loading

About BILM

At BILM, we believe in the power of education, accreditation and excellence in longevity medicine for doctors in the UK.

Doctor reading a medical journal on a laptop in a hospital library

Contact Info

Narrative ReviewBritish Journal of Longevity Medicine · Autumn–Winter 2026 · Volume 1 · Issue 1

Loneliness, Brain Ageing and Cognitive Healthspan:
Neurobiological Mechanisms, Cognitive Consequences and Opportunities for Prevention

Nathan Roy

University of Bristol Medical School, Bristol, United Kingdom

Corresponding author: Nathan Roy, University of Bristol Medical School, Bristol, United Kingdom.

Article type
Narrative Review
Published
18 October 2026
Volume / Issue
Volume 1, Issue 1 (Autumn–Winter 2026)
DOI
DOI pending
e-ISSN
e-ISSN pending
Licence
This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Abstract

Loneliness is a subjective and distressing experience arising from a perceived discrepancy between desired and actual social relationships. Increasing evidence suggests that its relevance extends beyond psychological well-being to brain health, cognition and healthy ageing. This narrative review examines evidence linking loneliness with neural systems involved in social cognition, memory and stress regulation, and evaluates its association with cognitive impairment and dementia. Neuroimaging studies implicate prefrontal and default-mode networks, the hippocampal and temporal systems, anterior insula, amygdala and subcortical structures, although findings are heterogeneous and do not support a single ‘lonely brain’ phenotype. Longitudinal studies and recent meta-analyses associate persistent loneliness with poorer cognitive performance and increased risks of cognitive impairment and dementia. Potential pathways include altered social-information processing, chronic stress responses, inflammation, sleep disturbance, cerebrovascular disease, adverse health behaviours and reduced cognitive reserve. However, reverse causation and confounding by depression, sensory impairment, multimorbidity and socioeconomic factors remain important limitations. Interventions can reduce loneliness, but current evidence does not establish that doing so prevents neurodegeneration or dementia. Persistent loneliness is therefore best regarded as a potentially modifiable marker and risk factor relevant to cognitive healthspan rather than an established direct cause of neurodegeneration. Future longitudinal and interventional research should determine whether sustained reductions in loneliness translate into preservation of cognitive and brain health.

Keywords: loneliness; social isolation; brain ageing; cognition; dementia; neuroimaging; cognitive healthspan; prevention

1. Introduction

Loneliness is commonly defined as an aversive subjective state arising from a discrepancy between the social relationships an individual desires and those they perceive themselves to have.¹ It is related to, but distinct from, social isolation, which describes the objective extent of social contact or integration. A person may have relatively few social contacts without feeling lonely, while another may experience profound loneliness despite frequent interaction. This distinction is clinically important because subjective loneliness and objective social-network measures do not necessarily show the same relationships with brain structure, cognition or health outcomes.²

Historically approached predominantly as a psychosocial problem, loneliness is increasingly studied as a determinant of physical and neurological health. This is particularly relevant to longevity medicine. Extending lifespan without preserving cognition, independence and psychological well-being would represent limited success; cognitive healthspan is therefore an important component of healthy ageing. Longitudinal and pooled studies associate loneliness with poorer cognitive performance and subsequent cognitive impairment or dementia.³,⁴ Neuroimaging reviews report differences in distributed regions and networks involved in memory, executive function, reward, threat processing and social cognition.²,⁵

This narrative review examines the relationship between loneliness, brain structure and cognitive function, with particular attention to social-cognitive networks, hippocampal and temporal systems, stress-related circuitry, cerebrovascular mechanisms, cognitive decline and dementia. It also evaluates evidence for interventions and considers the implications for preventive and longevity medicine. The central question is not simply whether loneliness accompanies poorer brain health, but whether persistent loneliness represents a clinically meaningful and potentially modifiable target for preserving cognitive healthspan.

2. Approach to the Literature

This narrative review drew on targeted searches for PubMed-indexed studies and reviews, supplemented by examination of relevant publisher records and article abstracts. Search concepts included loneliness, social isolation, neuroimaging, brain structure, cognition, cognitive impairment, dementia, ageing, inflammation and interventions. The evidence was selected purposively to prioritise systematic reviews and meta-analyses, longitudinal cohorts, population-based studies and neuroimaging research relevant to the review question. Studies were considered through 12 October 2026. This was not an exhaustive systematic search: no prespecified protocol, comprehensive database-search strategy, duplicate screening or formal risk-of-bias assessment was undertaken. Accordingly, the selection of evidence may be incomplete and the conclusions should be interpreted as a critical narrative synthesis rather than a systematic review.

3. Loneliness, Social Cognition and the Prefrontal Cortex

Social cognition encompasses processes that allow individuals to attend to socially relevant stimuli, infer the intentions and emotions of others, evaluate reward and threat, represent the self and regulate behaviour within social environments. The prefrontal cortex (PFC), together with the anterior insula, temporal regions, striatum and other distributed networks, contributes to these functions.⁶ Experimental and neuroimaging work therefore provides a plausible basis for examining whether persistent loneliness is associated with altered processing within social-cognitive networks.²,⁵

A systematic review by Lam and colleagues identified associations between loneliness and functional and structural differences across neural systems implicated in social cognition, including the default mode network (DMN).² The DMN contributes to self-referential processing, autobiographical memory and representations of other people. One influential model proposes that loneliness is accompanied by increased vigilance for social threat and altered interpretation of social information. More recent synthesis of functional neuroimaging studies supports context-dependent differences in attention, sensitivity to negative social information and mentalising, while cautioning against a single neural signature of loneliness.⁵ These adaptations may promote self-preservation or reconnection in the short term, but persistent negative expectations and withdrawal could reinforce perceived disconnection.

Lesion-based evidence also supports involvement of the PFC and anterior insula. Cristofori and colleagues reported lower loneliness scores among individuals with penetrating traumatic brain injuries involving the right anterior insula and right PFC.⁷ This is intriguing because these regions contribute to emotional awareness and social-affective processing. Nevertheless, lesion findings cannot be translated straightforwardly into a model in which reduced activity in these regions is beneficial: injury to the same networks can produce executive dysfunction, apathy and impaired quality of life. Collectively, the evidence favours distributed and context-dependent neural processes rather than a discrete ‘loneliness centre’ in the brain.

The hippocampus and surrounding medial temporal structures are central to episodic memory and are vulnerable to ageing and neurodegenerative disease. Several studies have therefore examined whether loneliness is associated with structural changes in these regions. Düzel and colleagues studied 319 adults aged 61–81 years and found that higher loneliness was associated with smaller grey-matter volumes in clusters involving the left amygdala/anterior hippocampus, left posterior parahippocampus and left cerebellum. Importantly, associations persisted after accounting for social-network size, reinforcing the distinction between subjective loneliness and objective isolation.⁸

Longitudinal evidence extends this observation into midlife. In the Framingham Heart Study, persistent rather than transient loneliness was associated with poorer memory and executive function and with smaller temporal-lobe volume. Among women, persistent loneliness was also associated with smaller frontal, temporal and hippocampal volumes and larger lateral ventricular volume. Greater cumulative loneliness was associated with smaller parietal, temporal and hippocampal volumes, particularly in the presence of APOE ε4.⁹ These findings suggest that chronicity may matter: a transient episode of loneliness may not carry the same implications as a sustained subjective state.

However, hippocampal atrophy is unlikely to explain the relationship by itself. Park and colleagues found that white-matter hyperintensities may contribute to the relationship between loneliness and memory, while hippocampal volume did not emerge as the principal mediator in their analysis.¹⁰ Other community-based imaging work has associated loneliness with lower frontal white-matter and basal-ganglia volumes but has not consistently identified differences across all cortical or cerebrovascular measures.¹¹ Similarly, Salinas and colleagues reported associations with lower total cerebral volume and greater white-matter-hyperintensity burden, but no significant association with hippocampal volume.¹² The overall literature is therefore heterogeneous and better supports interacting neural and vascular pathways than a single structural mechanism.

5. Stress, Inflammation and Other Neurobiological Pathways

Loneliness may influence cognition through chronic stress and altered processing of social threat. The amygdala contributes to salience, fear learning and emotional processing, while the hypothalamic–pituitary–adrenal axis coordinates neuroendocrine responses to stress. Morr and colleagues reviewed evidence linking chronic loneliness with altered social-threat processing and neuroendocrine function, while emphasising that mechanistic findings remain heterogeneous.¹³ Persistent loneliness may therefore be associated with a state in which social information is processed through heightened threat sensitivity, potentially reinforcing avoidance and distress.

Several pathways could connect prolonged loneliness with later cognitive vulnerability. Chronic stress may influence sleep, vascular risk, inflammatory signalling and neuroplasticity. Behavioural pathways may operate simultaneously: loneliness can coexist with reduced physical activity, poorer sleep, depression and other health behaviours that themselves influence cognition. In 383,421 UK Biobank participants, Zhao and colleagues reported associations of social isolation and loneliness with multiple neurological and psychiatric outcomes, worse cognitive and emotional performance and structural brain differences; peripheral inflammatory and biochemical markers partially mediated some associations.¹⁴ Mendelian-randomisation analyses in the same study provided evidence for putative causal relationships with selected psychiatric and neurological outcomes, but did not establish a direct causal pathway from loneliness to dementia.¹⁴

These findings strengthen biological plausibility but also illustrate the difficulty of isolating one mechanism. Stress, inflammation, vascular health, sleep, mood and behaviour are interdependent. Loneliness may influence these processes, be influenced by them, or both. A multidimensional model is therefore more appropriate than attributing cognitive decline to a single neurobiological pathway.

6. Cognitive Decline and Dementia

The most clinically important question is whether loneliness predicts meaningful cognitive decline. Earlier longitudinal work associated loneliness with incident dementia and markers of neurocognitive vulnerability. Salinas and colleagues found that loneliness was associated with subsequent dementia risk and with early markers of vulnerability to neurocognitive decline, including lower total cerebral volume and greater white-matter injury rather than a significant difference in hippocampal volume.¹² Sundström and colleagues likewise reported an association between loneliness and subsequent all-cause dementia and Alzheimer’s disease.¹⁵ Such studies helped move the field beyond cross-sectional associations.

More recent pooled evidence strengthens the association. Luchetti and colleagues combined 21 samples comprising 608,561 participants and found that loneliness was associated with higher risk of all-cause dementia (hazard ratio [HR] 1.306, 95% confidence interval [CI] 1.197–1.426), Alzheimer’s disease (HR 1.393, 95% CI 1.290–1.504) and vascular dementia (HR 1.735, 95% CI 1.483–2.029); the vascular dementia estimate, however, was based on only three studies and was heavily influenced by one large cohort. Across 16 samples comprising 103,387 participants, loneliness was also associated with cognitive impairment (HR 1.150, 95% CI 1.113–1.189). In analyses using fully adjusted estimates, associations were attenuated but remained statistically significant for all-cause dementia (HR 1.189, 95% CI 1.101–1.285) and cognitive impairment (HR 1.093, 95% CI 1.045–1.143). Between-study heterogeneity remained substantial.⁴

A separate 2024 meta-analysis of eight longitudinal studies involving 45,032 participants found an association between loneliness and subsequent mild cognitive impairment or cognitive decline (overall odds ratio 1.14, 95% CI 1.05–1.23).³ Harmonised cognitive data from more than 20,000 participants across the United States, England, India, China, South Africa, Mexico and Chile have subsequently shown that loneliness is associated with poorer overall cognitive performance and informant-rated cognitive decline, with associations also observed across several cognitive domains.¹⁶ This cross-national replication reduces the likelihood that the association is confined to a single healthcare system or cultural setting.

Nevertheless, association should not be equated with causation. Prodromal neurodegenerative disease may reduce social engagement, impair communication or alter motivation, increasing loneliness before dementia is diagnosed. Depression can influence both loneliness and cognition; in the Framingham analyses by Salinas and colleagues, the association between loneliness and cognitive performance was no longer statistically significant after adjustment for depressive symptoms.¹² Hearing or visual impairment, multimorbidity, bereavement, socioeconomic disadvantage and reduced mobility may create additional common causes. Persistent loneliness may therefore function simultaneously as a risk factor, mediator, consequence and early marker of declining brain health. This bidirectionality is central to interpretation of the literature.

7. Can Loneliness Be Modified to Preserve Cognitive Health?

Interventions for loneliness can be broadly divided into approaches that increase opportunities for social contact and approaches that target the cognitive and emotional processes maintaining perceived disconnection. Community groups, befriending, structured activities and social-prescribing approaches may increase opportunities for connection, while psychological interventions can address maladaptive expectations, negative social interpretations and avoidance. Technology-assisted, reminiscence, physical-activity and multicomponent approaches have also been studied.

Contemporary meta-analytic evidence suggests that loneliness itself is modifiable. A 2024 systematic review and meta-analysis of interventions in older adults included 67 studies; meta-analysis of 27 studies involving 1,756 participants suggested a moderate overall reduction in loneliness (standardised mean difference [SMD] −0.47, 95% CI −0.62 to −0.32), although between-study heterogeneity was substantial.¹⁷ Another review of community-living older adults found moderate-certainty evidence that group-based interventions reduced loneliness, with a smaller pooled effect among randomised trials (SMD −0.27, 95% CI −0.48 to −0.08).¹⁸ A subsequent network meta-analysis also found an overall reduction in loneliness across intervention studies, while highlighting substantial variation in content and delivery.¹⁹

The crucial distinction for longevity medicine is between reducing loneliness and preventing cognitive decline. A 2024 systematic review specifically examining cognitive outcomes of interventions targeting loneliness or social isolation identified only nine studies involving 1,025 participants; most randomised trials were judged to have high risk of bias.²⁰ This evidence base is insufficient to conclude that reducing loneliness preserves hippocampal volume, prevents cerebrovascular injury or lowers dementia incidence. Similarly, experimental pharmacological approaches directed at social-affective neurobiology remain investigational and should not be presented as established neuroprotective treatments.

The current evidence therefore supports a proportionate clinical position. Identifying persistent loneliness may reveal an important psychosocial vulnerability and an opportunity to improve quality of life, mental health and engagement with health-promoting behaviours. It is not yet justified to promise that treating loneliness will prevent dementia. Trials with longer follow-up, validated loneliness measures, objective cognitive outcomes and mechanistic biomarkers are needed to determine whether sustained reductions in loneliness translate into preservation of cognitive health.

8. Implications for Longevity Medicine

Longevity medicine frequently focuses on measurable biological risk factors such as blood pressure, lipids, glycaemia, body composition and physical activity. The evidence reviewed here suggests that social and psychological health should not be ignored when evaluating healthspan. Persistent loneliness may identify people with poorer cognitive trajectories and may interact with sleep, mood, physical activity, vascular health and cognitive reserve. The 2024 Lancet Commission on dementia prevention, intervention and care provides an important evidence-based framework for considering modifiable dementia risk factors in UK clinical practice: it includes social isolation among 14 potentially modifiable risk factors.²¹ Crucially, the Commission identifies objective social isolation, not subjective loneliness, as that risk factor; the two constructs should not be treated as interchangeable.

A pragmatic longevity assessment could therefore include brief enquiry about social connection alongside established medical and behavioural risk factors, potentially using the Office for National Statistics (ONS) recommended loneliness questions or the three-item UCLA Loneliness Scale where structured assessment is appropriate, particularly in older adults or individuals experiencing bereavement, functional decline or mental-health symptoms. The purpose should be to identify unmet need rather than to medicalise normal periods of solitude. Where significant loneliness is present, clinicians should consider potentially remediable contributors such as depression, anxiety, sensory impairment, mobility limitations and adverse social circumstances, and signpost or refer to appropriate psychological, community and social support.

This framing also guards against overstatement. Loneliness is not a biomarker in the same sense as serum cholesterol, and no validated threshold currently identifies a level at which a specific neuroprotective intervention should be prescribed. Its value may instead lie in broadening preventive medicine from a predominantly biological model towards one that recognises social connection as one component of healthy ageing. In this model, loneliness is clinically relevant not because it provides a deterministic prediction of dementia, but because it may identify a modifiable dimension of health that intersects with several established determinants of cognitive ageing.

9. Limitations of the Evidence and Future Directions

Several limitations constrain interpretation of the literature. First, loneliness is measured inconsistently, ranging from single questions to validated multi-item scales. Second, social isolation and loneliness are sometimes conflated despite representing different constructs. Third, many neuroimaging studies are cross-sectional and cannot establish temporal direction. Fourth, animal models of social isolation provide mechanistic insights but cannot be assumed to reproduce the subjective human experience of loneliness. Fifth, residual confounding and reverse causation remain major concerns in dementia research. Finally, intervention studies frequently prioritise loneliness scores rather than long-term cognitive or neurological outcomes.

Future research should distinguish transient from persistent loneliness, incorporate repeated measurements across the life course and recruit culturally and socioeconomically diverse populations. Prospective studies integrating neuroimaging, vascular measures, inflammatory markers, sleep, psychiatric symptoms and detailed cognitive phenotyping could clarify pathways. Genetic and quasi-experimental approaches may help strengthen causal inference, although these methods have their own assumptions and limitations. Most importantly, adequately powered intervention trials should test whether reducing persistent loneliness changes cognitive trajectories rather than loneliness scores alone.

10. Conclusion

Loneliness is more than an unpleasant emotional state: it is consistently associated with differences in social-cognitive processing, poorer cognitive performance and increased risk of later cognitive impairment and dementia. The neurobiological picture is complex. Evidence implicates prefrontal and default-mode networks, hippocampal and temporal structures, stress-related circuitry, inflammatory pathways and cerebrovascular mechanisms, but no single pathway explains the association and causality remains uncertain.

For longevity medicine, the most defensible conclusion is that persistent loneliness is a potentially modifiable marker and risk factor relevant to cognitive healthspan. Addressing loneliness is already worthwhile for psychological and social well-being; whether doing so can prevent neurodegeneration remains an important unanswered question. Incorporating social health into a multidimensional model of prevention, while avoiding causal overstatement, may help longevity medicine pursue not merely longer life but longer life with preserved cognition, independence and meaningful connection.

Declarations

Funding: No specific funding was received for this work.

Conflicts of interest: The author declares no conflicts of interest.

Ethics approval: Not applicable. This article is a narrative review of published literature.

Consent: Not applicable.

Data availability: No new datasets were generated or analysed for this article.

Acknowledgements: None.

References

1. Peplau LA, Perlman D, editors. Loneliness: A Sourcebook of Current Theory, Research and Therapy. New York: Wiley; 1982.

2. Lam JA, Murray ER, Yu KE, et al. Neurobiology of loneliness: a systematic review. Neuropsychopharmacology. 2021;46:1873–1887. doi:10.1038/s41386-021-01058-7.

3. Fan K, Seah B, Lu Z, Wang T, Zhou Y. Association between loneliness and mild cognitive impairment in older adults: a meta-analysis of longitudinal studies. Aging Ment Health. 2024;28(12):1650–1658. doi:10.1080/13607863.2024.2358079.

4. Luchetti M, Aschwanden D, Sesker AA, et al. A meta-analysis of loneliness and risk of dementia using longitudinal data from >600,000 individuals. Nat Ment Health. 2024;2(11):1350–1361. doi:10.1038/s44220-024-00328-9.

5. Blandl F, Eisenberger NI. The lonely brain: aligning theories of loneliness with data from neuroimaging studies. Ann N Y Acad Sci. 2025;1547(1):75–87. doi:10.1111/nyas.15330.

6. Bicks LK, Koike H, Akbarian S, Morishita H. Prefrontal cortex and social cognition in mouse and man. Front Psychol. 2015;6:1805. doi:10.3389/fpsyg.2015.01805.

7. Cristofori I, Pal S, Zhong W, Gordon B, Krueger F, Grafman J. The lonely brain: evidence from studying patients with penetrating brain injury. Soc Neurosci. 2019;14(6):663–675. doi:10.1080/17470919.2018.1553798.

8. Düzel S, Drewelies J, Gerstorf D, et al. Structural brain correlates of loneliness among older adults. Sci Rep. 2019;9:13569. doi:10.1038/s41598-019-49888-2.

9. Tao Q, Akhter-Khan SC, Ang TFA, et al. Different loneliness types, cognitive function, and brain structure in midlife: findings from the Framingham Heart Study. EClinicalMedicine. 2022;53:101643. doi:10.1016/j.eclinm.2022.101643.

10. Park HY, Kim H, Kwak S, Youm Y, Chey J. Association between loneliness and memory function through white matter hyperintensities in older adults: the moderating role of gender. Behav Sci (Basel). 2023;13(10):869. doi:10.3390/bs13100869.

11. Lee H, Yong SY, Choi H, Yoon GY, Koh S. Association between loneliness and cognitive function, and brain volume in community-dwelling elderly. Front Aging Neurosci. 2024;16:1389476. doi:10.3389/fnagi.2024.1389476.

12. Salinas J, Beiser AS, Samra JK, et al. Association of loneliness with 10-year dementia risk and early markers of vulnerability for neurocognitive decline. Neurology. 2022;98(13):e1337–e1348. doi:10.1212/WNL.0000000000200039.

13. Morr M, Liu X, Hurlemann R, Becker B, Scheele D. Chronic loneliness: neurocognitive mechanisms and interventions. Psychother Psychosom. 2022;91(4):227–237. doi:10.1159/000524157.

14. Zhao YL, Zhang DD, Gao PY, et al. Associations of social isolation and loneliness with neurological disorders, psychiatric disorders, brain structures and behavioural phenotypes among UK Biobank participants. Nat Commun. 2026;17(1):5818. doi:10.1038/s41467-026-72529-y.

15. Sundström A, Adolfsson AN, Nordin M, Adolfsson R. Loneliness increases the risk of all-cause dementia and Alzheimer’s disease. J Gerontol B Psychol Sci Soc Sci. 2020;75(5):919–926. doi:10.1093/geronb/gbz139.

16. Lee JH, Sutin AR, Hajek A, et al. Loneliness and cognition in older adults: a meta-analysis of harmonized studies from the United States, England, India, China, South Africa, Mexico, and Chile. Psychol Med. 2025;55:e58. doi:10.1017/S003329172500011X.

17. Duffner LA, Janssen N, Deckers K, et al. Facing the next ‘geriatric giant’—a systematic literature review and meta-analysis of interventions tackling loneliness and social isolation among older adults. J Am Med Dir Assoc. 2024;25(9):105110. doi:10.1016/j.jamda.2024.105110.

18. Shekelle PG, Miake-Lye IM, Begashaw MM, et al. Interventions to reduce loneliness in community-living older adults: a systematic review and meta-analysis. J Gen Intern Med. 2024;39(6):1015–1028. doi:10.1007/s11606-023-08517-5.

19. Sun Y, Huang H, Cai G, et al. Interventions to reduce loneliness among community-dwelling older adults: a network meta-analysis and systematic review. J Am Med Dir Assoc. 2025;26(3):105441. doi:10.1016/j.jamda.2024.105441.

20. Baptista C, Silva AR, Lima MP, Afonso RM. Evaluating the cognitive effects of interventions to reduce social isolation and loneliness in older adults: a systematic review. Aging Ment Health. 2024;28(12):1767–1776. doi:10.1080/13607863.2024.2374933.

21. Livingston G, Huntley J, Liu KY, et al. Dementia prevention, intervention, and care: 2024 report of the Lancet standing Commission. Lancet. 2024;404(10452):572–628. doi:10.1016/S0140-6736(24)01296-0.

How to cite

Roy N. Loneliness, Brain Ageing and Cognitive Healthspan: Neurobiological Mechanisms, Cognitive Consequences and Opportunities for Prevention. British Journal of Longevity Medicine. 2026;1(1). Autumn–Winter 2026. DOI pending.

Copyright and licence

Copyright: BILM (British Institute for Longevity Medicine)
Licence: This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Download PDF

PDF of this article: BJLM 2026 1(1) Roy – Loneliness, Brain Ageing and Cognitive Healthspan (PDF)

View PDF Download PDF