Errors are necessary for learning: What the science of memory actually says

UCLA memory expert Dr. Alan Castel explains why mistakes are essential for learning, how superagers defy cognitive aging, and practical tools to improve memory at any age.
The most efficient way to learn something is to get it wrong first. That counterintuitive idea sits at the center of a new episode of the Huberman Lab podcast, where host Andrew Huberman sat down with Dr. Alan Castel, a professor of psychology at the University of California, Los Angeles and a leading researcher in human memory and cognitive aging.
Castel didn't come to praise mistakes for their own sake. He came armed with decades of research showing that errors โ when made in the right context โ actively strengthen the neural pathways required for long-term recall. The conversation covered everything from how memory actually works to why some older adults become "superagers" who improve their cognitive capacity with age. For anyone who has ever felt frustrated by forgetting a name, a fact, or a face, the episode offers a scientifically grounded path forward.
Errors as a learning tool
The premise of the episode is embedded in its headline: "Errors Are Necessary for Learning." That idea directly challenges the instinct most of us have to avoid mistakes. We study for a test by reading notes over and over. We practice a skill by repeating the same correct motion. But Castel's research suggests that the brain learns more effectively when it is forced to retrieve information, even โ and especially โ when that retrieval fails.
The mechanism is called retrieval practice. Instead of passively reviewing material, you actively attempt to recall it. If you get it wrong, the error signals to your brain that the information is both important and missing. That signal triggers a cascade of corrective processes that embed the correct answer more deeply than a dozen rereads ever could. Self-testing, in other words, is not just a way to measure what you know. It is the very act that builds memory.
Castel emphasized that the emotional response to an error โ mild frustration, surprise, or curiosity โ also plays a role. Emotions act as a chemical bookmark, telling the brain that this particular piece of information matters. The combination of active retrieval, failure, and emotional tagging creates a memory that is both durable and accessible.
How memory works and why it fails
Memory is not a video recording. It is a reconstructive process. Every time you recall something, you rebuild it from fragments, and each reconstruction changes the memory slightly. This is why false memories are so common and so convincing. Castel addressed this directly: the same mechanisms that allow us to imagine the future and plan โ drawing on past experiences โ also make us vulnerable to inserting details that never happened.
The brain's default mode network, which is active during daydreaming and future planning, relies entirely on stored memories. You cannot plan a vacation without remembering what a beach looks like. You cannot imagine a conversation without recalling how language sounds. That dependence on the past means that memory errors are baked into human cognition. The trick is not to eliminate errors entirely โ that is impossible โ but to use them as data.
Superagers: the outliers who improve with age
One of the most striking segments of the episode focused on "superagers," a term researchers use for older adults whose cognitive performance doesn't just hold steady but actually improves. These individuals defy the typical trajectory of cognitive aging. Their memory, attention, and processing speed remain closer to those of people decades younger.
Castel discussed what superagers do differently. They tend to stay intellectually curious, engaging in activities that require active learning โ not just crossword puzzles, but new skills that demand real cognitive effort. They often maintain strong social connections, which provide both emotional engagement and the need to remember names, faces, and shared histories. They also practice retrieval regularly, often without knowing they are doing it: they quiz themselves, they engage in debate, they teach others.
The implication is clear. Cognitive decline is not a one-way street. While genetics and brain health matter, the behaviors that superagers adopt are available to almost anyone. The key is consistent, deliberate practice of the kind that produces errors.
Practical tools for any age
Castel's advice for maintaining and improving memory across the lifespan is grounded in the same principles that govern his research. The most powerful tool is retrieval practice. Instead of rereading a list of names before a meeting, try to list them from memory first. Instead of reviewing slides, cover them up and summarize the key points out loud. The errors that emerge are not failures โ they are the raw material for stronger encoding.
Curiosity is another lever. When you are genuinely interested in something, your brain allocates more resources to encoding it. That attention is not passive; it involves asking questions, making predictions, and checking whether you were right. The prediction error โ getting it wrong โ again drives learning.
Emotion, too, can be harnessed deliberately. If you need to remember something important, attach a vivid or emotional cue to it. The brain prioritizes emotional content because it was historically linked to survival. A boring fact paired with a funny image or a strong feeling will outlast the same fact presented dryly.
Finally, Castel addressed the role of sleep. While the episode did not dive into sleep physiology in detail, he noted that consolidation โ the process by which short-term memories become long-term โ happens primarily during sleep. Errors made during the day are sorted and corrected overnight. Without adequate sleep, even the best retrieval practice loses much of its effect.
What this means for the rest of us
The episode positions memory not as a fixed trait but as a skill. It can be trained, maintained, and even improved at any age. The superagers prove that cognitive decline is not inevitable. The research on errors shows that avoiding mistakes โ the very thing most of us strive to do โ is exactly the wrong strategy for building lasting knowledge.
For students, professionals, and older adults alike, the message is the same: test yourself often, embrace the errors, and use the frustration as fuel. The science is clear. A memory that cost you something to build is the one you will keep.
Staff Writer
Lauren covers medical research, public health policy, and wellness trends.
Comments
Loading commentsโฆ



