The scientific literature has long demonstrated cognitive change as a natural part of aging. The dynamic and variable longitudinal changes in cognitive function that occur inherently during the aging process are referred to as cognitive aging (Harada et al., 2013). However, those who maintain their cognitive function at high levels, even with advancing age, are categorized as successful cognitive aging (Daffner, 2010).

The cascade model of cognitive aging suggested by Birren and Cunningham (1985) emphasized a life course approach to cognitive aging and cognitive performance. According to the model, primary aging is characterized by a steady deterioration in mental function, which is often accompanied by problems with memory (particularly new learning and retention), information processing, language, and other cognitive skills. Secondary aging refers to a loss of fluid and crystallized cognitive capacities caused by a disease process such as dementia, whereas tertiary aging refers to impairments in cognitive function caused by total biological devitalization of the organism before death (Birren and Cunningham, 1985).

Brayne and Calloway (1988), on the other hand, depict cognitive decline as a continuum, ranging from normal and successful aging to moderate cognitive impairment and dementia. This model depicts a general downward trend in all elements of cognitive ability, regardless of the competing danger of neuropathological alterations like dementia. However, Krivanek et al. (2021) suggested a new model of cognitive decline that depicts the progression of cognitive decline. In people with neurodegenerative diseases, this theoretical curve would shift to the left. On the other hand, boosting cognitive or brain reserve would shift this theoretical curve to the right, allowing patients to reach this threshold later in life (Figure 1).

Earlier neurocognitive models, which blend behavioral and neurological evidence to develop a conceptual model of cognitive aging, are based on neural compensation. These models, such as the Hemispheric Asymmetry Reduction in Older Adults (HAROLD; Cabeza, 2002), the Compensation-Related Utilization of Neural Circuits Hypothesis (CRUNCH, Reuter-Lorenz and Cappell, 2008), the Posterior-Anterior Shift in Aging (PASA; Davis et al., 2008), and the Scaffolding Theory of Aging And Cognition (STAC and STAC-r; Park and Reuter-Lorenz, 2009; Reuter-Lorenz and Park, 2014), postulate that older adults can perform as well as young adults on cognitive tasks depending on their capacity to recruit additional neural networks. These models, however, were unable to fully explain the cognitive deterioration that happens with healthy aging. Ebaid and Crewther (2020) then offered a theory of cognitive aging based on a system biology approach that combines the sensory deprivation hypothesis, the information degradation hypothesis, and the common-cause hypothesis. The theory stressed the significance of including all of the biological changes that frequently occur at a later age (Ebaid and Crewther, 2020).

Many studies on cognitive aging including its theories have been conducted in the past but only a few have kept track of the literature. The impact of literature on future research could be determined by bibliometric analysis, which is a quantitative analysis of publication metadata. The application of bibliometric approaches in the scientific and professional community has progressed much beyond the basic concept of simple lists of scientific production or citation indexing, and there is a wide range of applications across disciplines (Ellegaard, 2018). This is owing to advancements in bibliometric software such as VOSviewer, Gephi, and Leximencer, as well as the availability and accessibility of scientific databases like Web of Science and Scopus (Donthu et al., 2021).

Previous bibliometric studies have investigated general aspects of aging, namely, aging or oldest age or geriatric (Lund and Wang, 2020; Gonzalez-Alcaide et al., 2021), healthy aging (Gu et al., 2019), aging in combination with other issues such as reception by the scientific community (Glänzel and Schoepflin, 1995), physical therapy (Arnal-GÓmez et al., 2020), geriatric nursing (Ghamgosar et al., 2021), mobile technologies (Tajudeen et al., 2022), safety in-home care (Cao et al., 2021), subjective well-being (Dominko and Verbič, 2019), and specific to aging policies in China (Nan et al., 2020). This bibliometric analysis and network visualization, on the other hand, was carried out to explore the literature on cognitive aging in the Scopus database. It aimed to answer the following research questions:

Therefore, in the present study, we attempted to reveal the publishing trends, scientific productivity patterns, the most productive authors, collaboration status, research patterns across the sources, and the major areas of cognitive aging research.

This study analyzed the main bibliometric indicators to profile the research landscape on cognitive aging from 1956 to 2021.

The cognitive aging theory was first introduced by Welford and Birren in 1965 (Birren, 1965). Before 1965, cognitive aging research was descriptive, determining which areas of intellectual performance are impaired in older vs. younger persons (Anderson and Craik, 2017). Over the past three decades, the growing number of publications meant that research on cognitive aging was gaining traction around the world. This was in line with the findings on healthy aging and geriatric nursing research reported by Gu et al. (2019) and Ghamgosar et al. (2021), respectively. Furthermore, this continual increase in research has important clinical and intellectual implications, as it aids clinicians in better measuring, preventing, and treating cognitive aging by establishing individualized risk profiles connected to a personalized intervention strategy (Ryan et al., 2019). The total citations peaked in 2011, suggesting this to be the key year for the development of the field (Cao et al., 2021). Further investigation revealed that the National Institute on Aging-Association Alzheimer’s working groups on Alzheimer’s disease diagnostic criteria issued a set of recommendations for diagnosing mild cognitive impairment attributable to Alzheimer’s disease in 2011 (Albert et al., 2011).

The United States was the most prolific country followed by the United Kingdom, Canada, Germany, and Australia. The productivity of the top five countries was in line with the recent study conducted by Arnal-GÓmez et al. (2020). The authors suggested that productivity was related to the aging of their population, as shown by the positive correlation between productivity and the aged population. However, in terms of scientific productivity pattern, as the number of publications contributed increased, the number of authors declined. This was consistent with Lotka’s Law (Lotka, 1926), which stated that single-publication authors are far more likely to conduct subsequent research on similar research areas (Rowlands, 2005; Kushairi and Ahmi, 2021).

The collaboration index was on the rise, reflecting the growing complexity of multidisciplinary research and the increasing quantity and quality of the resultant publications in cognitive aging (Stallings et al., 2013). According to several studies, research conducted by larger groups is more influential and impactful (Crane, 1972; Goffman and Warren, 1980). Furthermore, articles co-authored by international collaborators receive more citations than articles co-authored by domestic collaborators, implying that internationally co-authored articles represent a larger segment of global science (Narin et al., 1991). Dutt and Nikam (2015), however, reported that publications from certain prolific countries and institutions emerging from domestic collaboration resulted in a higher impact than those from international collaboration.

Based on the keyword co-occurrence analysis, there are three main clusters. In the analytical (individual-based) studies cluster, the main keywords were grouped into: (i) neuroimaging which includes various modalities; (ii) cognitive trajectories that are related to the major study groups; and (iii) cognitive domains that are related to the cognitive assessment frequently used (Table 7). The mature keywords are Alzheimer disease, mild cognitive impairment (MCI), magnetic resonance imaging (MRI), functional MRI (fMRI), and working memory, while the core keywords include cognitive aging, executive function, cognitive reserve, healthy aging, and diffusion tensor imaging (DTI). These keywords highlight the main group of individuals, established neuroimaging, and cognitive domains related to analytical (individual-based) studies.

Individuals in the aging population vary greatly, and while some develop cognitive impairment (including mild cognitive impairment), Alzheimer’s disease, and other types of dementia, others may retain their cognitive function to a major extent well into old age, which is also known as healthy aging (Nyberg et al., 2012). Reduced brain volume, cortical thinning, and deterioration in white matter microstructure are common age-related structural alterations (Fjell and Walhovd, 2010), which can contribute to lower cognitive performance in domains like executive function, memory, and processing speed (Nyberg et al., 2012; Grady et al., 2016; Cabeza et al., 2018). The cognitive reserve theory attempts to explain why some people can sustain cognitive performance while having a disease or aging-related brain abnormalities. Individuals with a larger cognitive reserve are thought to process information more efficiently, allowing them to functionally adapt to brain aging and sustain greater disease before cognitive deficits appear (Stern, 2002).

While the emerging keywords include functional connectivity, human immunodeficiency virus (HIV), decision making, gender, compensation, and default mode network, it is generally known that functional connectivity alterations associated with Alzheimer’s disease start years before structural changes and clinical symptoms are noticed (Cieri and Esposito, 2018). In persons at risk of developing Alzheimer’s disease, some resting-state fMRI studies have found increased functional connectivity between certain regions of the default network, while others have found decreased connectivity (Cieri and Esposito, 2018). Overactivation in functional connectivity across resting-state networks may be related to compensatory mechanisms even in cognitively preserved older adults, according to some studies (Li et al., 2015; Grady et al., 2016; Fjell et al., 2017). More advanced neuroimaging techniques with a higher spatial-temporal resolution, as well as methods to measure neurotransmitter activity or gene expression in real-time, may be developed, allowing for a better knowledge of the brain factors associated with cognitive aging and a new avenue for intervention (Anderson and Craik, 2017).

With a higher number of older individuals living with HIV in the era of antiretroviral therapy, there is a higher likelihood of cognitive decline, particularly in executive function, processing speed, vocabulary, recollection, and motor/psychomotor domains (Deng et al., 2021). In the available research, there is some evidence for premature and accelerated cognitive aging among HIV individuals, particularly in large and longitudinal studies and those with a higher number of older samples. Future HIV and cognitive aging studies will need to standardize neuropsychological testing methodologies and outcomes, as well as use a large sample from collaborative multi-centers (Aung et al., 2021).

Decision-making deficit has been shown even in cognitively healthy older adults (Spreng et al., 2016; Bangma et al., 2017). It may increase vulnerability to fraud (Duke Han et al., 2016; Lamar et al., 2020), including financial exploitation. Weissberger et al. (2020) found that perceived financial exploitation in old age is linked to differences in whole-brain functional connectivity involving the hippocampus, insula, and medial frontal cortex, which is consistent with models linking age-related changes in decision-making and social cognition to financial exploitation.

There is still debate on the gender difference in cognitive performance with a particular interest in the older population. Previous studies have indicated that sex variations in cognitive performance remain until late adulthood (Siedlecki et al., 2019), as well as an unbalanced prevalence of neurodegenerative diseases associated with different cognitive impairments, for example, males are more likely to suffer from MCI and Parkinson’s disease, while females are more likely to suffer from Alzheimer’s disease (Cholerton et al., 2018; Sohn et al., 2018). Different interrelationships between cognitive functions could potentially explain sections of these different age-related trajectories, presenting a promising study topic.

Large-scale functional brain networks have also been used to investigate neurocognitive aging (Damoiseaux, 2017). Internally directed cognitive processes that rely on access to prior-knowledge representations to guide goal-directed behaviors generally engage the default network (Andrews-Hanna et al., 2014). During the performance of externally directed tasks, however, default-network are suppressed (Buckner et al., 2008). Reduced suppression, decreased within-network connectivity, and increased between-network connectivity are all age-related alterations in the default network, all of which are minimally controlled by task context (Spreng and Schacter, 2012; Rieck et al., 2017). These led to a default-executive coupling hypothesis of aging proposed by Spreng and Turner (2019). This hypothesis was based on findings that the lateral prefrontal cortex, responsible for executive function and cognitive control, is functionally coupled with engagement of the default network in old age.

In the descriptive (population-based) studies cluster, the main keywords were grouped into: (i) risk factors; and (ii) behavioral assessment. The mature keywords in this cluster include cognition, depression, neuropsychology, normal aging, and risk factor, while the core keywords were cognitive decline, dementia, cognitive function, elderly, and epidemiology. These keywords highlight the common types of population-based studies (epidemiology and longitudinal), established risk factors (elderly, depression, etc.), and behavioral assessments (cognitive function, frailty, mini-mental state examination (MMSE), etc.).

The emerging keywords in this cluster were neuropsychological assessment, frailty, mortality, MMSE, anxiety, and gait speed. These keywords reflect different behavioral assessments, frailty, gait speed, and MMSE, frequently used in population-based studies. The International Academy on Nutrition and Aging and the International Association of Gerontology and Geriatrics defined cognitive frailty as comorbid physical frailty (>1 Fried criteria) and mild cognitive impairment (Petersen criteria; Kelaiditi et al., 2013; Rivan et al., 2020). Physical frailty such as gait speed and handgrip strength has been linked to cognitive decline in older persons in many previous studies (Demnitz et al., 2016; Kobayashi-Cuya et al., 2018; Chou et al., 2019). Furthermore, a recent theory has suggested a link between cognitive impairment, sensory deprivation, and common-cause hypotheses (Ebaid and Crewther, 2020).

In the experimental studies cluster, the main keywords were grouped into: lifestyle, animal/human models, cognitive markers, and hippocampal functions. The mature keywords in this cluster include aging, memory, hippocampus, diabetes, and learning, while the core keywords were inflammation, brain aging, exercise, oxidative stress, and neuroinflammation. This type of study normally assesses the role of lifestyle in affecting cognitive markers as well as hippocampal functions in animals and humans (Fordyce and Wehner, 1993; Vaynman et al., 2004; Gow et al., 2012; Woodard et al., 2012). The Cam-CAN data set provides a valuable resource that contributes to the expanding understanding of cognitive aging as a lifetime developmental process characterized by intricate interactions across life stages and cognitive domains. Thus, there is a need for large-scale cognitive aging experimental studies to include a wider range of ages and cognitive tasks (Shafto et al., 2020).

Learning and memory, brain-derived neurotrophic factor (BDNF), obesity, D-galactose, and epigenetics were among the emerging keywords in the experimental studies cluster. BDNF, a protein that regulates synaptic transmission and induces long-term changes in excitability and synaptic plasticity in the adult brain, has been shown to have a prominent role in neuron survival, growth, and function in experimental models (Miranda et al., 2019). The BDNF Val66Met polymorphism, which regulates BDNF expression, has been linked to resilience toward the effects of aging on cognition (Collins et al., 2021). In addition, epigenetics has been studied as a possible relationship between environmental/lifestyle factors (hormone status, food, stress, and exercise) and the variability of cognitive function as people age (Barter and Foster, 2018; Beydoun et al., 2020).

With a few exceptions, we have addressed all of the research objectives. To begin with, we only searched one database, Scopus, because it is the most comprehensive database (Zhu and Liu, 2020; Pranckute, 2021) and to avoid variations in data formats and field tags that would occur if we used data from multiple databases. Second, in order to prevent finding unnecessary documents, we run our search in the article title; nonetheless, we may overlook certain significant documents. Third, keyword cleaning and statistics are tailored to our specific needs, which may be limited by our professional knowledge and experience.

The United States continues to dominate in terms of publication and research collaboration in cognitive aging. The journals publishing themes relevant to aging research are the top sources of cognitive aging research. In the coming decades, the hot topics in cognitive aging research would be neuropsychological assessment, functional connectivity, HIV, decision-making, gender, compensation, default mode network, learning and memory, BDNF, obesity, D-galactose, epigenetics, frailty, mortality, MMSE, anxiety, and gait speed. These study findings provide useful references to health practitioners and researchers who are involved in cognitive aging management.

ZO and ASAH planned the study. RZ, KFA, and KNSS collected the data and drafted the manuscript. AHA, AW, and AA revised the manuscript and language. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.