Most of us regard forgetting as an annoyance because our memory fails us. This also has been the most prominent view in memory research in the last century. But during the past decade, we have learned that this is not true. In fact, most of everyday forgetting is not a glitch in the normal functioning of the brain, but a feature of memory. Forgetting is so fundamental to memory that without it, we would have no memory at all.

This sounds a bit counterintuitive, which is probably why forgetting has become an attractive topic in current memory research (many scientists like surprise). Imagine you would have perfect memory, and you would never forget anything. What would life be like? The Argentinian poet Jorge Luis Borges described such a life in his biography of the fictional character Funes, an Uruguayan who could not forget. Funes was unable to develop conceptual ideas because he could not abstract the general from the particular. Not only were all dogs for Funes as different from each other as dogs are from cats, but each individual dog would not even be self-similar, as a particular dog would change from moment to moment, from viewpoint to viewpoint, and these differences Funes would note and memorize and since he could not forget, Funes could not merge these instances to connect them to one particular dog, let alone to the category of dogs. His perfect memory made life nearly impossible for Funes.

Although Funes and his unfailing memory are a thought experiment, there are humans who retain certain forms of memories to a much greater extend than the average population. People with highly superior autobiographical memory (or hyperthymesia), for example, can recall an astonishing number of events of their lives in minute detail. Jill Price, who was the first person identified with this condition, can recall the details of any day of her life since she was fourteen, and, similar to Funes, the diminished forgetting is one reason why she resides in an assisted living arrangement.

A perfect memory may therefore be more burden than asset, and forgetting may be an essential, necessary component of any natural memory system as otherwise it would fail. We have argued that because the way memories must be formed, a functional memory system requires constitutional forgetting. When experiencing the present, it is largely unclear what aspects of the experience should be retained in long term in memory. And often, it may require some time to decide what to keep and what to ignore. Experiences, however, are fleeting because the particular events in the world causing these experiences themselves are transient, and in most cases, they do not repeat, and if they do, they may not unfold exactly as before. If no memory of an experience is formed while it occurs, there will be no record of it anywhere (our brains evolved before smartphones and surveillance technology was invented). One way to address this problem is to engage in “promiscuous” encoding – basically, forming a memory that captures as much as possible of whatever one experiences right now, and later decide whether that memory is useful. We proposed that therefore natural forgetting processes are a fundamental aspect of memory, as without them, we would all be like Funes.

That this likely is the case can be seen when looking at our own lives. At the end of a day, can you give a detailed report of what has happened since you woke up? For humans with normal memory, this is an easy task, and seems to form the basis for a variety of evening rituals, such as writing a diary, or family conversations during dinner. Perhaps we have been telling each other the remains of our days since we sat in groups around a fire at night. As we all know, most of our recounted experiences are mundane, not worth retaining, yet, without even trying or any awareness, we automatically form detailed and long-lasting memories of these experiences. The vast majority of these event memories will not last more than a couple of days, but some will be retained for longer, maybe a lifetime, and how to explain this is one of the biggest and unresolved challenges in memory research.