A new study from Stanford University sheds light on why some older adults maintain strong memory while others struggle, even when both groups appear equally healthy. Researchers found that differences in memory performance among cognitively healthy seniors can be explained by two separate brain-based pathways: one involving early biological changes associated with Alzheimer’s disease, and another related to the brain’s capacity to focus attention. The study, published in Science Advances, shows that both factors independently influence how sharply the brain encodes new experiences, which ultimately affects how well they are remembered.
Alzheimer’s disease is a progressive neurological condition that affects memory, thinking, and behavior. It is the most common cause of dementia in older adults. Long before symptoms emerge, the brain begins accumulating abnormal proteins, particularly tau and amyloid beta. These proteins disrupt normal brain function, especially in areas critical for memory. Even when individuals seem cognitively normal, early biological signs of Alzheimer’s can already be present and silently influence how the brain processes information.
Memory tends to decline with age, but not all older adults experience this decline in the same way. Some remain sharp well into their 80s, while others struggle with daily memory tasks despite showing no signs of dementia. This variability intrigued the Stanford research team. Working within the Knight Initiative for Brain Resilience, the team aimed to understand why these differences arise. They focused on “neural selectivity,” or how precisely the brain represents different types of information, as a key factor in memory performance.
“As individuals age, declines in episodic memory are often one of the first observable signs of cognitive changes. However, there is considerable variability in memory abilities among older adults,” explained study author Jintao Sheng, a postdoctoral fellow. “While some maintain good memory performance, others experience significant decline. Understanding the factors that contribute to these individual differences is crucial, as it can guide interventions aimed at improving the life quality for older adults. In this study, we focus on understanding the causes of these differences in episodic memory, particularly from the perspectives of age, brain function, and early Alzheimer’s disease risk factors.”
The study drew on data from 166 older adults, aged 60 to 88, who participated in the Stanford Aging and Memory Study. All participants were cognitively normal, with no signs of dementia. During the study, participants completed a memory task while undergoing brain scans using functional magnetic resonance imaging. In the task, they were shown pairs of words and pictures—some of famous faces like Queen Elizabeth, others of iconic places like the Golden Gate Bridge. They were asked to form associations between the words and images. Later, they were tested on how well they could recall those associations.
The researchers measured activity in brain areas known to specialize in processing faces and places. They looked at how selectively these areas responded to their preferred category, such as whether a “place region” showed greater activity for places than for faces. They also examined a network of fronto-parietal regions known as the dorsal attention network, which helps people focus on relevant information. In addition to brain scans, the researchers collected blood and spinal fluid samples to measure levels of tau and amyloid proteins, which are early markers of Alzheimer’s disease.
The results revealed that older adults with sharper and more distinct brain responses to faces or places—what the researchers call higher neural selectivity—were better at remembering the associations. These distinct neural patterns were more likely to occur when participants were paying close attention during the task. In other words, when the brain’s attention system was more active, the memory-related areas of the brain were better “tuned” to process and encode new information.
“High neural selectivity helps form clear and stable memories, much like tuning a radio to a clear signal,” Sheng told PsyPost. “However, as we age, this tuning becomes less precise, leading to ‘blurred’ memory traces, which we refer to as neural dedifferentiation. We found that neural selectivity declines with age and toxic protein accumulation, but it improves with higher attention activity, which ultimately influences memory performance. By systematically exploring the interplay among these factors, we provide comprehensive evidence for why some cognitively unimpaired older adults remember better than others.”
However, the researchers also found that higher levels of a specific form of tau protein, known as phosphorylated Tau181, a marker of Alzheimer’s pathology, were linked to lower neural selectivity, especially in brain regions involved in place recognition. This means that even in people who had no symptoms of dementia, early signs of Alzheimer’s biology were already making the brain’s memory encoding less precise.
Attention and Alzheimer’s pathology influenced memory independently. One pathway to poor memory involved lower activity in the brain’s attention network, which reduced how clearly information was encoded. The other pathway involved elevated tau protein levels, which also interfered with encoding by degrading the quality of the brain’s representations of events. Some individuals showed both problems, while others had one or neither.
“Memory decline in older adults can result from various factors, such as the buildup of toxic proteins in the brain or declines in the brain’s ability to differentiate between different types of information (at least these two factors have been supported by our study),” Sheng explained. “However, from a healthy aging perspective, staying engaged in learning, such as by paying close attention and consistently reminding yourself of your learning goals, can activate the brain’s attention centers, helping you continue learning, even if it’s a bit challenging. This approach remains effective, regardless of your age or the accumulation of toxic proteins in your brain.”
Statistical models confirmed that these two pathways—attention and tau pathology—each explained unique portions of the variability in memory performance. In fact, the strength of the brain’s attention network was the strongest predictor of neural selectivity, followed by age and tau levels. This suggests that even in the presence of early Alzheimer’s-related changes, strong attentional engagement can partially protect memory performance.
“One thing that surprised me was that factors like attention and goal-setting seemed to actually mitigate some of the negative effects of aging,” Sheng said. “This suggests that, even with the accumulation of toxic proteins in the brain, individuals can still perform well if their brain’s attention centers are more active during the task. Furthermore, within each individual, attention-related brain activity varied over time. Better memory encoding was linked to periods when the brain’s attention centers were more active. These findings open up new directions for future research on how cognitive strategies might help protect against memory decline.”
The researchers also found that neural selectivity not only predicted how well participants performed on the immediate memory task, but also how they did on other memory assessments given weeks apart. These included tests of delayed recall and the ability to distinguish similar objects—skills important in everyday life. Neural selectivity was not linked to other mental skills like task switching or verbal fluency, suggesting it specifically supports memory.
But as with all research, there are limitations to consider. “One limitation is that the measure of attention used in this study is indirect,” Sheng noted. “Specifically, we relied on neural data from the dorsal attention network, which supports top-down attention in the brain. While this network is associated with attention regulation, it doesn’t directly or purely measure attention itself, and other factors may influence attention that we didn’t capture.”
“As a result, the conclusions we draw about the role of attention are based on an inferred relationship rather than a direct measurement of attentional processes. However, we do have behavioral data that measures sustained attention, and we plan to explore this further in future research.”
The Stanford Aging and Memory Study is ongoing, and researchers plan to follow these participants for several more years to see how their memory and brain health change over time. This will allow them to explore whether neural selectivity can predict who maintains memory performance and who experiences decline.
“The Stanford Aging and Memory Study, a landmark longitudinal project on cognitive aging co-led by Anthony Wagner and Elizabeth Mormino and supported by the National Institute on Aging, continues to gather long-term follow-up data (including imaging and cognitive tests over seven years),” Sheng explained. “Our goal is to answer a crucial question in aging research: Which early brain and biological signals predict memory decline or resilience over time?”
“We aim to deepen our understanding of the mechanisms behind changes in brain function and identify additional factors contributing to memory changes. A key next step is using tau PET imaging to explore how toxic tau proteins accumulate in the brain and disrupt memory circuits. We are also investigating how the hippocampus interacts with cortical areas in aging brain to influence memory encoding and retrieval.”
The study, “Top-down attention and Alzheimer’s pathology affect cortical selectivity during learning, influencing episodic memory in older adults,” was authored by Jintao Sheng, Alexandra N. Trelle, America Romero, Jennifer Park, Tammy T. Tran, Sharon J. Sha, Katrin I. Andreasson, Edward N. Wilson, Elizabeth C. Mormino, and Anthony D. Wagner.
