University of Warwick research has shown that the cognitive slowness and disjointedness that comes with aging can be better explained as a symptom of a brain that knows too much (‘cluttered wisdom’) instead of a symptom of a brain that is declining.
Traditional views of cognitive aging suggest that declines in memory, processing speed, and problem-solving are caused by brain degradation. When cognitively tested, older adults struggle to quickly recall specific words, will describe related objects as less similar than younger adults and give more unpredictable answers when asked what they associate with a specific word.
Professor Thomas Hills from the Department of Psychology at The University of Warwick and sole author of this study said: “Degradation is often the default theory of age-related cognitive decline. But it is well established that as people age, they also know more, which is called crystallized intelligence. This research set out to test whether too much knowledge is a better explanation of the evidence for decline than degradation. And as it turns out, it is.”
In this research, published in Psychological Review, Professor Hills used computational modelling to mimic the process of human learning over a lifetime. Over a thousand learning trials later, the model had accumulated knowledge like human memories, with lots of associations between ‘memories’, producing an imitation of an enriched human mind.
However, as the model ‘aged’, the increasing mental complexity produced more competition among memories. Navigating these denser mental connections created similar cognitive effects to aging in this computational model. When tested on the same association tasks often used in clinical aging research, the model had slower performance, saw related things as less similar and produced more unpredictable responses, identical to the performance of older adults.
Professor Hills added: “This work is the result of a decade of research, involving empirically testing older adults, done by myself and many others. As we age, the fruits of life and learning enrich our mental representations, causing increased competition between memories and words in people’s minds. This often leads to better decision making using more integrated knowledge. But the trade-off is slower performance and greater difficulty in learning new things, especially when those things are unrelated to things that are already known well.”
Cognitive decline in older age is not necessarily the result of decay and a faltering memory; instead, it can be seen as the unavoidable consequence of cognitive enrichment. As we accumulate knowledge, it weighs us down, becomes slower to access, and produces more unpredictable results.
Professor Hills said: “Consider what you know about the city in which you live. If you haven’t lived there long, you may only know a few routes to take. Probably the quickest roads from point A to point B. But the longer you live there, the more you will have learned about alternate routes, and interesting places along the way. That adds up to more enriched knowledge, with all of its many assets. But it also requires a greater consideration of this interconnectedness, and the trade-offs that go along with that”.
While this model captures healthy aging as an over-enriching of the mental network, it is different from dementia-associated aging, with well-defined pathological brain matter loss. Dementia-related cognition is still best explained by a degradation of the network of memories.
Intriguingly, as Thomas points out in his recent book, “Behavioral Network Science: Language, Mind, and Society” (Cambridge University Press, 2025), similar consequences of enrichment exist in other complex systems (e.g., ecosystems, economic markets, IT systems), and influence behaviours such as creativity, culture, and innovation.
As computers fill up with interconnected files, and as organisations grow, they slow and take more time performing basic functions that used to be done at speed. This suggests that cognitive aging may be one example of a universal trade-off in highly connected networks.
Professor Hill’s full research paper in Psychological Review can be accessed here.
