A new study published in the * Journal of Affective Disorders* suggests that the advanced brain aging often seen in people with anxiety and depression is partly associated with variations in their cognitive performance. By accounting for cognitive skills like memory and processing speed, scientists observed that the apparent effect of these psychiatric conditions on brain aging decreased by roughly twenty to twenty-five percent. This indicates that cognitive differences play an important role in understanding brain health in individuals with neuropsychiatric conditions.
Biological aging of the brain can sometimes diverge from chronological aging. Using structural magnetic resonance imaging, machine learning algorithms can predict a person’s brain age by comparing their brain structure to a large dataset of healthy individuals. The difference between this predicted age and the person’s actual age is called the brain age gap. A positive gap indicates an older-appearing brain, which is linked to cognitive decline, health risks, and various neurological conditions.
Neuropsychiatric disorders like depression and anxiety are associated with increased brain age gaps. People with these conditions also frequently experience cognitive difficulties that affect their attention, executive function, and processing speed. Executive function refers to a set of mental skills that include working memory, flexible thinking, and self-control. Because cognitive decline is a common feature of mood disorders, it is often difficult to tell if brain aging differences reflect the psychiatric diagnosis itself or the accompanying cognitive variation.
Previous neuroimaging research typically looked at global brain age, which assumes aging happens uniformly across the entire brain. This approach can obscure specific regional effects.
“Our main motivation was that most previous brain-age studies summarize the entire brain using a single number,” said Owen M. Vega, a doctoral candidate in the neuroscience graduate program at the University of Southern California and a researcher at the Ethel Percy Andrus Gerontology Center in the Leonard Davis School of Gerontology.
“While useful, that approach assumes the brain ages uniformly and provides little insight into why certain disorders are associated with advanced brain aging. We wanted to move beyond prediction toward biological understanding,” Vega told PsyPost.
By mapping brain age at a regional level and accounting for cognitive performance, the researchers aimed to identify the specific neural systems affected and link them to underlying cellular processes.
“Ultimately, this brings us closer to understanding the biological mechanisms that contribute to psychiatric brain aging rather than simply measuring that it exists,” Vega explained.
Add PsyPost to your preferred sources The authors analyzed data from 21,424 older adult participants in the UK Biobank. Participants were classified into four mutually exclusive groups based on their diagnostic status. The sample included 12,285 individuals with no psychiatric diagnosis, 1,746 with anxiety only, 4,267 with depression only, and 1,563 with comorbid anxiety and depression. Comorbidity means the individual met the criteria for both conditions.
To estimate regional brain ages, the scientists processed structural brain scans through a deep neural network, breaking the brain down into 187 distinct cortical and subcortical regions. Participants also completed several cognitive assessments measuring fluid intelligence, reaction time, and symbol substitution. These test results were statistically combined into a single principal component score representing general cognitive performance. The models also controlled for participant sex, years of education, and socioeconomic deprivation to isolate the variables of interest.
The researchers first ran a statistical model that did not account for cognitive performance. They found widespread regional brain age gap elevations across the psychiatric groups compared to the diagnosis-free participants. On average, brains in the anxiety group appeared about 1.01 years older than chronological age. Brains in the depression group appeared 1.05 years older, and brains in the comorbid group appeared 1.14 years older.
These elevated brain ages were widely distributed but particularly pronounced in specific areas. The largest gaps were observed in the anterior frontal and orbitofrontal regions, as well as the temporal pole. These areas are heavily involved in emotion regulation and reward processing.
Vega noted that while the overall increases are relatively small, they provide a starting point for exploring the biology of mental health.
“The effects are statistically robust but modest in size, with average differences of about one year. However, they should not be interpreted as the whole story,” Vega told PsyPost. “Averaging across the entire brain masks much larger regional differences. The real significance of this work lies in identifying where these changes occur.”
By pinpointing these spatial patterns, scientists can relate them to specific genes and molecular pathways, moving the field past simple summary measures.
“This moves brain-age research beyond a single summary measure toward understanding the mechanisms that may contribute to psychiatric illness and cognitive vulnerability,” Vega added.
Next, the authors ran a second model that included the participants’ general cognitive performance scores. Factoring in cognition reduced the magnitude of the brain age gaps by approximately twenty to twenty-five percent. The mean gap dropped to 0.80 years for the anxiety group, 0.84 years for the depression group, and 0.78 years for the comorbid group. Despite this reduction, the effects remained present, indicating that diagnostic status contributes to brain aging independent of cognitive ability.
“The main takeaway is that anxiety and depression are associated with subtle but measurable differences in how the brain ages, and those differences are not spread evenly across the brain,” Vega said.
By showing how these estimates change when mental skills are factored into the equations, the study refines how scientists understand brain health in clinical populations.
“We also found that part of the observed brain-age signal is associated with cognitive performance, showing that cognition is an important piece of the picture,” Vega explained. “More broadly, our work suggests that brain aging in psychiatric disorders reflects specific biological patterns rather than a single, uniform process, which may ultimately help researchers develop more biologically meaningful biomarkers.”
Higher cognitive performance was associated with a younger-looking brain, suggesting a protective effect. This association was noticeably stronger in all three psychiatric groups compared to the diagnosis-free participants. Interestingly, the brain regions most strongly associated with cognitive performance differed from the regions most affected by the psychiatric diagnoses.
Cognitive associations were strongest in subcortical and ventral regions of the brain. These included the thalamus, pallidum, and hippocampus, which are structures located deep beneath the cerebral cortex that are essential for memory formation and information integration. This dissociation suggests that psychiatric status and cognition exert distinct but overlapping influences on different neural systems.
The researchers also looked beyond the magnetic resonance imaging scans to see if their regional brain age maps aligned with other biological data, such as transcriptomics. Transcriptomics is the study of RNA molecules in cells, which reveals how specific genes are turned on or off to drive cellular activity.
“One of the most striking findings was that several independent biological analyses converged on the same underlying systems,” Vega said. “Regional brain-aging patterns identified from MRI aligned with transcriptomic enrichment and biological pathways in a remarkably consistent way.”
This overlap suggests that the structural differences visible on brain scans are directly tied to cellular and genetic changes.
“That convergence gives us greater confidence that these patterns reflect meaningful biology rather than isolated statistical findings, and suggests that regional brain age can serve as a bridge between neuroimaging and molecular neuroscience,” Vega added.
The cross-sectional design of the study relies on data collected at a single point in time. This prevents researchers from establishing the sequence of events.
“A key caveat is that these results are not causal. Our findings do not demonstrate that anxiety or depression directly accelerate brain aging,” Vega said. “Instead, they identify patterns of brain-aging vulnerability associated with psychiatric illness and cognitive performance.”
Tracking individuals over multiple years is necessary to determine if cognitive differences precede advanced brain aging or reflect the downstream consequences of an aging brain. Bidirectional influences are highly likely in these conditions.
“Longitudinal studies will be needed to determine how these relationships evolve over time and whether they predict future cognitive decline,” Vega explained. “The goal was to refine the interpretation of previous brain-age findings and pave the way to clinical research, not to claim a direct mechanism.”
The diagnostic classifications were derived from a combination of self-reported surveys and clinician-confirmed records. The available data lacked details regarding symptom severity, illness duration, and the age of onset. The researchers were unable to determine if the older brain ages were linked to more severe, chronic, or recurrent forms of mental illness. Residual misclassification or reporting bias might also introduce variability into the data.
The UK Biobank predominantly consists of White European participants who are often healthier than the general population. This demographic makeup limits how well these findings apply to more diverse groups worldwide. Environmental factors, cultural differences, and early-life stressors that influence brain aging were not fully captured in the dataset. Future research should prioritize replicating these findings in more ethnically diverse cohorts.
Future research will continue to explore the genetic and molecular factors that drive these localized brain changes.
“Our next step is to relate regional brain-age maps to other spatially organized biological data,” Vega said. “We are now integrating regional brain-age maps with transcriptomic, genetic, and cellular datasets to identify the biological pathways associated with vulnerability to psychiatric brain aging.”
By building a more comprehensive biological profile, the team aims to improve risk assessments for aging adults.
“Ultimately, we hope this work will improve biologically informed risk stratification, help identify individuals at greatest risk for later cognitive decline, and reveal biological systems that may become targets for future therapeutic interventions,” Vega concluded.
The study, “Cognitive performance modulates regional brain age differences in clinical anxiety and depression,” was authored by Owen M. Vega, Phoebe Imms, Nikhil N. Chaudhari, Wendy J. Mack, Nahian F. Chowdhury, and Andrei Irimia.