students applied these principles to design their own Learning Units
(LUs), initially with the assistance of the CONALI Ontology, and later with the support provided by
ChatGPT. This process allowed them to explore both the advantages and challenges of integrating
AI into educational design.
The potential of AI in enhancing instructional design lies in its ability to personalize learning
experiences, simplify repetitive or administrative tasks, and foster creativity in the development of
educational materials. For instance, previous research has demonstrated that AI-based systems can
provide personalized learning environments that cater to individual students' needs, thus promoting
more inclusive and effective learning experiences [14]. This is particularly relevant in the context of
designing Learning Units, where the combination of the CONALI Ontology’s structured approach
with ChatGPT’s generative capabilities can help future educators articulate SMART objectives more
clearly, leading to more focused and effective instructional plans [15].
Despite the benefits, there are also notable challenges in the integration of AI in education. One of
the key issues is the lack of comprehensive training for educators in the effective use of AI tools [8].
While AI can automate many aspects of instructional design, such as generating lesson ideas or
providing feedback on assessments, educators still need to develop a deep understanding of these
technologies to use them effectively. This requires not only technical proficiency but also an
understanding of the pedagogical implications of AI use in classrooms. The results of this study
highlight the importance of providing future educators with adequate training in both AI tools and
pedagogical strategies to ensure they can harness the full potential of these technologies in their
teaching practices.
Furthermore, the findings from the questionnaires administered at the conclusion of the study
revealed that students generally perceived the integration of ChatGPT and the CONALI Ontology as
beneficial, particularly in terms of increasing the efficiency and creativity of their design processes.
However, they also identified several challenges, including the need for more training on how to
critically assess and refine AI-generated content to ensure its pedagogical relevance and quality [7].
As AI continues to play a growing role in education, it is crucial to address these challenges to
maximize its potential benefits.
The integration of AI in education, particularly through tools like ChatGPT and structured
frameworks such as the CONALI Ontology, holds significant promise for enhancing the instructional
design process and preparing future educators for the demands of a technology-enhanced learning
environment. By streamlining the design of Learning Units and promoting educational innovation,
these tools can help educators create more personalized, inclusive, and effective learning
experiences. However, to fully realize the benefits of AI in education, it is essential to address the
current gaps in training and provide educators with the skills they need to critically engage with AI
technologies in their professional practice [16].
2. Methodology
The aim of this study was to evaluate how the integration of the CONALI Ontology and ChatGPT
could enhance the instructional design process among students in Primary Education Sciences.
Conducted with 110 students enrolled in the Primary Education Sciences Laboratory at the
University of Foggia, the study focused on guiding students through the design of Learning Units
(LUs) by emphasizing the identification of educational objectives, activities, and assessment methods
using the CONALI Ontology, which is based on Constructive Alignment (CA) principles [11]. The
research aimed to capture both perceived benefits and challenges associated with using AI tools like
ChatGPT in combination with structured ontological frameworks to support instructional design.
2.1. Study design and participant training
To introduce participants to the concepts underpinning the study, a one-hour lecture was
conducted on the principles of Constructive Alignment and the CONALI Ontology. This session
highlighted the importance of creating SMART (Specific, Measurable, Attainable, Realistic, and
Time-bound) objectives [17][18] and aligning these with Teaching and Learning Activities (TLAs)
and Assessment Tasks (ATs) to achieve coherent instructional design.
Following this, students were randomly divided into two groups to engage in practical activities:
• Experimental Group: Used both the CONALI Ontology and ChatGPT for instructional
design.
• Control Group: Used only the CONALI Ontology for the same tasks.
2.2. Intervention phases
2.2.1. Phase 1: Initial design without AI support
In the first phase, both groups participated in a frontal lecture specifically on the CONALI Ontology,
followed by an exercise in which they independently designed their LUs without using AI tools. This
phase served to familiarize students with the instructional design process based solely on the
ontology framework. Students were allotted 45 minutes to complete this task, followed by a 30-
minute group discussion to address common challenges and clarify any misunderstandings.
2.2.2. Phase 2: Introduction and use of ChatGPT
After completing the initial task, a detailed 30-minute training session introduced ChatGPT as a
generative AI tool to support instructional design. This training emphasized how ChatGPT could
assist in brainstorming ideas, refining educational objectives, and suggesting suitable activities and
assessments.
For the subsequent design exercise, the experimental group was instructed to use ChatGPT in
combination with the CONALI Ontology, while the control group continued to work with only the
ontology. Both groups were given 60 minutes to design new Learning Units (LUs), with the
experimental group encouraged to use ChatGPT to streamline idea generation and optimize their
instructional plans. After this exercise, a classroom discussion was conducted to gather qualitative
feedback from both groups and to identify any notable differences in their approaches and
experiences with instructional design.
2.2.3. Data collection and evaluation
At the conclusion of the exercises, a comprehensive questionnaire was administered to all
participants to capture their perceptions of the instructional design process, focusing on ChatGPT’s
impact on efficiency, creativity, and clarity. To ensure comparability:
• The control group questionnaire focused solely on their experience using the CONALI
Ontology.
• The experimental group questionnaire specifically explored how ChatGPT contributed to
their ability to identify educational objectives, TLAs, and appropriate assessment methods.
The structured design of this study allowed for a direct comparison between the two groups, enabling
an in-depth analysis of the unique contributions of ChatGPT to instructional design. The data
collected from these questionnaires provided valuable insights into how future educators perceive
and benefit from AI-driven tools in planning personalized and innovative learning experiences.
2.2.4. Limitations and considerations for future research
One limitation of this study was the lack of a pure control group that did not use either tool, which
might have helped isolate the distinct effects of ChatGPT and the CONALI Ontology. Future studies
should consider incorporating a baseline group to better quantify the individual impacts of each tool.
Additionally, the study relied on self-reported data, which, while insightful, would benefit from being
supplemented by expert evaluations or rubric-based assessments to objectively measure the quality
of the produced Learning Units.
2.3. Intervention schedule
The intervention schedule was structured as follows:
1. Introduction to Constructive Alignment and the CONALI Ontology (1 hour): Presentation on
aligning educational objectives and Educational Goal Verbs (EGVs) with TLAs and ATs.
2. Exercise 1 (45 minutes): Students independently designed LUs using only the CONALI
Ontology.
3. Group Discussion (30 minutes): Addressing challenges and experiences from the initial
exercise.
4. Introduction to ChatGPT (30 minutes): Training on using ChatGPT in educational design.
5. Exercise 2 (60 minutes): The experimental group used ChatGPT alongside the CONALI
Ontology, while the control group used only the ontology to design new LUs.
6. Final Discussion and Assessment (20 minutes): Questionnaire to evaluate participants'
perceptions and experiences.
7. The questionnaire responses, which include detailed analyses of participants' perceptions
regarding the integration of AI tools in instructional planning, are discussed in the “Results”
section.
3. Results
The results of this study are organized into two main sections, each focusing on the impact of
either the CONALI framework or ChatGPT in enhancing students' instructional design skills. Both
sections address quantitative data from the questionnaire responses, supplemented by qualitative
insights from classroom discussions. Each figure is referenced with accompanying captions for
clarity and relevance.
3.1. Design results after support from the CONALI Ontology
Download The application of the CONALI Ontology, specifically structured around
Constructive Alignment (CA), proved to significantly enhance students' confidence and clarity
in designing Learning Units (LUs). Key findings are organized as follows:
3.1.1. Confidence levels and understanding of CA principles
Download A notable increase in student confidence and understanding was observed. Prior to
training, only 27.3% of students felt confident in their ability to design a UDA independently.
Following the structured support of the CONALI Ontology, this figure rose to 85%, indicating a
substantial increase in self-efficacy (Fig. 1). Moreover, familiarity with CA principles increased
from 27.3% to 83.6%, demonstrating that the ontology effectively communicated complex concepts
(Fig. 2).
Figure 1: Percentage of students confident in designing a UDA pre-CONALI Ontology training.
Figure 2: Student understanding of CA principles after the CONALI Ontology training.
3.1.2. Application of knowledge and instructor support
The questionnaire responses revealed that 92% of students felt that practical examples in the CONALI
training bridged theoretical and practical aspects of instructional design, directly applicable to future
UDA planning (Fig. 3). Additionally, 88% reported that the instructional support provided during the
sessions was crucial in enabling them to apply these concepts independently. This indicates that the
ontology, combined with structured support, facilitated a productive learning environment.
Figure 3: Percentage of students finding practical examples in the training applicable to future
UDA design.
3.1.3. Collaborative learning environment
The training’s collaborative aspect also had a positive impact, with approximately 80% of students
indicating that the course encouraged idea exchange and teamwork, which they found beneficial to
their learning process (Fig. 4). Many students attributed their understanding of complex CA concepts
to these collaborative sessions, as supported by the high satisfaction rate (85%) regarding course
content and interactive format (Fig. 5).
Figure 4: Percentage of students who found collaboration beneficial to their learning.
Figure 5: Student satisfaction with course content and interactive approach to teaching CA
principles.
3.2. Design results after support from ChatGPT
The integration of ChatGPT as a supplemental design tool yielded a generally favorable response,
with students noting improvements in creativity, efficiency, and task clarity during UDA planning.
3.2.1. Integration into teaching practice, task understanding and satisfaction
An impressive 95% expressed intentions to apply constructive alignment principles when designing
their future UDAs. This commitment suggests that students not only learned but also valued the
knowledge gained during the course (Fig.6)
Approximately 90% of students reported a clear understanding of the task requirements, with
ChatGPT effectively helping them delineate objectives and design requirements (Fig. 7).
Furthermore, 85% expressed satisfaction with their performance in creating UDAs with ChatGPT’s
support, indicating its positive impact on the overall design process.
Figure 6: Percentage of students intending to use the CA to design future UDAs
Figure 7: Percentage of students satisfied with their UDA design process using ChatGPT.
3.2.2. Time efficiency and accessibility
In terms of time efficiency, students reported an average completion time of two hours for UDA
design, noting that ChatGPT expedited brainstorming and helped them organize their ideas more
effectively. Most students accessed the free version of ChatGPT, which underlines its accessibility as
an educational resource.
3.2.3. Impact on objective and organization
Approximately 75% of students acknowledged that ChatGPT facilitated the generation of
instructional objectives and alignment with suitable assessment systems, a critical component in
instructional design (Fig. 8). Additionally, 80% reported that ChatGPT assisted in structuring their
ideas, making it easier to organize UDAs coherently (Fig. 9), with 70% indicating that the AI’s
suggestions encouraged them to pursue more refined strategies and assessments (Fig. 10).
Figure 8: Percentage of students reporting improved objective generation using ChatGPT.
Figure 9: Students’ perception of ChatGPT’s aid in organizing their UDA structure.
Figure 10: Percentage of students pursuing refined objectives and assessments with ChatGPT’s
support.
3.2.4. Limitations and critical thinking challenges
Despite these positive outcomes, some limitations were reported. Approximately 60% of students
noted that ChatGPT’s suggestions sometimes lacked contextual relevance, necessitating a careful
evaluation of AI-generated content to ensure pedagogical alignment (Fig. 11). Additionally, 70%
emphasized the importance of critical thinking in validating ChatGPT’s output to maintain
instructional accuracy and relevance.
Figure 11: Students emphasizing the need for critical thinking when using ChatGPT.
3.2.5. Technical limitations and feedback on AI accuracy
Technical issues were also highlighted, particularly by students using the free version, who reported
slow response times and occasional service unavailability, which disrupted their workflow.
Moreover, 75% rated ChatGPT’s accuracy as satisfactory, though some students noted that certain
responses were generic and did not fully address their instructional goals (Fig. 12).
Figure 12: Percentage of students rating ChatGPT’s accuracy as satisfactory or higher.
3.2.6. Technical limitations and feedback on AI accuracy
Reflecting on their experience, 85% of students considered ChatGPT a useful tool for UDA design
and expressed an intent to continue using it as a supplementary resource in future instructional
planning (Fig. 13). However, many cautioned against excessive reliance on AI tools, stressing the
need for human oversight to preserve quality and depth in educational planning.
Figure 13: Percentage of students who consider ChatGPT valuable and intend to use it in future
projects.
3.3. Summary of findings
In conclusion, both the CONALI Ontology and ChatGPT contributed to the enhancement of
instructional design skills among students, each providing distinct but complementary benefits.
While the CONALI framework improved foundational knowledge of CA principles and practical
application, ChatGPT offered efficiencies in creativity and organization. However, the need for
critical evaluation of AI output remains essential, as does the risk of over-reliance on AI. These
results underscore the potential for combining ontological frameworks with AI tools to enhance
instructional design, provided that students are trained to engage critically and independently with
AI-generated content.
4. Discussion
The integration of artificial intelligence (AI) tools, such as ChatGPT, alongside the Constructive
Alignment Ontology (CONALI), represents a promising approach for designing Units of Didactic
Activities (UDAs) in primary education training. This section explores the benefits, limitations,
and challenges associated with the adoption of ChatGPT and CONALI, along with future
implications for educational practices.
4.1. Summary of findings
The findings of this study highlight several distinct benefits stemming from the combined use of
ChatGPT and CONALI in UDA design. Firstly, the structured support provided by CONALI
significantly improved students' comprehension and application of Constructive Alignment (CA)
principles. Approximately 90% of participants reported increased clarity in designing SMART
(Specific, Measurable, Attainable, Realistic, and Time-bound) objectives, which directly supported
their ability to align learning outcomes with targeted instructional strategies and assessments. This
reflects an essential goal of CA by ensuring coherence between objectives, teaching methodologies,
and assessments, a principle foundational to effective instructional design.
The introduction of ChatGPT further augmented these benefits by enhancing efficiency and
creativity. As noted by 85% of students, ChatGPT helped streamline the ideation process, allowing
them to focus on the depth and quality of their instructional units rather than getting bogged down
by initial brainstorming challenges. In particular, 75% of students highlighted ChatGPT’s
effectiveness in facilitating the generation of instructional objectives and assessment frameworks.
This suggests that ChatGPT’s generative capacity is especially valuable during the preliminary
design phase, helping students concentrate on more nuanced and pedagogically aligned choices (Fig.
7).
The observed time efficiency also represents a crucial benefit in the educational context, where
both students and instructors manage demanding schedules. As most students reported completing
their UDA designs more quickly than anticipated, ChatGPT demonstrated potential as a time-saving
tool, supporting its role as a complementary resource for educators and students alike (Fig. 8).
Moreover, students valued the accessibility of ChatGPT, with most relying on the free version,
highlighting the tool’s potential as a scalable and inclusive educational asset.
4.2. Limitation and challenges in AI integration
Despite these positive outcomes, integrating AI in UDA design introduces notable limitations and
challenges. A primary concern is the need for critical thinking when using AI-generated content.
While 70% of students found ChatGPT’s suggestions useful, many noted that personal judgment
remained crucial to ensure the relevance and accuracy of AI-generated ideas. ChatGPT’s suggestions,
though helpful, require students to interpret and adapt them thoughtfully, underscoring a critical
challenge: while AI can generate content ideas, it cannot replace human insight, which is necessary
for aligning instructional design with specific educational contexts.
Technical limitations also impacted the user experience. Several students using the free version of
ChatGPT encountered slow response times, which affected the fluidity of their design process (Fig.
9). Furthermore, approximately 60% of students reported that some AI-generated responses lacked
specificity, indicating that the tool’s generative capabilities may sometimes produce generic or
contextually inadequate responses. This limitation suggests that while ChatGPT can support UDA
design, it requires refinement to better address diverse and specific educational needs.
An additional challenge is the risk of over-reliance on AI. As reported by some students, ChatGPT’s
supportive role could inadvertently lead to dependence, with students potentially prioritizing AI-
generated ideas over personal critical analysis (Fig. 11). This concern aligns with existing literature
on AI in education, which emphasizes the importance of fostering independent critical thinking skills
among students. If AI tools like ChatGPT are used as a substitute rather than a support, they risk
stifling students' development of essential instructional design competencies.
4.3. Future implications for educational practice
The integration of ChatGPT and CONALI presents substantial implications for educational practices.
As AI technology evolves, it is essential to consider strategies for its effective incorporation in
educational settings without compromising pedagogical integrity. Enhancing AI’s contextual
understanding is one avenue that holds promise for educational applications; tailored AI responses
could improve students’ ability to align UDA design elements with nuanced educational objectives.
This advancement would likely require ongoing collaboration between AI developers and educators
to refine algorithms that account for varied educational scenarios and learning environments.
Additionally, future training programs for educators should focus on equipping them to use AI tools
as enhancements to, rather than replacements for, their instructional expertise. Approximately 85%
of students expressed interest in continuing to use ChatGPT as a supplementary resource,
emphasizing the need for educators to guide students on how to critically engage with AI-generated
content. Encouraging reflective practices—where students assess AI’s contributions in light of their
instructional goals—will help maintain a balance between technological assistance and critical
pedagogical thinking (Fig. 13). These practices will be crucial in ensuring that AI serves as a
constructive tool that enhances, rather than undermines, educators’ role in shaping students’
learning experiences.
Ethical considerations must also be addressed as institutions increasingly adopt AI technologies.
Concerns surrounding data privacy and algorithmic bias are particularly salient in educational
settings, where trust between students and educators is paramount. Institutions should establish
transparent guidelines regarding the handling of data and the functionality of AI tools, fostering a
secure and trusted educational environment.
The integration of ChatGPT alongside the CONALI framework demonstrates considerable promise
for enriching the instructional design process. The observed benefits—such as increased
understanding of task requirements, improved efficiency, and high student satisfaction—highlight
the potential for these tools to support the creation of well-structured and innovative UDAs.
However, as demonstrated by the challenges related to critical thinking, technical limitations, and
the risk of AI dependency, it is essential to navigate AI integration with caution.
As AI technologies continue to shape educational practices, it is imperative that educators approach
these innovations thoughtfully, prioritizing both ethical use and pedagogical depth. By fostering an
environment where AI serves as a supportive, rather than a replacement, tool, educators can prepare
future generations to navigate an increasingly complex educational landscape. These findings
advocate for a balanced, critically engaged approach to AI in education, one that respects the insights
of human educators while embracing the efficiencies AI offers.
The ongoing dialogue surrounding AI’s role in education will undoubtedly influence instructional
design and teaching strategies in the years to come.
5. Conclusion
The integration of artificial intelligence (AI) tools such as ChatGPT, in conjunction with the
Constructive Alignment Ontology (CONALI), represents a significant innovation in instructional
design for primary education science students. This concluding analysis synthesizes the findings
from the results and discussion, highlighting the dual potential of these tools to support instructional
clarity and creativity while raising crucial considerations for their responsible integration in
educational practice.
5.1. Future implications for educational practice
The data reveals a positive reception among students towards combining ChatGPT with the CONALI
framework, underscoring the potential of these tools to facilitate effective instructional design.
Specifically, 90% of students reported an enhanced understanding of learning objectives, and
attributed this improvement to ChatGPT’s capability to clarify task requirements. By supporting the
alignment of learning outcomes with appropriate teaching strategies, ChatGPT demonstrated its
utility as a complementary resource, aligning well with Constructive Alignment (CA) principles to
enhance instructional coherence. Moreover, the satisfaction rate of 85% among students in using
ChatGPT further underscores its positive impact on their UDA design experience. The efficiency
gains associated with AI were particularly noteworthy; a substantial number of students completed
their designs faster than anticipated due to ChatGPT’s support in generating objectives and
organizing ideas. Approximately 75% of participants acknowledged ChatGPT’s role in streamlining
the ideation process, allowing them to prioritize quality and depth rather than getting stalled in
initial brainstorming phases (Fig. 8). This finding reinforces the importance of integrating AI as a
strategic support mechanism in education, particularly where time constraints are a factor. However,
alongside these benefits, the study revealed specific challenges that warrant careful consideration. A
significant portion of students (70%) emphasized the necessity of applying critical judgment when
working with AI-generated content. While ChatGPT provided a foundation for ideas, students noted
that personal insight remained essential to ensuring relevance and pedagogical accuracy,
highlighting the need for critical thinking skills in educational technology use (Fig. 11). Moreover,
technical issues such as limited server availability and delayed responses with the free ChatGPT
version detracted from the user experience, underscoring the importance of reliable, accessible tools
in educational settings (Fig. 12). These limitations suggest that while AI is a valuable asset, it is not
without its constraints and requires integration with deliberate oversight.
5.2. Implications for educational practice
These findings suggest several key implications for the use of AI and structured frameworks like
CONALI in instructional design. First, educators must guide students on how to effectively engage
with AI-generated content without compromising their analytical capabilities. Training programs
should place a strong emphasis on critical evaluation, encouraging students to critically assess AI
inputs to cultivate pedagogical judgment. Reflective practices, where students are prompted to
consider the AI’s outputs within their instructional context, will be fundamental to ensuring robust
instructional competencies.
To maximize the contextual relevance of AI-generated suggestions, further advancements in AI’s
understanding capabilities are essential. The development of algorithms that better account for
diverse educational contexts could provide more tailored responses, directly addressing specific
instructional needs. Collaboration between educators and AI developers could facilitate these
improvements, ultimately creating a more adaptable AI resource suited to varied pedagogical
scenarios.
Ethical considerations are also paramount, particularly as AI becomes more integrated into
educational settings. Transparency in AI’s data handling and algorithmic functions will be essential
for fostering trust among educators and students. Institutions should proactively establish clear
ethical guidelines for AI use, including data privacy measures and policies for mitigating algorithmic
bias. These protocols will support responsible AI integration and reinforce the credibility of
educational technology in institutional practices.
Additionally, while AI can serve as a powerful aid in UDA design, it is essential that it not replace
traditional instructional methods or human insight. Educators remain central to guiding students
through the intricacies of instructional design, and their role in fostering critical thinking skills is
irreplaceable. By positioning AI as an enhancement to traditional pedagogical strategies, educators
can ensure that students receive a balanced approach, benefiting from technological support while
developing their independent analytical skills.
5.3. Future directions
Looking forward, further research is necessary to explore the long-term effects of integrating AI
tools like ChatGPT into teacher education programs. Longitudinal studies could provide insights into
how sustained AI use shapes pedagogical skills over time, and whether it leads to improved
educational outcomes among future educators. Additionally, examining how AI influences student
perceptions and experiences with technology in education can inform best practices for effective AI
integration. Qualitative methods, such as interviews or focus groups, could offer deeper insights into
how students perceive and utilize AI, shaping a more nuanced understanding of its role in
instructional design.
In conclusion, the combination of ChatGPT and the CONALI framework presents a valuable
enhancement for instructional design among primary education science students, promoting
improved clarity, efficiency, and satisfaction in the UDA design process. However, challenges such
as fostering critical thinking, addressing technical limitations, and preventing over-reliance on AI
must be navigated with care. As technological advancements continue to reshape educational
practices, it is imperative that educators approach these tools thoughtfully and ethically, fostering
an environment where AI serves as a supportive tool that complements, rather than replaces, human
expertise.
By cultivating a balanced approach that embraces AI’s efficiencies while prioritizing critical
engagement, educators can effectively prepare future teachers for an increasingly complex
educational landscape. The ongoing dialogue surrounding AI’s role in education will undoubtedly
shape the trajectory of instructional design, setting the stage for thoughtful and effective technology
integration in the years to come.
References
[1] L. Chen, P. Chen, Z. Lin, Artificial intelligence in education: A review, IEEE Access 8 (2020)
75264–75278. doi:10.1109/ACCESS.2020.2984651.
[2] A. Vazhayil, R. Shetty, R. R. Bhavani, N. Akshay, Focusing on teacher education to introduce AI
in schools: Perspectives and illustrative findings, in: Proceedings of the 2019 IEEE Tenth
International Conference on Technology for Education (T4E), IEEE, 2019, pp. 71–77.
doi:10.1109/T4E.2019.00019.
[3] K. Mehmood, K. Verleye, A. De Keyser, B. Larivière, Piloting personalization research through
data-rich environments: A literature review and future research agenda, Journal of Service
Management 33 (2022). doi:10.1108/josm-10-2021-0405.
[4] A. Maffei, L. Daghini, A. Archenti, N. Lohse, CONALI ontology: A framework for design and
evaluation of constructively aligned courses in higher education: Putting in focus the educational
goal verbs, Procedia CIRP 50 (2016) 765–772. doi:10.1016/j.procir.2016.04.119.
[5] A. Maffei, E. Boffa, F. Lupi, M. Lanzetta, On the design of constructively aligned educational unit,
Education Sciences 12 (7) (2022) 438. doi:10.3390/educsci12070438.
[6] S. Russell, P. Norvig, Artificial intelligence: A modern approach, fourth ed., Pearson, 2020.
[7] W. Holmes, M. Bialik, C. Fadel, Artificial intelligence in education: Promises and implications for
teaching and learning, Center for Curriculum Redesign, 2019.
[8] O. Zawacki-Richter, V. I. Marín, M. Bond, F. Gouverneur, Systematic review of research on
artificial intelligence applications in higher education—Where are the educators?, International
Journal of Educational Technology in Higher Education 16 (1) (2019) 1–27. doi:10.1186/s41239-019-
0171-0.
[9] R. Luckin, W. Holmes, M. Griffiths, L. B. Forcier, Intelligence unleashed: An argument for AI in
education, Pearson Education, 2016.
[10] T. Anderson, J. Dron, Three generations of distance education pedagogy, International Review
of Research in Open and Distributed Learning 12 (3) (2011) 80–97. doi:10.19173/irrodl.v12i3.890.
[11] J. Biggs, C. Tang, Teaching for quality learning at university: What the student does, fourth ed.,
Open University Press, 2011.
[12] B. P. Woolf, Building intelligent interactive tutors: Student-centered strategies for
revolutionizing e-learning, Morgan Kaufmann, 2010.
[13] J. Biggs, Constructive alignment in university teaching, HERDSA Review of Higher Education
1 (2014) 5–22. Available: HERDSA.
[14] G.-J. Hwang, Definition, roles, and potential research issues of smart learning environments—A
context-aware ubiquitous learning perspective, Smart Learning Environments 7 (1) (2020) 1–11.
doi:10.1186/s40561-020-00134-5.
[15] F. Ouyang, P. Jiao, Artificial intelligence in education: The three paradigms, Computers and
Education: Artificial Intelligence 2 (2021) 100020. doi:10.1016/j.caeai.2021.100020.
[16] R. Luckin, W. Holmes, M. Griffiths, L. B. Forcier, Intelligence unleashed: An argument for AI in
education, Pearson Education, 2016.
[17] G. T. Doran, There’s a S.M.A.R.T. way to write management's goals and objectives, Management
Review 70 (11) (1981) 35–36.
[18] G. Brown, C. Leonard, M. Arthur-Kelly, Writing SMARTER goals for professional learning and
improving classroom practices, Reflective Practice 17 (5) (2016) 621–635.
doi:10.1080/14623943.2016.1187120.
[19] P. S. Abril, R. Plant, The patent holder’s dilemma: Buy, sell, or troll?, Communications of the
ACM 50 (2007) 36–44. doi:10.1145/1188913.1188915.