arxiv: 2604.08594 · v1 · submitted 2026-04-02 · 🧬 q-bio.NC · cs.AI· cs.HC

Recognition: 2 theorem links

· Lean Theorem

Mapping generative AI use in the human brain: divergent neural, academic, and mental health profiles of functional versus socio emotional AI use

Junjie Wang , Xianyang Gan , Dan Liu , Jingxian He , Stefania Ferraro , Keith M. Kendrick , Weihua Zhao , Shuxia Yao

show 2 more authors

Christian Montag Benjamin Becker

Authors on Pith no claims yet

Pith reviewed 2026-05-13 20:18 UTC · model grok-4.3

classification 🧬 q-bio.NC cs.AIcs.HC

keywords generative AIAICAbrain structuremental healthacademic performancefunctional usesocio-emotional useMRI

0 comments

The pith

Functional AI chatbot use correlates with better grades and larger prefrontal brain volumes, while socio-emotional use links to poorer mental health and smaller social-processing regions.

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

This paper examines how university students' patterns of using generative AI conversational agents relate to their academic performance, mental health, and brain structure. Combining self-report surveys with high-resolution MRI scans from 222 participants, it separates general and functional uses from socio-emotional ones. Functional uses appear tied to higher GPA scores, greater gray matter volume in dorsolateral prefrontal and calcarine areas, and stronger hippocampal network efficiency. Socio-emotional uses instead track with elevated depression and social anxiety scores plus reduced volume in superior temporal and amygdala regions. The work shows that identical AI tools produce divergent neural and behavioral signatures depending on the user's goals and interaction style.

Core claim

Across computational anatomy, meta-analytic network level, and behavioral decoding analyses, higher general and functional AICA use frequencies were linked to better academic outcomes (GPA), larger dorsolateral prefrontal and calcarine gray matter volume, and enhanced hippocampal network clustering and local efficiency. In contrast, more frequent socio-emotional AICA use was associated with poorer mental health (depression, social anxiety) and lower volume of superior temporal and amygdalar regions central to social and affective processing.

What carries the argument

The separation of AICA usage into functional versus socio-emotional categories, which maps onto distinct prefrontal-hippocampal cognitive networks versus superior temporal-amygdalar social-affective systems.

If this is right

Functional AICA use may support academic performance through structural expansion in dorsolateral prefrontal and hippocampal systems.
Socio-emotional AICA use may coincide with reduced gray matter volume in regions supporting social and affective processing.
The same AI tools can produce opposing associations with cognition and mental health depending on usage motivation.
Network clustering and local efficiency gains in the hippocampus accompany higher functional AICA frequency.
Lower superior temporal and amygdalar volumes track with elevated socio-emotional AICA frequency and poorer mental health scores.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

If the associations prove directional, educational policies could encourage functional AI prompts while monitoring socio-emotional reliance.
The observed brain differences raise the possibility that habitual AI interaction styles influence ongoing maturation of cognitive versus social circuits in young adults.
Replication in non-student samples or with objective usage logs rather than self-reports would test whether the pattern generalizes beyond university settings.
Pre-existing differences in social skills or cognitive style might still contribute to both usage preferences and brain measures, suggesting value for future controlled or longitudinal designs.

Load-bearing premise

The measured associations reflect the effects of different AI usage patterns on brain structure and mental health rather than pre-existing traits that simultaneously shape both usage choices and brain metrics.

What would settle it

A longitudinal follow-up that measures the same participants' brain volumes and mental health scores before and after documented shifts in functional versus socio-emotional AICA use frequencies, then finds no corresponding structural or symptom changes.

read the original abstract

The widespread adoption of generative artificial intelligence conversational agents (AICAs) among university students constitutes a novel cognitive social environment whose impact on the maturing brain remains elusive. Combining surveys with high resolution structural MRI, we examined patterns of general, functional, and socio emotional AICA use, academic performance, mental health, and brain structural signatures in a comparatively large sample of 222 young individuals. Across computational anatomy, meta analytic network level, and behavioral decoding analyses, we observed use specific associations. Higher general and functional AICA use frequencies were linked to better academic outcomes (GPA), larger dorsolateral prefrontal and calcarine gray matter volume, and enhanced hippocampal network clustering and local efficiency. In contrast, more frequent socio emotional AICA use was associated with poorer mental health (depression, social anxiety) and lower volume of superior temporal and amygdalar regions central to social and affective processing. These findings indicate that the same class of AI tools exerts distinct effects depending on usage patterns and motivations, engaging prefrontal hippocampal systems that support cognition versus socio emotional systems that may track distress linked usage. These heterogeneities are crucial for designing environments that harness the educational benefits of AI while mitigating mental health risks.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

The paper splits generative AI use into functional versus socio-emotional patterns and links them to opposing brain and outcome measures in 222 students, but the cross-sectional design leaves causation wide open.

read the letter

The core finding is that functional AI use (study help, problem-solving) tracks with higher GPA, larger dorsolateral prefrontal and calcarine gray matter, and tighter hippocampal networks, while socio-emotional use (chatting for support) tracks with worse depression and anxiety plus smaller superior temporal and amygdala volumes. That split is the main new angle compared with earlier broad AI or screen-time work. The sample of 222 is decent for an MRI study, and they ran standard voxel-based morphometry plus network metrics, which gives a clean initial map of associations. The behavioral decoding and meta-analytic network steps add some interpretability without overcomplicating things. The obvious limitation is the design. Everything is measured at one time point, so the data cannot separate whether AI habits shape brain structure or whether pre-existing differences in prefrontal volume or social traits steer people toward one usage style over the other. Self-report categories can blur too, and no strong controls for IQ, personality, or prior mental health appear in the abstract. These are standard issues for observational brain-behavior work rather than fatal ones. The paper is aimed at researchers tracking how digital tools intersect with adolescent and young-adult brain development, education outcomes, and mental health. A reader already working on technology effects or policy around AI in schools would get the most out of it. I would send it for peer review. The question is timely and the usage distinction is worth testing further, even if the current evidence stays correlational and needs longitudinal or experimental follow-up to carry real weight.

Referee Report

3 major / 3 minor

Summary. The paper reports a cross-sectional observational study of 222 university students combining self-reported frequencies and motivations for generative AI conversational agent (AICA) use—categorized as general, functional, or socio-emotional—with structural MRI metrics, GPA, and mental health measures. It claims that higher functional AICA use correlates with better academic outcomes, larger dorsolateral prefrontal and calcarine gray matter volumes, and enhanced hippocampal network clustering and efficiency, whereas higher socio-emotional AICA use correlates with poorer mental health (depression, social anxiety) and reduced superior temporal and amygdalar volumes.

Significance. If the reported associations are robust and the design limitations are addressed, the work would offer timely correlational evidence on how distinct patterns of AI tool use relate to brain structure and behavioral outcomes in young adults, potentially informing educational and mental-health guidelines. The sample size and multi-modal approach (survey + MRI + network metrics) are strengths, but the purely observational nature means the findings remain descriptive rather than mechanistic.

major comments (3)

[Abstract, Discussion] Abstract and Discussion: The central claim that AICAs 'exert distinct effects depending on usage patterns' (and that functional use 'engages prefrontal hippocampal systems' while socio-emotional use 'may track distress') interprets cross-sectional correlations as directional impacts. The data consist of single-time-point self-reports and MRI; no longitudinal component, instrumental variables, or confounder adjustment is described, so the associations are equally consistent with pre-existing neural or trait differences selecting for usage type. This interpretation is load-bearing for the paper's conclusions and requires either causal language to be removed or explicit discussion of the reverse-causation alternative.
[Methods, Results] Methods/Results: No details are provided on control for potential confounders (e.g., socioeconomic status, baseline cognitive ability, screen time unrelated to AI, or personality traits) that could jointly influence both AICA usage patterns and the reported brain volumes or GPA. Without such controls or sensitivity analyses, the specificity of the functional vs. socio-emotional distinctions cannot be evaluated.
[Results] Results: The manuscript reports associations between usage frequencies and MRI/network metrics but does not include confidence intervals, effect sizes, or correction for multiple comparisons across the numerous brain regions and network measures examined. This weakens the reliability of the highlighted 'larger DLPFC/calcarine' and 'lower superior temporal/amygdala' findings.

minor comments (3)

[Methods] Clarify the exact survey items and validation used to separate 'functional' from 'socio-emotional' AICA motivations; self-report bias could blur the categories.
[Methods] Add a power analysis or justification for the sample size of 222 given the number of statistical tests performed.
[Figures, Results] Ensure all figures include error bars or confidence intervals and that network metrics (clustering, local efficiency) are defined with reference to the specific parcellation or atlas used.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for their thoughtful and constructive comments. We have revised the manuscript to address all major concerns by removing causal language, adding explicit discussion of limitations including reverse causation and incomplete confounder control, and enhancing the statistical reporting with effect sizes, confidence intervals, and multiple-comparison correction. Our point-by-point responses follow.

read point-by-point responses

Referee: [Abstract, Discussion] The central claim that AICAs 'exert distinct effects depending on usage patterns' (and that functional use 'engages prefrontal hippocampal systems' while socio-emotional use 'may track distress') interprets cross-sectional correlations as directional impacts. The data consist of single-time-point self-reports and MRI; no longitudinal component, instrumental variables, or confounder adjustment is described, so the associations are equally consistent with pre-existing neural or trait differences selecting for usage type. This interpretation is load-bearing for the paper's conclusions and requires either causal language to be removed or explicit discussion of the reverse-causation alternative.

Authors: We agree that the original wording risked implying causality. We have revised the abstract and discussion to use strictly associational language, replacing 'exert distinct effects' with 'are associated with distinct patterns' and 'engages' with 'correlates with'. We have added a dedicated paragraph discussing the cross-sectional design and explicitly noting that reverse causation—pre-existing differences in brain structure, cognitive ability, or mental health traits selecting for usage type—remains a plausible alternative explanation that cannot be ruled out without longitudinal data. revision: yes
Referee: [Methods, Results] No details are provided on control for potential confounders (e.g., socioeconomic status, baseline cognitive ability, screen time unrelated to AI, or personality traits) that could jointly influence both AICA usage patterns and the reported brain volumes or GPA. Without such controls or sensitivity analyses, the specificity of the functional vs. socio-emotional distinctions cannot be evaluated.

Authors: We acknowledge this limitation. The survey did not collect personality traits or detailed non-AI screen time. We have added a limitations subsection in the Discussion describing these gaps and their implications for specificity. We performed sensitivity analyses controlling for available covariates (age, sex, total intracranial volume, and parental education as a proxy for socioeconomic status); the main associations remained stable. We agree that richer covariate data would be valuable and have highlighted this for future work. revision: partial
Referee: [Results] The manuscript reports associations between usage frequencies and MRI/network metrics but does not include confidence intervals, effect sizes, or correction for multiple comparisons across the numerous brain regions and network measures examined. This weakens the reliability of the highlighted 'larger DLPFC/calcarine' and 'lower superior temporal/amygdala' findings.

Authors: We have updated the Results to report Pearson's r effect sizes and 95% confidence intervals for all primary associations. We applied FDR correction across the brain regions and network metrics; the key DLPFC, calcarine, superior temporal, amygdala, and hippocampal network findings survive correction and are now clearly labeled as such, with uncorrected results moved to supplementary materials. revision: yes

Circularity Check

0 steps flagged

No circularity: empirical associations only

full rationale

The paper reports direct statistical associations from cross-sectional survey and MRI data in 222 participants. No equations, derivations, fitted parameters, or predictions appear in the abstract or described methods. Claims rest on observed correlations between usage frequencies, GPA, mental health scores, and brain volumes/network metrics without any reduction to inputs by construction or self-citation load-bearing steps. This is a standard empirical neuroimaging study with no mathematical chain to inspect for circularity.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The study rests on established neuroimaging analysis pipelines and standard statistical assumptions for volume and graph-theoretic metrics without introducing new free parameters or postulated entities.

axioms (1)

standard math Standard assumptions of linear models and network efficiency calculations in structural MRI hold for the sample.
Invoked for gray matter volume and hippocampal network clustering analyses.

pith-pipeline@v0.9.0 · 5556 in / 1287 out tokens · 45743 ms · 2026-05-13T20:18:28.248581+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

IndisputableMonolith/Foundation/RealityFromDistinction.lean reality_from_one_distinction unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

Higher general and functional AICA use frequencies were linked to better academic outcomes (GPA), larger dorsolateral prefrontal and calcarine gray matter volume, and enhanced hippocampal network clustering and local efficiency.
IndisputableMonolith/Cost/FunctionalEquation.lean washburn_uniqueness_aczel unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

more frequent socio-emotional AICA use was associated with poorer mental health (depression, social anxiety) and lower volume of superior temporal and amygdalar regions

What do these tags mean?

matches: The paper's claim is directly supported by a theorem in the formal canon.
supports: The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends: The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses: The paper appears to rely on the theorem as machinery.
contradicts: The paper's claim conflicts with a theorem or certificate in the canon.
unclear: Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

200 extracted references · 200 canonical work pages

[1]

Will our social brain inherently shape and be shaped by interactions with AI? Neuron 113, 2037–2041 (2025)

Becker, B. Will our social brain inherently shape and be shaped by interactions with AI? Neuron 113, 2037–2041 (2025)

work page 2037
[2]

& Becker, B

Montag, C., Spapé, M. & Becker, B. Can AI really help solve the loneliness epidemic? Trends Cognit. Sci. 29, 869–871 (2025)

work page 2025
[3]

& Gašević, D

Yan, L., Greiff, S., Teuber, Z. & Gašević, D. Promises and challenges of generative artificial intelligence for human learning. Nat. Hum. Behav. 8, 1839–1850 (2024)

work page 2024
[4]

& Sharot, T

Glickman, M. & Sharot, T. How human–AI feedback loops alter human perceptual, emotional and social judgements. Nat. Hum. Behav. 9, 345–359 (2025)

work page 2025
[5]

& West, R

Salvi, F., Horta Ribeiro, M., Gallotti, R. & West, R. On the conversational persuasiveness of GPT-4. Nat. Hum. Behav. 9, 1645–1653 (2025)

work page 2025
[6]

& Gibb, R

Kolb, B. & Gibb, R. Searching for the principles of brain plasticity and behavior. Cortex 58, 251–260 (2014)

work page 2014
[7]

& Cheung, A

Chen, S. & Cheung, A. C. K. Effect of generative artificial intelligence on university students learning outcomes: a systematic review and meta-analysis. Educ. Res. Rev. 49, 100737 (2025)

work page 2025
[8]

Doshi, A. R. & Hauser, O. P. Generative AI enhances individual creativity but reduces the collective diversity of novel content. Sci. Adv. 10, eadn5290 (2024)

work page 2024
[9]

& Fan, W

Wang, J. & Fan, W. The effect of ChatGPT on students’ learning performance, learning perception, and higher-order thinking: insights from a meta-analysis. Humanit. Soc. Sci. Commun. 12, 621 (2025)

work page 2025
[10]

G., Bradbury, T

Smith, M. G., Bradbury, T. N. & Karney, B. R. Can generative AI chatbots emulate human connection? A relationship science perspective. Perspect. Psychol. Sci. 20, 1081–1099 (2025)

work page 2025
[11]

& Zhou, G

Wang, X., Zhou, Y. & Zhou, G. The application and ethical implication of generative AI in mental health: systematic review. JMIR Ment Health 12, e70610 (2025)

work page 2025
[12]

Use of large language models might affect our cognitive skills

Heersmink, R. Use of large language models might affect our cognitive skills. Nat. Hum. Behav. 8, 805–806 (2024)

work page 2024
[13]

Kosmyna, N. et al. Your brain on ChatGPT: accumulation of cognitive debt when using an AI assistant for essay writing task. arXiv prepr. arXiv:2506,08872 4, (2025)

work page 2025
[14]

Chen, Y. et al. Effects of generative artificial intelligence on cognitive effort and task performance: study protocol for a randomized controlled experiment among college students. Trials 26, 244 (2025)

work page 2025
[15]

Fan, Y. et al. Beware of metacognitive laziness: effects of generative artificial intelligence on learning motivation, processes, and performance. Br. J. Educ. Technol. 56, 489–530 (2025)

work page 2025
[16]

Phang, J. et al. Investigating affective use and emotional well-being on ChatGPT. arXiv prepr. arXiv:2504,03888 (2025)

work page 2025
[17]

& Markett, S

Montag, C. & Markett, S. Social media use and everyday cognitive failure: investigating the fear of missing out and social networks use disorder relationship. BMC Psychiatry 23, 872 (2023)

work page 2023
[18]

Mueller, S. C. et al. Gray matter volume in adolescent anxiety: an impact of the brain-derived neurotrophic factor Val66Met polymorphism? J. Am. Acad. Child Adolesc. Psychiatry 52, 184–195 (2013)

work page 2013
[19]

& Wegner, D

Sparrow, B., Liu, J. & Wegner, D. M. Google effects on memory: cognitive consequences of having information at our fingertips. Science 333, 776–778 (2011)

work page 2011
[20]

Wei, Z. et al. Social media distractions alter behavioral and neural patterns to global-local attention: the moderation effect of fear of missing out. Comput. Hum. Behav. 157, 108258 (2024)

work page 2024
[21]

Klugah-Brown, B. et al. Associations between levels of internet gaming disorder symptoms and striatal morphology–replication and associations with social anxiety. Psychoradiology 2, 207–215 (2022)

work page 2022
[22]

Nagata, J. M. et al. Social media use and depressive symptoms during early adolescence. JAMA Netw. Open 8, e2511704 (2025)

work page 2025
[23]

Klugah-Brown, B. et al. Common neurofunctional dysregulations characterize obsessive–compulsive, substance use, and gaming disorders—an activation likelihood meta-analysis of functional imaging studies. Addict. Biol. 26, e12997 (2021)

work page 2021
[24]

& Becker, B

Montag, C. & Becker, B. Neuroimaging the effects of smartphone (over-) use on brain function and structure—a review on the current state of MRI-based findings and a roadmap for future research. Psychoradiology 3, kkad001 (2023)

work page 2023
[25]

C., Lindenberger, U

Kühn, S., Gleich, T., Lorenz, R. C., Lindenberger, U. & Gallinat, J. Playing super mario induces structural brain plasticity: gray matter changes resulting from training with a commercial video game. Mol. Psychiatry 19, 265–271 (2014)

work page 2014
[26]

Maza, M. T. et al. Association of habitual checking behaviors on social media with longitudinal functional brain development. JAMA pediatr. 177, 160–167 (2023)

work page 2023
[27]

Zhou, F. et al. Orbitofrontal gray matter deficits as marker of internet gaming disorder: converging evidence from a cross-sectional and prospective longitudinal design. Addict. Biol. 24, 100–109 (2019)

work page 2019
[28]

Herbener, A. B. & Damholdt, M. F. Are lonely youngsters turning to chatbots for companionship? The relationship between chatbot usage and social connectedness in danish high-school students. Int. J. Hum.-Comput. Stud. 196, 103409 (2025)

work page 2025
[29]

& Kim, K

Hu, B., Mao, Y. & Kim, K. J. How social anxiety leads to problematic use of conversational AI: the roles of loneliness, rumination, and mind perception. Comput. Hum. Behav. 145, 107760 (2023)

work page 2023
[30]

& Liu, K

Peng, C., Zhang, S., Wen, F. & Liu, K. How loneliness leads to the conversational AI usage intention: the roles of anthropomorphic interface, para-social interaction. Curr. Psychol. 44, 8177–8189 (2025)

work page 2025
[31]

& Marín-López, I

Rodríguez-Ruiz, J., Espejo-Siles, R. & Marín-López, I. Are psychopathy traits related to the use of artificial intelligence tools among university students? The mediating effect of self-control. Deviant Behav. 46, 1510–1519 (2025)

work page 2025
[32]

& Espejo-Siles, R

Rodríguez-Ruiz, J., Marín-López, I. & Espejo-Siles, R. Is artificial intelligence use related to self-control, self-esteem and self-efficacy among university students? Educ. Inf. Technol. 30, 2507–2524 (2025)

work page 2025
[33]

& Luo, W

Chen, J., Xue, J., Li, Y. & Luo, W. Impact of different employee–AI interaction: instrumental vs. Emotional support and gender differences. Int. J. Hum.–Comput. Interact. 1–14 (2025)

work page 2025
[34]

Yankouskaya, A. et al. LLM-D12: a dual-dimensional scale of instrumental and relational dependencies on large language models. ACM Trans. Web (2025) doi:10.1145/3765895

work page doi:10.1145/3765895 2025
[35]

Zhuang, Q. et al. Segregating domain-general from emotional context-specific inhibitory control systems - ventral striatum and orbitofrontal cortex serve as emotion-cognition integration hubs. Neuroimage 238, 118269 (2021)

work page 2021
[36]

Montag, C., Marciano, L., Schulz, P. J. & Becker, B. Unlocking the brain secrets of social media through neuroscience. Trends Cognit. Sci. 27, 1102–1104 (2023)

work page 2023
[37]

Zhou, X. et al. Higher levels of (internet) gaming disorder symptoms according to the WHO and APA frameworks associate with lower striatal volume. J. behav. addctn. 9, 598–605 (2020)

work page 2020
[38]

& Schmiedek, F

Lövdén, M., Bäckman, L., Lindenberger, U., Schaefer, S. & Schmiedek, F. A theoretical framework for the study of adult cognitive plasticity. Psychol. Bull. 136, 659–676 (2010)

work page 2010
[39]

& Whishaw, I

Kolb, B. & Whishaw, I. Q. Brain plasticity and behavior. Annu. Rev. Psychol. 49, 43–64 (1998)

work page 1998
[40]

& Bäckman, L

Lövdén, M., Wenger, E., Mårtensson, J., Lindenberger, U. & Bäckman, L. Structural brain plasticity in adult learning and development. Neurosci. Biobehav. Rev. 37, 2296–2310 (2013)

work page 2013
[41]

& Sharma, R

Bozkurt, A. & Sharma, R. C. Generative AI and prompt engineering: the art of whispering to let the genie out of the algorithmic world. Asian J. Distance Educ. 18, i–vii (2023)

work page 2023
[42]

& Kurasinski, A

Korzynski, P., Mazurek, G., Krzypkowska, P. & Kurasinski, A. Artificial intelligence prompt engineering as a new digital competence: analysis of generative AI technologies such as ChatGPT. Entrep. Bus. Econ. Rev. 11, 25–37 (2023)

work page 2023
[43]

& Trajkovik, V

Pesovski, I., Santos, R., Henriques, R. & Trajkovik, V. Generative AI for customizable learning experiences. Sustainability 16, 3034 (2024)

work page 2024
[44]

& Su, H.-C

Li, C.-J., Hwang, G.-J., Chang, C.-Y. & Su, H.-C. Generative AI-supported progressive prompting for professional training: effects on learning achievement, critical thinking, and cognitive load. Br. J. Educ. Technol. https://doi.org/10.1111/bjet.13594 (2025) doi:10.1111/bjet.13594

work page doi:10.1111/bjet.13594 2025
[45]

& Gašević, D

Yan, L., Pammer-Schindler, V., Mills, C., Nguyen, A. & Gašević, D. Beyond efficiency: empirical insights on generative AI’s impact on cognition, metacognition and epistemic agency in learning. Br. J. Educ. Technol. 56, 1675–1685 (2025)

work page 2025
[46]

Executive Functions

Diamond, A. Executive Functions. Annual Review of Psychology vol. 64 135–168 (2013)

work page 2013
[47]

Miller, E. K. & Cohen, J. D. An integrative theory of prefrontal cortex function. Annu. Rev. Neurosci. 24, 167–202 (2001)

work page 2001
[48]

Niendam, T. A. et al. Meta-analytic evidence for a superordinate cognitive control network subserving diverse executive functions. Cogn. Affect. Behav. Neurosci. 12, 241–268 (2012)

work page 2012
[49]

C., Cai, Y

Kwok, S. C., Cai, Y. & Buckley, M. J. Mnemonic introspection in macaques is dependent on superior dorsolateral prefrontal cortex but not orbitofrontal cortex. J. Neurosci. 39, 5922–5934 (2019)

work page 2019
[50]

& Rahnev, D

Shekhar, M. & Rahnev, D. Distinguishing the roles of dorsolateral and anterior PFC in visual metacognition. J. Neurosci. 38, 5078–5087 (2018)

work page 2018
[51]

Shaked, D. et al. Dorsolateral prefrontal cortex volume as a mediator between socioeconomic status and executive function. Neuropsychology 32, 985 (2018)

work page 2018
[52]

& Raz, N

Yuan, P. & Raz, N. Prefrontal cortex and executive functions in healthy adults: a meta-analysis of structural neuroimaging studies. Neurosci. Biobehav. Rev. 42, 180–192 (2014)

work page 2014
[53]

M., Weil, R

Fleming, S. M., Weil, R. S., Nagy, Z., Dolan, R. J. & Rees, G. Relating introspective accuracy to individual differences in brain structure. Science 329, 1541–1543 (2010)

work page 2010
[54]

& Vuilleumier, P

Sinanaj, I., Cojan, Y. & Vuilleumier, P. Inter-individual variability in metacognitive ability for visuomotor performance and underlying brain structures. Conscious. Cogn. 36, 327–337 (2015)

work page 2015
[55]

Kwok, V. et al. Learning new color names produces rapid increase in gray matter in the intact adult human cortex. Proc. Natl. Acad. Sci. 108, 6686–6688 (2011)

work page 2011
[56]

Xia, Z. et al. Neural correlates of oral word reading, silent reading comprehension, and cognitive subcomponents. Int. j. behav. dev. 42, 342–356 (2018)

work page 2018
[57]

Burgess, N., Maguire, E. A. & O’Keefe, J. The human hippocampus and spatial and episodic memory. Neuron 35, 625–641 (2002)

work page 2002
[58]

J., Keane, M

Palombo, D. J., Keane, M. M. & Verfaellie, M. How does the hippocampus shape decisions? Neurobiology of Learning and Memory 125, 93–97 (2015)

work page 2015
[59]

D., Watson, P

Rubin, R. D., Watson, P. D., Duff, M. C. & Cohen, N. J. The role of the hippocampus in flexible cognition and social behavior. Front. Hum. Neurosci. 8, 742 (2014)

work page 2014
[60]

& Dunbar, R

Shultz, S. & Dunbar, R. I. M. Species differences in executive function correlate with hippocampus volume and neocortex ratio across nonhuman primates. J. Comp. Psychol. 124, 252 (2010)

work page 2010
[61]

J., Rus, O

Koch, K., Reess, T. J., Rus, O. G. & Zimmer, C. Extensive learning is associated with gray matter changes in the right hippocampus. Neuroimage 125, 627–632 (2016)

work page 2016
[62]

Lee, B. C. & Chung, J. (Jae). An empirical investigation of the impact of ChatGPT on creativity. Nat. Hum. Behav. 8, 1906–1914 (2024)

work page 1906
[63]

& Drigas, A

Pergantis, P., Bamicha, V., Skianis, C. & Drigas, A. AI chatbots and cognitive control: enhancing executive functions through chatbot interactions: a systematic review. Brain Sci. 15, 47 (2025)

work page 2025
[64]

Urban, M. et al. ChatGPT improves creative problem-solving performance in university students: an experimental study. Comput. Educ. 215, 105031 (2024)

work page 2024
[65]

& Zhao, X

Zhang, Q., Li, L., Xu, C., Qi, Y. & Zhao, X. How can artificial intelligence help college students develop entrepreneurial ability? Evidence from China. Int. J. Manag. Educ. 23, 101244 (2025)

work page 2025
[66]

Ballard, I. C. et al. Dorsolateral prefrontal cortex drives mesolimbic dopaminergic regions to initiate motivated behavior. J. Neurosci. 31, 10340–10346 (2011)

work page 2011
[67]

Gruber, M. J. & Ranganath, C. How curiosity enhances hippocampus-dependent memory: the prediction, appraisal, curiosity, and exploration (PACE) framework. Trends Cognit. Sci. 23, 1014–1025 (2019)

work page 2019
[68]

Holton, E. et al. Goal commitment is supported by vmPFC through selective attention. Nat. Hum. Behav. 8, 1351–1365 (2024)

work page 2024
[69]

W., Yorke, J

Feng, X., Tian, L., Ho, G. W., Yorke, J. & Hui, V. The effectiveness of AI chatbots in alleviating mental distress and promoting health behaviors among adolescents and young adults: systematic review and meta-analysis. J. Med. Internet Res. 27, e79850 (2025)

work page 2025
[70]

Heinz Michael V. et al. Randomized trial of a generative AI chatbot for mental health treatment. NEJM AI 2, AIoa2400802 (2025)

work page 2025
[71]

Perlis, R. H. et al. Generative AI use and depressive symptoms among US adults. JAMA Netw. Open 9, e2554820–e2554820 (2026)

work page 2026
[72]

Zhang, X. et al. Exploring artificial intelligence (AI) chatbot usage behaviors and their association with mental health outcomes in chinese university students. J. Affect. Disord. 380, 394–400 (2025)

work page 2025
[73]

& Jiang, Y

Lai, L., Pan, Y., Xu, R. & Jiang, Y. Depression and the use of conversational AI for companionship among college students: the mediating role of loneliness and the moderating effects of gender and mind perception. Front. Public Health 13, 1580826 (2025)

work page 2025
[74]

Davis, R. A. A cognitive-behavioral model of pathological internet use. Comput. Hum. Behav. 17, 187–195 (2001)

work page 2001
[75]

The social brain: neural basis of social knowledge

Adolphs, R. The social brain: neural basis of social knowledge. Annu. Rev. Psychol. 60, 693–716 (2009)

work page 2009
[76]

& Chang, E

Bhaya-Grossman, I. & Chang, E. F. Speech computations of the human superior temporal gyrus. Annu. Rev. Psychol. 73, 79–102 (2022)

work page 2022
[77]

Insel, T. R. & Fernald, R. D. How the brain processes social information: searching for the social brain. Annu, Rev, Neurosci, 27, 697–722 (2004)

work page 2004
[78]

Kennedy, D. P. & Adolphs, R. The social brain in psychiatric and neurological disorders. Trends Cognit. Sci. 16, 559–572 (2012)

work page 2012
[79]

Mihov, Y. et al. Mirroring fear in the absence of a functional amygdala. Biol. Psychiatry 73, e9–e11 (2013)

work page 2013
[80]

Zhang, R. et al. A neurofunctional signature of affective arousal generalizes across valence domains and distinguishes subjective experience from autonomic reactivity. Nat. Commun. 16, 6492 (2025)

work page 2025

Showing first 80 references.