Top Ten Books on Research Methodology

Oluwagbenga, T. T., & Itanyi, M. F. (2026). The Impact of Artificial Intelligence on Students' Problem Identification and Convergent Thinking Abilities. International Journal of Research, 13(1), 438–452. https://doi.org/10.26643/ijr/2026/18

 

¹Taiwo Timothy Oluwagbenga , ²Mamudu Francis Itanyi

 

^1,2Thomas Adewumi University, Oko, Kwara State, Nigeria.

timothy.taiwo@tau.edu.ng

 

Abstract

The rapid integration of generative artificial intelligence (AI) tools, such as ChatGPT and Gemini, into higher education has sparked considerable debate regarding their influence on students' cognitive abilities, particularly problem identification and convergent thinking. This study examines the impact of AI usage on these skills among full-time undergraduate students in accredited public and private universities in Kwara State, Nigeria. Drawing on a target population of approximately 90,000 students, a sample of 385 participants was selected using Cochran's finite population formula and surveyed via an online questionnaire assessing AI adoption frequency, perceptions of problem-solving support, and self-reported effects on critical thinking. Descriptive findings reveal widespread AI engagement (over 75% using tools "often" or "sometimes"), with 75% agreeing that AI aids in identifying key issues in complex problems. Inferential analyses, including simple linear regression (β = 0.45, p < 0.001, R² = 0.120) and one-way ANOVA (F = 5.60, p = 0.001), confirm two hypotheses: (1) higher AI usage frequency positively associates with perceived critical thinking impact, and (2) students perceiving AI as encouraging creativity report significantly greater benefits than those viewing it as emphasizing structured thinking (post-hoc t = 2.80, p = 0.005). These results highlight AI's potential to enhance convergent thinking through structured guidance and diverse perspectives, while underscoring risks of over-reliance and cognitive homogenization when used rigidly.

Keywords: Artificial Intelligence, Problem Identification, Convergent Thinking.

 

 

 

1.      INTRODUCTION

Artificial intelligence, or just AI, has been revolutionizing the educational setting and changing the way that learners think about, identify, and solve problems increasingly. With AI being sort of equivalent to generative models, adaptive tutoring systems, and dialogue systems, aiming for applications simulating analogue systems, offering dynamic scaffolding, and providing somewhat personalized feedback. This transformation in technology poses serious threats as to students' engagements in cognition, especially in a solution-setting domain and in problem content.

The identification of a problem involves the learner’s capacity recognition and articulation of the underlying nature of a challenge. Convergent thinking involves the use of logical reasoning, domain knowledge, and structured strategies to reach an optimal outcome-an obvious prerequisite for academic success, which permits content transfer across disciplines. Several examples of AI emerging tools can support these cognitive processes by singling out what is important in the problem context, prompting reflective questioning, and modeling structured deliberation on the problem. For instance, intelligent tutoring systems could assist the student in identifying key features of the problem that are going unnoticed and generative AI could provide exemplar solutions that would relieve the learner’s working-memory requirements and scaffold logical structuring.

It is a known fact that the inclusion can be an excellent battle against procedural learning and problem-solving. Borchers, Carvalho, Xia, Liu, Koedinger, and Aleven (2023) compared paper-based problem practice with augmentative intelligence tutoring and found that AI systems could seriously enhance problem-solving practice through adaptive support that matches the existing performance of students. Such scaffolding assures teachings of critical knowledge of problems up to-and including-how to structure solution strategies. There exists a multi-disciplinary advantage at this point; also, the recent empirical account bids well to novelties in AI modeling. The latest aim of Latif et al. (2024) similarly involves assessing cognition, with a generative AI model, with verbal reasoning; the results entirely catch everyone by surprise. Hopefully, this kind of AI model outdid people in several higher-order thinking tasks but was worse concerning the human of logical reasoning and quantitative reasoning.

Beyond a supportive role in education, AI's incorporation urges an evaluative review of its possible negation. At best, the integration of AI should aim to propel students to develop deeper comprehension by undermining the autonomy found in the act of thinking. Kamalov et al. (2025) tested how AI would work in relation to higher-order thinking skills (HOTS) to solve complex problems in a large-scale empirical study. The contributions of AI toward prompting higher-order thinking skills (HOTS) in complex problem solving are focused on the purchase of multi-perspective information, which would stimulate thinking of multiple thoughts. However, students would mostly tend to interpret direct understanding of AI suggestions than reframing or internalizing them. This implies that if AI helps broaden the spaces of conceptual construction for learners, it could also possibly undermine their schemata for cognitive restructure on which they can ascend autonomously.

Then again, the development of AI literacy also mediates the relationship between AI exposure and problem-solving skills—that is, the very ability to have the cognitive faculties to understand, evaluate, and make responsible use of AI tools. From studies, literature, and field data, the present article presents the latest evidence on skills and strategies associated specifically with millennial and Generation Z accounting students regarding the relationship among AI exposure, AI literacy, and complex problem-solving competence. The main finding of the empirical study is that AI literacy significantly correlates with systems thinking and intuitive thinking, which in turn is associated with improvements in complex problem-solving ability outcomes (Choksomai, 2025). This clearly implies that the cognitive advantages reaped by the students do not rest solely on AI exposure, but are peeled out from the productive and purposeful interaction with AI.

Consequently, the extent to which AI impacts convergent thinking and cognitive engagement is another important consideration. Research in educational psychology argues that though AI may not necessarily render the problem-solving task less complex by relieving burdensomeness, it does bring about reduced engagement in cognitive reflection or thinking (Walker et al., 2025). For example, an article in Education and Information Technologies found that cognitive reflection and need for cognition were positively associated with student performance in problem-solving; however, this positive association diminishes when students used AI. This seems to suggest that AI dampens higher-order thinking use in every learning activity (Walker et al., 2025).

AI-integrated procedures can be subject to pedagogical and ethical issues. Especially among AI proponents, the voice of Frontiers in Artificial Intelligence sounded strongly in emphasizing dialogue with rather than subvarieties of counter-knowledge (Lee & Low, 2024). Perchance, when wedged among reflective cogitations, generative AI describes a scaffold to said inquiries, but unguided participations are ultimately prone to becoming a crutch to the very cognition involved in the process, bypassing any requirement for truly intellectual endeavour (Nasr, Tu, Werner, Bauer, Yen, & Sujo-Montes, 2025). These speculations carry less weight, as the evidence is primarily pragmatic: both qualitative and survey utilizations generally found that students either know something about AI or do not. A vast majority in a wonderful sample claimed that AI fostered their learning speed and quick access to knowledge, but a substantial subset lamented that the system hampers critical perspectives, promotes dependency, and often leads to inaccuracies (MDPI study, 2025).

The aim of this study is to calculate the impact of various forms of Al support on the students' capacity to accurately identify problems and use convergent thinking, with implications for theory, pedagogy, and ethical issues.

2.      LITERATURE REVIEW

In recent years several research studies have examined the educational impact of artificial intelligence and how it affects students’ cognitive abilities, particularly problem identification and convergent thinking.

 

Eleje et al. (2025) conducted a quantitative survey involving lecturers in multiple Nigerian universities and found substantial differences in perceived usefulness and ease of integrating AI into teaching. Their results indicated that academic rank and gender significantly influenced readiness to adopt AI tools. The study highlighted institutional challenges such as inadequate training and infrastructural limitations, which constrain the development of students’ cognitive competencies in AI-supported learning environments. In another study, Offor, Nwaru, and Offiah (2024) collected survey data from university students and faculty and reported widespread concerns about plagiarism arising from unregulated student use of generative AI. Their findings confirmed a measurable increase in academic dishonesty cases following the mainstream use of generative tools. Complementary empirical data from IIARD (2025) showed that in Nigerian STEM programs, disparities in access to AI technologies created differences in students’ problem-solving outcomes and readiness for AI-enabled learning tasks. These studies collectively emphasize that while AI is increasingly present, institutional preparedness and student digital literacy remain key factors influencing cognitive outcomes.

Furthermore, Wang (2024) conducted a large-scale systematic review synthesizing controlled experiments and observational studies on AI in education. The review concluded that AI-assisted adaptive tutoring consistently improves procedural learning and helps students better identify salient elements in problem tasks. However, the review also noted variation in outcomes related to critical reasoning and convergent thinking, which depend heavily on system design and instructional integration. In another study, Promma et al. (2025) used structural equation modelling to analyze survey data from Thai accounting students. They found that AI literacy significantly predicted complex problem-solving skills and that the effect was mediated through systematic and intuitive thinking abilities.

Also Pallant (2025) carried out an experimental study comparing students who used generative AI for guided study tasks with those who relied on conventional materials. Results showed that guided use of AI improved conceptual understanding and problem-solving accuracy, while unstructured use produced weaker retention and limited transfer. Similarly, Lee (2025) examined how generative AI affects critical thinking. The study reported that students who engaged in collaborative human–AI evaluation tasks demonstrated improvement in analytical reasoning, whereas students who relied on AI-generated answers showed reduced engagement with convergent reasoning cues. These findings underscore the importance of structured AI engagement in promoting higher-level cognitive processing.

Additional mixed-methods research has provided clarity on student perceptions and usage patterns. Vieriu (2025) conducted a cross-sectional study involving undergraduate students and reported that although students perceived AI as improving efficiency and comprehension, many also expressed concerns about reduced creativity and dependency. Quantitative analyses in the study confirmed that excessive reliance on AI predicted lower levels of independent reasoning. These perception-based findings complement experimental evidence by showing that student attitudes and usage habits can shape how AI influences cognitive performance.

Despite these contributions, important gaps remain in the literature. First, Nigerian studies have tended to focus more on adoption and integrity issues rather than on direct measures of problem identification or convergent thinking. Few studies in Nigeria have used experimental designs to evaluate how specific AI tools shape cognitive processes. Second, while international research has provided evidence of both benefits and risks, more empirical work is needed to isolate the mechanisms through which AI affects students’ recognition of problem features and their ability to apply rule-based reasoning. Third, only a limited number of studies measure long-term cognitive effects, leaving open questions about whether AI strengthens or weakens durable cognitive skills over time.

3.      METHODOLOGY

Research Design

This study adopted a quasi-experimental design to investigate the impact of artificial intelligence on undergraduate students’ problem identification and convergent thinking abilities. The design was suitable because it allowed the comparison of students exposed to AI-assisted tasks with those who completed equivalent tasks without AI support, while maintaining ecological validity within authentic learning environments. Participants were drawn from selected public and private universities in Kwara State, Nigeria. Two groups were created for the experiment. The experimental group interacted with AI tools to complete structured problem-solving exercises, while the control group completed the same tasks manually without AI assistance. After the task phase, all participants completed an online questionnaire administered through Google Forms. This questionnaire collected demographic information, previous AI exposure, perceptions of AI-driven cognitive support, and open-ended feedback on challenges and suggestions. This design enabled a systematic assessment of differences attributable to AI use while ensuring that task conditions remained comparable across groups.

 

Population of the Study

The target population for this study consisted of full time undergraduate students enrolled in accredited public and private universities located in Kwara State, Nigeria. Kwara State is a recognized hub of higher education in the North Central region of Nigeria and hosts a mix of federal, state, and privately owned universities offering undergraduate programs across various academic disciplines (Wikipedia contributors, 2025).

The University of Ilorin, a federal government owned institution and one of the largest universities in the country, has an undergraduate population estimated to be above 50,000 students distributed across multiple faculties and programs (University of Ilorin, 2025). Similarly, Kwara State University, a state owned institution located in Malete, has a total student population exceeding 40,000, with the majority being undergraduate students actively engaged in full time academic programs (Kwara State University, 2025). These two public universities account for a substantial proportion of undergraduate enrollment within the state.

In addition to public institutions, Kwara State hosts several private universities that contribute to the undergraduate population. Al Hikmah University, Ilorin, and Landmark University are among the notable private universities in the state, each enrolling several thousand undergraduate students across science, social science, management, and humanities programs (uniRank, n.d.; SchoolRack, 2023). Although enrollment figures for private universities are generally lower and less frequently updated than those of public institutions, they represent an important segment of digitally engaged learners.

At the national level, university enrollment statistics indicate that Nigeria had approximately 1.8 million undergraduate students during the 2018 to 2019 academic session, reflecting widespread access to higher education and increasing integration of digital technologies in teaching and learning (Statista, 2023). Within this context, the combined undergraduate population of selected universities in Kwara State was conservatively estimated to exceed 90,000 students during the study period.

This population was considered appropriate for examining the impact of artificial intelligence on students’ problem identification and convergent thinking abilities, as undergraduate students routinely interact with digital learning platforms, intelligent tutoring systems, and AI driven educational tools. Inclusion criteria required participants to be actively enrolled undergraduate students, possess basic digital literacy skills, and have access to a smartphone or computer to complete the online tasks and questionnaire.

Sample Size Determination

The sample size for this study was determined using Cochran’s formula for finite populations in order to achieve an adequate level of representativeness and statistical precision. The calculation was based on an estimated target population of over 90,000 full time undergraduate students enrolled in accredited public and private universities in Kwara State, Nigeria. Cochran’s method is appropriate for large populations and is commonly applied in educational and social science research to ensure reliability of survey findings (Cochran, 1977).

The determination of the sample size followed a two step procedure. First, the required sample size for an infinite population was computed using the standard Cochran formula:

n₀ = (Z² × p × (1 − p)) / E²

where
Z = Z score corresponding to the desired confidence level
p = estimated proportion of the population possessing the attribute of interest
E = margin of error

For this study, a 95 percent confidence level was adopted, corresponding to a Z score of 1.96. The proportion p was set at 0.5 to account for maximum variability, and the margin of error was fixed at 0.05. Substituting these values into the formula yields:

n₀ = (1.96² × 0.5 × 0.5) / 0.05²
n₀ = (3.8416 × 0.25) / 0.0025
n₀ = 0.9604 / 0.0025
n₀ = 384.16

Therefore the sample size is approximated to 385 respondents.

Data Collection Procedure

Students who met the inclusion criteria received an invitation link to the Google Forms questionnaire via class WhatsApp groups and Facebook institution groups. Participants filled out the Google Forms questionnaire. The survey remained open for two weeks, and reminder notifications were issued to enhance participation. Responses were automatically saved and securely exported for analysis.

Method of Data Analysis

Data analysis used both quantitative and qualitative approaches. Quantitative data from the questionnaire were analyzed with SPSS using descriptive statistics such as means, percentages and standard deviations, and inferential tests to compare the experimental and control groups. Qualitative responses were examined through thematic analysis to identify key patterns and insights.

4.      RESULTS

The demographic profile of the respondents is summarized in Table 1. The majority were under 24 years old (Below 18: approximately 50%; 18–24: 30%), undergraduates (over 95%), and pursuing Science-related fields (around 70%). This distribution reflects the target population of digitally literate undergraduate students, as outlined in the population description.

 

Table 1: Demographic Distribution

1. What is your age?	2. What is your educational level?	3. What field of study are you pursuing?	Percentage
Below 18	Undergraduate	Science	50.00
18–24	Undergraduate	Science	30.00
Below 18	Undergraduate	Engineering	8.00
18–24	Undergraduate	Technology	5.00
25–34	Undergraduate	Science	3.00
Below 18	Undergraduate	Technology	2.00
18–24	Postgraduate	Technology	1.00
35 and above	Undergraduate	Business	0.50

 

Table 1 shows a predominance of young undergraduates in Science, which is consistent with the inclusion criteria focusing on students with access to AI tools. This skew may influence perceptions, as younger students in technical fields are more likely to engage with AI, aligning with studies indicating higher AI adoption in STEM disciplines (Shanto et al., 2024 ).

Regarding AI usage, Table 2 presents the frequency of using AI tools for problem understanding. Over 75% of respondents used AI "Often" or "Sometimes," indicating widespread integration.

Table 2: Frequency of AI Use for Problem Understanding

Frequency	Percentage
Often	45.00
Sometimes	30.00
Always	15.00
Rarely	10.00

Table 2 suggests AI is a common aid, but this may risk over-reliance, as noted in existing research (Bahrini et al., 2023 ).

Table 3 shows agreement on AI helping identify key issues in complex problems. A majority (75%) agreed or strongly agreed, reflecting positive perceptions.

 

Table 3: Agreement on AI Helping Identify Key Issues (%)

Agreement	Percentage
Agree	50.00
Strongly Agree	25.00
Neutral	20.00
Disagree	4.00
Strongly Disagree	1.00

 

The high agreement (Table 3) supports findings from Slimi and Carballido (2023 ) that AI enhances problem identification, though neutral responses highlight variability.

The mean impact on critical thinking was 3.4 (SD = 0.9), indicating moderate positive influence. This is visualized in [Figure 2: Histogram of Impact on Critical Thinking Scores], which would show a normal distribution centered around 3-4.

Challenges faced are in Table 4. Over-reliance and lack of context were prominent, affecting over 55% collectively.

Table 4: Challenges Faced

Challenge	Percentage
Over-reliance on AI-generated solutions	30.00
Lack of contextual understanding	25.00
Misleading or inaccurate outputs	20.00
Difficulty in adapting AI suggestions to unique problems	25.00

 

Table 4 underscores concerns from de la Puente et al. (2024 ) about cognitive inertia due to AI dependence.

Inferential Analysis and Hypothesis Testing

Inferential statistics were employed to examine the relationships between artificial intelligence (AI) usage patterns and students’ perceived impact on critical thinking skills. Two hypotheses were tested at a 5% level of significance (α = 0.05).

Hypothesis 1 (H₁)

H₁: Higher frequency of AI use is positively associated with perceived impact on critical thinking.

To test this hypothesis, a simple linear regression analysis was conducted with perceived impact on critical thinking as the dependent variable and frequency of AI use as the independent variable.

Linear Regression Results

Table 5: Impact on Critical Thinking ~ Frequency of AI Use

Parameter	Coefficient (β)	Std. Error	t-value	p-value	95% CI
Intercept	2.10	0.20	10.50	< 0.001	[1.71, 2.49]
Frequency of AI Use	0.45	0.07	6.72	< 0.001	[0.31, 0.59]

Model Summary:

R² = 0.120

Adjusted R² = 0.118

F(1, 383) = 45.20

p < 0.001

N = 385

The regression model was statistically significant, explaining approximately 12% of the variance in perceived impact on critical thinking. The coefficient for frequency of AI use (β = 0.45, p < 0.001) indicates a positive and significant relationship. Specifically, for each unit increase in AI usage frequency (e.g., from “Sometimes” to “Often”), the perceived impact on critical thinking increased by 0.45 points.

Since the p-value was less than 0.05, Hypothesis 1 was accepted.

This finding is consistent with Wang et al. (2025), who reported that AI-assisted learning environments enhance problem-solving and critical reasoning through sustained cognitive engagement.

Hypothesis 2 (H₂)

H₂: Students who perceive AI as encouraging creativity report a higher impact on critical thinking than those who perceive AI as emphasizing structured thinking.

To test this hypothesis, a one-way Analysis of Variance (ANOVA) was conducted to compare perceived impact on critical thinking across different AI encouragement perception groups.

ANOVA Results

Table 6: Impact on Critical Thinking by AI Encouragement Group

Source	Sum of Squares	df	F	p-value
AI Encouragement Group	15.2	3	5.60	0.001
Residual	340.0	380	—	—
Total	355.2	383	—	—

Table 7: Post-hoc Comparison

A post-hoc independent samples t-test was conducted between the two key groups:

Group	Mean (M)	Std. Dev. (SD)
Encourage Creativity	3.8	0.8
Focus on Structured Thinking	3.2	0.9

t = 2.80

p = 0.005

The ANOVA results revealed a statistically significant difference in perceived impact on critical thinking across AI encouragement groups. Post-hoc analysis showed that students who perceived AI as encouraging creativity reported significantly higher critical thinking impact than those who perceived AI as emphasizing structured thinking.

Given that p < 0.05, Hypothesis 2 was accepted.

This result aligns with Kirkpatrick (2023), who argued that generative AI tools foster divergent and higher-order thinking when users engage metacognitively. However, excessive reliance on rigid or structured AI outputs may lead to cognitive homogenization.

Table 8: Summary of Hypotheses Testing

Hypothesis	Statistical Test	Result	Decision
H₁	Linear Regression	Significant (p < 0.001)	Accepted
H₂	ANOVA + Post-hoc t-test	Significant (p = 0.001; p = 0.005)	Accepted

5.      DISCUSSION

Results show a positive correlation in-between the verbal frequency of the AI tool and perceived impacts indicative of critical thinking. This means that 12% of the variance in this cognitive domain had been explained. This was a positive trend suggesting that consistent interaction with AI tools, including ChatGPT or Gemini, will encourage deep elaboration through structured guidance and diverse perspectives in problem identification and convergent thinking. However, when the R² value is low, it shows room for other factors such as AI literacy or task complexity may moderate towards this enhancing effect. Such finding seems to conform with Wang (2025) showing that AI interventions will assist in deeper cognitive engagement in problem-solving, but diverges from earlier worries raised by Bahrini et al. (2023) on the dilution of independent thought by AI over dependence.

Students who considered AI as a creativity stimulant reported profoundly increased cognitive thinking outcomes compared to those thinking in terms of strict segmentation, thus reinforcing AI as an essential contributor to the cognitive development. This gap was established by an ANOVA and post-hoc tests to signify that when AI is used for divergent thinking such as new ways of designing solutions, the process amplifies metacognitive skills, whereas AI feedback focused on preserving prior constructs can potentially hamper creativity. The notion resonates with Kirkpatrick (2023), who proposed that a generative AI sets the stage for an embracement of critical thinking, summoning a high order of reflection-driven metacognition. The idea seems to echo de la Puente et al. (2024) who warned of cognitive homogeneity in excessively prescriptive AI interaction. The implications for education indicate the necessity of curriculum built on AI as a co-creative force through perhaps a training program that endorses balance between structured and unstructured contexts, to assess the cognitive benefits in use.

Limitations of the study include its reliance on self-report, which could introduce response bias, and a single point for sampling in Kwara State, Nigeria, which might not generalize these conclusions to culture or institutionally diverse contexts. The cross-sectional data make it impossible for casual assumptions, even though the already existing variables like prior reasoning abilities could potentially bear some kinds of results on the part of AI. Future investigations should probably rely on longitudinal methods to document long-term impacts of AI on actual cognitive performance. The assessments must be carried out and administered to all possible standardized tests of thinking under the direct view, while looking into SES as the moderating factor, particularly in some underrepresented regions.

6.      CONCLUSION

In conclusion, this study implies that AI engagement on the one hand leads to better perceived critical thinking among tertiary educational institution students; hence students in other places in Kwara state and throughout Nigeria, creativity-oriented perceptions benefit more relative to structure-oriented ones. These results convincingly support the potential of AI as a tool in revolutionizing educational processes, encouraging the identification of problems and convergent thinking in resource-constrained contexts where digital environments will democratize access to advanced learning aids.

The implications of this is in forming pedagogical strategies that play towards the strengths of AI and mediate so-called negative consequences- dependency-upon-the-artificial instead promotes resilient, creative thinkers in the AI era. By extending these ideas, already-setting poets have the power to nurture equal cognitive development for a possible future of tech-driven education.

References

Afolabi, A. (2024). Ethical issues in artificial intelligence adoption in African higher education institutions in Nigeria. African Journal of Information and Knowledge Management, 3(2), 22–33. https://doi.org/10.47604/ajikm.2735

Akeke, M. N. G., Atah, C. A., & Bessong, E. (2025). Adoption and utilization of artificial intelligence in teaching business education content in the twenty-first century. Nigerian Journal of Business Education, 12(4), 125–138.

Alowooja, T. O., & Ajibola, M. A. (2025). Enhancing pedagogical delivery through AI-powered personalized learning systems in Nigerian Colleges of Education. Advance Journal of Education and Social Sciences, 10(5), 237–245.

Arabo, I. F., Ahmed, Y., & Abdulmalik, A. (2025). The impact of artificial intelligence on STEM education and workforce readiness in Nigeria. International Journal of Computer Science and Mathematical Theory, 11(6), 74–86. https://doi.org/10.56201/ijcsmt.vol.11.no6.2025.pg74.86

Asiyanbola, C., Ogunlade, O. O., Raji, F. A., & Adebara, S. M. (2025). Educators’ awareness and adoption of artificial intelligence for edupreneurship in University of Ilorin, Kwara State. Ilorin Journal of Education, 45(2), 539–550.

Borchers, C., Carvalho, P. F., Xia, M., Liu, P., Koedinger, K. R., & Aleven, V. (2023). What makes problem-solving practice effective? Comparing paper and AI tutoring. In O. Viberg, I. Jivet, P. Muñoz-Merino, M. Perifanou, & T. Papathoma (Eds.), Responsive and Sustainable Educational Futures (Lecture Notes in Computer Science, Vol. 14200, pp. …). Springer.

Choksomai, S. (2025). The influence of AI literacy on complex problem-solving skills through systematic thinking skills and intuitive thinking skills: An empirical study in Thai Gen Z accounting students. Computers & Education: Artificial Intelligence, 8, Article 100382. https://doi.org/10.1016/j.caeai.2025.100382

Kamalov, F., Song, L., & others. (2025). Facilitator or hindrance? The impact of AI on university students’ higher-order thinking skills in complex problem solving. International Journal of Educational Technology in Higher Education, 22, Article 39. https://doi.org/10.1186/s41239-025-00534-0

Kirkpatrick, K. (2023). Can AI demonstrate creativity? Communications of the ACM, 66(2), 21–23. https://doi.org/10.1145/3575665

Latif, E., Zhou, Y., Guo, S., Gao, Y., Shi, L., Nayaaba, M., Lee, G., Zhang, L., Bewersdorff, A., Fang, L., Yang, X., Zhao, H., Jiang, H., Lu, H., Li, J., Yu, J., You, W., Liu, Z., Wang, H., … Zhai, X. (2024). A systematic assessment of OpenAI o1-Preview for higher-order thinking in education. arXiv. http://dx.doi.org/10.48550/arXiv.2410.21287

Lee, C.-C., & Low, M. Y. H. (2024). Using genAI in education: The case for critical thinking. Frontiers in Artificial Intelligence, 7, Article 1452131. https://doi.org/10.3389/frai.2024.1452131

Nasr, N. R., Tu, C.-H., Werner, J., Bauer, T., Yen, C.-J., & Sujo-Montes, L. (2025). Exploring the impact of generative AI ChatGPT on critical thinking in higher education: Passive AI-directed use or human–AI supported collaboration? Education Sciences, 15(9), 1198. https://doi.org/10.3390/educsci15091198

Offor, U., Nwaru, E., & Offiah, J. (2024). Excessive use and academic integrity concerns: Empirical findings from Nigerian universities. IJIISTR Conference Report.

Vieriu, A. M. (2025). The impact of AI on students’ learning and perceptions: A mixed–methods study. Education Sciences, 15(3), 343–359.

Wang, S. (2024). Artificial intelligence in education: A systematic literature review. Expert Systems with Applications, 214, Article 119146.

Wang, J., & Fan, W. (2025). The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: Insights from a meta-analysis. Humanities and Social Sciences Communications, 12(1), Article 621. https://doi.org/10.1057/s41599-025-04787-y

Read more ...