A Conceptual Exploration of Generative AI-Induced Cognitive Dissonance and its Emergence in University-Level Academic Writing

Carl Errol Seran; Hezerul Abdul Karim; Myles Joshua Toledo Tan; Nouar AlDahoul

REVIEW 3 major objections 5 minor 3 references

A Conceptual Exploration of Generative AI-Induced Cognitive Dissonance and its Emergence in University-Level Academic Writing

T0 review · 3 major / 5 minor · reviewed 2026-08-08 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read This paper argues that generative AI is both a trigger and an amplifier of cognitive dissonance in university academic writing, and that the resulting tension is a single psychological construct.

desk verdict A tidy conceptual synthesis that overclaims its evidence: the trigger/amplifier framing is useful, but the cited studies never actually measure cognitive dissonance. read the letter →

arxiv 2502.05698 v1 pith:I2XR7KGS submitted 2025-02-08 cs.CY

Carl Errol Seran , Myles Joshua Toledo Tan , Hezerul Abdul Karim , Nouar AlDahoul This is my paper · ORCID

classification cs.CY

keywords cognitivedissonancegenerativeAIacademicwritinguniversitystudentsintegrityliteracyself-efficacyhighereducation

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper argues that the psychological unease students report when using generative AI in university writing is a recognizable cognitive-dissonance process, not just vague anxiety. It claims GenAI is both a trigger that creates new ethical conflicts and an amplifier that intensifies existing tensions around originality, effort, and intellectual ownership. The authors propose a single hypothetical construct of "GenAI-induced cognitive dissonance" and link it to behavioral outcomes such as avoiding effortful tasks, justifying AI use, and reassessing academic values. If the construct holds, institutions can design policies, transparency standards, and writing-task redesigns that reduce the tension instead of leaving students to resolve it privately.

What carries the argument

The central object is the hypothetical construct of GenAI-induced cognitive dissonance, a single tension-based construct in which GenAI is simultaneously a trigger (it introduces new contradictions between efficiency and integrity) and an amplifier (it heightens ongoing struggles with authorship, skill development, and self-efficacy). The machinery that carries the argument is Festinger's cognitive-dissonance theory, which connects reported states of discomfort to the drive to reduce inconsistency through justification, avoidance, or value reassessment. The paper's Figure 1 formalizes the construct as a cycle: GenAI disrupts core academic values, the resulting dissonance produces self-doubt and ethical dilemmas, those shape behaviors such as avoiding effortful tasks, and increasing dependence on GenAI feeds back to amplify the tension.

What would settle it

A study that measures students' tension during GenAI-assisted writing with a validated cognitive-dissonance scale, alongside measures of fear of detection, general anxiety, and performance pressure, would settle the claim. If the reported tension tracks fear of detection or performance pressure more closely than it tracks the conflict between AI use and academic values, or if students who use GenAI without believing it conflicts with integrity show the same tension, the paper's central claim would not hold.

Watch

Extended reading notes

Core claim

The paper's central claim is that generative AI tools do not merely sit alongside the traditional stresses of academic writing; they actively create and intensify a specific psychological conflict, cognitive dissonance. Using Festinger's theory, the authors argue that a student who uses GenAI while valuing originality, effort, and intellectual ownership holds contradictory cognitions, and that the resulting discomfort shows up in self-reported ambivalence, guilt, frustration, and self-doubt. The paper draws on survey and experimental evidence to portray GenAI as a trigger that introduces new ethical dilemmas and as an amplifier that deepens existing tensions about skills and authorship, and it condenses this into a hypothetical construct of "GenAI-induced cognitive dissonance" with identifiable manifestations and behavioral consequences.

Load-bearing premise

The argument assumes that the ambivalence, guilt, frustration, and self-doubt reported in the cited studies are expressions of cognitive dissonance rather than fear of detection, general anxiety, or performance pressure, since those studies rely on self-reported attitudes and behavior rather than validated cognitive-dissonance measures.

Editorial extensions

If this is right

Transparent declaration of GenAI use should reduce student guilt and anxiety, because it aligns AI-assisted work with institutional values and removes ambiguity.
AI literacy programs and reflective pedagogy should be adopted as dissonance-reduction strategies, shifting students from rationalizing AI use toward reassessing their academic values.
Discipline-specific task redesigns that require students to manually refine or expand AI-generated drafts should preserve intellectual ownership and slow the erosion of self-efficacy.
The paper's construct implies that vague or purely punitive GenAI policies may worsen dissonance by leaving the ethical boundary unclear.
If the construct is correct, student justifications such as "everyone does it" are signs of unresolved dissonance and should be addressed through education, not accepted as settled ethical positions.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

The paper does not test the trigger/amplifier distinction; an extension would compare students new to GenAI with heavy users to see whether the two proposed roles respond differently to the same intervention.
A stricter test of the Festingerian reading would use a validated cognitive-dissonance scale or a physiological measure of discomfort, since the cited studies infer dissonance from self-reported attitudes and behavior.
The construct yields a concrete prediction for learning analytics: students who report guilt about AI use should subsequently avoid effortful writing tasks more, and ownership-restoring interventions should reduce that avoidance.
A natural modeling extension, not pursued in the paper, is to treat dissonance reduction as a mediator between GenAI-use policies and academic-integrity outcomes.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

The paper is a clear, well-organized conceptual review of how generative AI might produce psychological tension in university writing. What's genuinely useful is the trigger/amplifier distinction: it gives institutions a simple vocabulary for talking about student discomfort that is otherwise vague. The mitigation section is also sensible, offering transparent AI-use policies, reflective pedagogy, and task redesign that could be implemented tomorrow. I'll credit the authors for labeling their construct "hypothetical" and for using hedged language like "potentially" in several places. That shows some awareness of the evidence gap.

The gap is still load-bearing, though. The central claim is that GenAI "serves as both a trigger and amplifier of CD," and the paper says this is supported by "robust evidence" from Chan, Ju, Playfoot and Quigley, Zhai, Ironsi, and Hutson. But none of those studies actually measures cognitive dissonance. They measure self-reported ambivalence, guilt, frustration, confidence loss, attitudes about integrity, and convenience preferences. Those are not the same as Festingerian dissonance, which requires the specific experience of discomfort from inconsistency between one's own behavior and one's own internalized values. A student can feel guilty about using AI because they're afraid of getting caught, or anxious because they didn't learn anything, without any dissonance between their action and a deeply held value. The paper never rules these alternatives out, and it never uses a validated dissonance measure. So the apparent empirical support is circular: the authors define the cited findings as evidence of CD, then use those same findings to validate the construct. Calling the construct "hypothetical" does not fix the overstatement when the conclusion says "empirical evidence suggests that GenAI can act as both a trigger and an amplifier of CD."

There are also smaller issues. The sample sizes are modest (Ju's experiment has 32 participants; Zhai's review is fine but the paper quotes specific percentages without appraising study quality). There is no baseline comparison to non-GenAI writing, so we don't know if the reported tension is new or just the ordinary stress of academic writing. The self-citation to Tan and Maravilla is minor and not a problem.

Who gets value from this? Educators, instructional designers, and institutional policy people who want a framework for discussing AI and student wellbeing. It is not a research contribution; it is a synthesis with a plausible but unproven construct. As a referee, I'd ask the authors to do one of two things: either reframe the paper strictly as a conceptual proposal with no claim of empirical validation, or add a serious research agenda for measuring CD directly and testing against alternative affective states.

I would send this to peer review. The topic is important, the writing is coherent, and the flaws are fixable. But the referee should insist on toning down the evidence claims and adding an explicit discussion of discriminant validity.

Referee Report

3 major / 5 minor

Summary. This conceptual paper argues that the use of generative AI (GenAI) in university-level academic writing can serve both as a trigger and as an amplifier of cognitive dissonance (CD). Drawing on Festinger's classical theory, the authors propose a hypothetical construct of 'GenAI-induced CD,' in which the tension between AI-driven efficiency and academic values of originality, effort, and integrity produces psychological discomfort, self-doubt, ethical dilemmas, and confidence erosion. The paper reviews several empirical studies (Chan, 2024; Ju, 2023; Playfoot & Quigley, 2024; Zhai et al., 2024; Ironsi & Ironsi, 2024; Hutson, 2024) that report student ambivalence, guilt, frustration, and reduced self-efficacy, and interprets these as manifestations of Festingerian dissonance. It then proposes mitigation strategies: AI literacy programs, reflective pedagogy, transparency standards, and discipline-specific task redesign. The paper explicitly labels the construct as 'hypothetical' and frames itself as a conceptual exploration rather than an empirical test, but it repeatedly describes the supporting evidence as 'robust' and 'well-documented.'

Significance. If the proposed construct can be validated, the paper would fill a genuine gap by connecting the growing literature on GenAI in education to a foundational social-psychological theory, with practical implications for pedagogy and institutional policy. The authors deserve credit for transparency in calling the construct 'hypothetical' and for proposing concrete, actionable interventions. The strength of the paper is its synthetic scope and its attention to an underexplored psychological dimension of AI-assisted writing. However, the paper's contribution is currently weakened by an evidentiary overreach: the central claim that GenAI 'acts as both a trigger and an amplifier of CD' rests on studies that do not directly measure cognitive dissonance, and the absence of discriminant validation means the construct could be a relabeling of generic negative affect. As a hypothesis-generating framework, the paper is useful; as currently argued, it does not support its 'robust evidence' language.

major comments (3)

[GenAI-induced Cognitive Dissonance: A Hypothetical Construct] The central claim that GenAI 'serves as a significant trigger for cognitive dissonance' is supported by citing Chan (2024), Ju (2023), and Playfoot and Quigley (2024), but none of these studies actually measures cognitive dissonance. Chan surveys attitudes toward 'AI-giarism' and reports ambivalence; Ju's 32-participant experiment measures writing accuracy and self-confidence; Playfoot and Quigley report beliefs about integrity and convenience. The mapping from these self-reported attitudes and affects to Festinger's construct is assumed, not demonstrated. Guilt, anxiety, self-doubt, and frustration could reflect fear of detection, social desirability concerns, or performance pressure—states with different antecedents and remedies. This measurement gap is load-bearing because the paper's conclusion that 'empirical evidence suggests that GenAI can act as both a trigger and an amplifier of CD' depends on this equivalence. The authors should either reframe the claim as a hypothesis requiring direct validation with dissonance-specific instruments (e.g., measures of psychological discomfort tied to inconsistency between one's own behavior and internalized values), or add a substantive limitation section acknowledging that the cited studies do not operationally distinguish GenAI-induced CD from other negative affect.
[GenAI as a Potential Amplifier of Pre-existing Tensions] The 'amplifier' claim is not empirically established. The supporting studies are small or self-report based (Ju, n=32; Ironsi and Ironsi, n=150; Hutson, n=200) and none includes a non-GenAI control or a pre/post design that compares the same students' tensions before and after GenAI use. To claim amplification of 'pre-existing conflicts' or 'pre-existing tensions,' the paper would need evidence that GenAI increases the frequency or intensity of these tensions relative to a baseline without GenAI. The current evidence shows only that students report tensions and concerns while using GenAI, which is consistent with a simple association, not with amplification. The terms 'robustly support' and 'well-documented' in this section overstate the evidentiary basis. Please temper the language and explicitly frame the amplifier claim as a plausible but untested mechanism that should be examined with suitable baseline comparisons.
[GenAI-induced Cognitive Dissonance: A Hypothetical Construct (concluding paragraphs)] There is a circularity concern in the way the construct is validated. The paper states that Festinger's theory predicts 'psychological discomfort, behavioral justifications, value reassessment' and that 'such are evident across studies,' but the studies cited were already used to define the construct in the preceding paragraphs. This means the 'predictions' are being confirmed by the same evidence that was used to build the model, rather than by independent tests. The construct would be strengthened by an explicit list of a priori, falsifiable predictions that go beyond the cited phenomenology—e.g., predictions about which students are most likely to show dissonance-reduction behaviors (justification vs. value change), or about dose-response relationships between GenAI reliance and dissonance scores. As it stands, the model cannot be falsified by the cited evidence, because the evidence is interpretative rather than confirmatory.

minor comments (5)

[Introduction] The sentence 'lesser attention to grammar instruction, and focus on the writing process often lead to poorly refined scientific documents' would benefit from rephrasing for clarity (e.g., 'lesser attention to grammar instruction and a strong focus on the writing process can lead...').
[References] The reference 'Strive, E. (2024)' appears in the text as 'Strive, 2024' but the reference list entry is 'Stride, E. (2024).' Please verify the correct author name and citation.
[References] Several reference entries lack complete bibliographic information or have inconsistent formatting (e.g., Hutson, 2024, and the Vaidis & Bran entry mixes journal volume and page numbering conventions). Please ensure all references match a consistent style and include full titles, volume, issue, and page ranges where available.
[Figure 1] Figure 1 is described in the text but not visible in the manuscript as provided; if it is a conceptual diagram, please ensure it is legible and that all labels (e.g., 'trigger,' 'amplifier,' 'behavioral outcomes') match the terminology used in the body text.
[GenAI as a Potential Amplifier of Pre-existing Tensions] The phrase 'a potential catalyst of CD' in the paragraph beginning 'GenAI tools such as ChatGPT have emerged...' introduces a third functional role ('catalyst') that is not defined separately from 'trigger' and 'amplifier.' Please clarify whether these terms are meant to be synonymous or distinct.

Circularity Check

1 steps flagged · score 4.0 of 10

The paper's empirical support for its central construct is partly circular: CD is operationalized as psychological discomfort or tension, and the cited 'evidence' consists of reports of guilt, ambivalence, and frustration; the trigger/amplifier framework then re-labels these same summarized findings.

self definitional [Sections 'Understanding Cognitive Dissonance in Traditional Academic Writing' and 'GenAI-induced Cognitive Dissonance: A Hypothetical Construct' (trigger and amplifier subsections)]
"For the purpose of clarity, CD is operationalized as a state of psychological discomfort or tension... This gap between perceived benefits and actual skills development created psychological tension, with 61% expressing frustration about their inability to improve independently while relying heavily on GenAI... This contradiction between practicality and integrity exemplifies CD... These predictions are in the form of psychological discomfort, behavioral justifications, value reassessment, such are evident across studies."

The paper defines CD as 'psychological discomfort or tension,' then treats survey reports of ambivalence, guilt, anxiety, and frustration as 'robust evidence' and 'well-documented' support that GenAI triggers and amplifies CD. Because the operational definition already includes the very states measured in the cited studies, the confirmation reduces to the definition: any report of discomfort around AI use automatically counts as CD. The trigger/amplifier taxonomy is then a re-description of the same summarized findings rather than an independent test. The cited instruments do not appear to administer a Festingerian dissonance measure, so the mapping is imposed by the paper's own broad operationalization.

full rationale

The paper is a conceptual exploration, and its central construct is explicitly labeled hypothetical, which mitigates the charge of circularity somewhat. However, the operational definition of CD as 'psychological discomfort or tension' is broad enough that the cited reports of guilt, ambivalence, frustration, and self-doubt become evidence for CD mainly by definition. The subsequent split into trigger and amplifier re-describes the same empirical summaries instead of testing the construct independently, so the apparent support is partly internal to the paper's definitions. The self-citation of Tan and Maravilla (2024) appears only in the mitigation section and is not load-bearing for the central claim. No fitted parameters, imported uniqueness theorems, or ansatz smuggling are present.

Assumptions & free parameters 0 free parameters · 3 assumptions · 1 invented entities

No free parameters are needed because the paper is a conceptual synthesis, not a quantitative model. The central claim rests on domain assumptions about CD theory and on uncritical acceptance of cited empirical studies. The only invented conceptual entity is the construct itself, which is clearly labeled hypothetical.

assumptions (3)

domain assumption Cognitive dissonance theory applies to GenAI-assisted writing in the same way as traditional academic dishonesty contexts.
The paper imports Festinger's framework and applies it to GenAI use without new evidence that the dissonance process is equivalent. Invoked in 'Understanding Cognitive Dissonance' and the hypothetical construct section.
domain assumption Students' self-reported ambivalence, guilt, and frustration are valid indicators of cognitive dissonance.
The supporting studies cited (Chan, Ju, Playfoot) measure attitudes and self-reports; the paper interprets these as CD without a validated CD scale. Invoked in section 'GenAI as a Potentially Trigger'.
domain assumption The cited surveys and experimental papers are reliable and representative.
The paper accepts the empirical claims (e.g., 25% reduction in writing accuracy, 68%/52%/61% proportions) at face value. Location: several sections reporting aggregate statistics.

invented entities (1)

GenAI-induced cognitive dissonance (hypothetical construct)
purpose: To unify observed student tensions under a single conceptual model with trigger and amplifier pathways.
The paper explicitly labels it a hypothetical construct (Figure 1) and offers no measurement instrument or falsifiable prediction beyond the cited studies.

how reviews work

0 comments

Cite this review

Pith. "Pith review of A Conceptual Exploration of Generative AI-Induced Cognitive Dissonance and its Emergence in University-Level Academic Writing." pith.science (2026). https://pith.science/paper/I2XR7KGS

@misc{pith2026250205698,
  author       = {Pith},
  title        = {Pith review of: A Conceptual Exploration of Generative AI-Induced Cognitive Dissonance and its Emergence in University-Level Academic Writing},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/I2XR7KGS}},
  note         = {Machine review of arXiv:2502.05698}
}

read the original abstract

The integration of Generative Artificial Intelligence (GenAI) into university-level academic writing presents both opportunities and challenges, particularly in relation to cognitive dissonance (CD). This work explores how GenAI serves as both a trigger and amplifier of CD, as students navigate ethical concerns, academic integrity, and self-efficacy in their writing practices. By synthesizing empirical evidence and theoretical insights, we introduce a hypothetical construct of GenAI-induced CD, illustrating the psychological tension between AI-driven efficiency and the principles of originality, effort, and intellectual ownership. We further discuss strategies to mitigate this dissonance, including reflective pedagogy, AI literacy programs, transparency in GenAI use, and discipline-specific task redesigns. These approaches reinforce critical engagement with AI, fostering a balanced perspective that integrates technological advancements while safeguarding human creativity and learning. Our findings contribute to ongoing discussions on AI in education, self-regulated learning, and ethical AI use, offering a conceptual framework for institutions to develop guidelines that align AI adoption with academic values.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

3 extracted references · 1 canonical work pages

[1]

AIgiarism,

A Conceptual Exploration of Generative AI-Induced Cognitive Dissonance and its Emergence in University-Level Academic Writing Carl Errol Seran 1,2,3 , Myles Joshua Toledo Tan 1,2,4,5,6,7,8,* , Hezerul Abdul Karim 9,* , Nouar AlDahoul 10 1 Biology Program, College of Arts and Sciences, University of St. La Salle, Bacolod, Philippines 2 Department of Natura...

work page 1957
[2]

The Will to Believe

Hypothetical construct of GenAI-induced cognitive dissonance (CD). GenAI is both a trigger and an amplifier of CD in academic writing. The figure illustrates how the integration of GenAI disrupts core academic values, leading to CD, which manifests through self-doubt, ethical dilemmas, and confidence erosion, among others. These psychological tensions inf...

arXiv 2024
[28]

https://doi.org/10.1186/s40561-024-00316-7

doi:10.1186/s40561-024-00316-7

Pith tools

Reviewed August 8, 2026 · model on record in the stance chip above.

[1] [1]

AIgiarism,

A Conceptual Exploration of Generative AI-Induced Cognitive Dissonance and its Emergence in University-Level Academic Writing Carl Errol Seran 1,2,3 , Myles Joshua Toledo Tan 1,2,4,5,6,7,8,* , Hezerul Abdul Karim 9,* , Nouar AlDahoul 10 1 Biology Program, College of Arts and Sciences, University of St. La Salle, Bacolod, Philippines 2 Department of Natura...

work page 1957

[2] [2]

The Will to Believe

Hypothetical construct of GenAI-induced cognitive dissonance (CD). GenAI is both a trigger and an amplifier of CD in academic writing. The figure illustrates how the integration of GenAI disrupts core academic values, leading to CD, which manifests through self-doubt, ethical dilemmas, and confidence erosion, among others. These psychological tensions inf...

arXiv 2024

[3] [28]

https://doi.org/10.1186/s40561-024-00316-7

doi:10.1186/s40561-024-00316-7