Why forgetting might actually be a good thing

January 9th, 2023

Written by: Jake Parker

One morning at a newspaper headquarters in mid-1920s Moscow, a young reporter and his colleagues were attending their daily meeting. While his colleagues were diligently taking notes on their assignments, the reporter simply listened. Upon noticing this, his editor confronted him, asking why he was not writing anything down. The reporter explained that he did not need to, since he could remember everything. The editor then challenged him to recite everything that he had said that morning. To the astonishment of everyone in the room, the young reporter repeated everything that was discussed in perfect detail1–3. In fact, subsequent studies by psychologists revealed that the reporter, whose real name is Solomon Shereshevsky, could memorize long lists of numbers and words in a matter of minutes. And unbelievably, he was reported to have remembered such sequences when tested over a decade later1–3 (!).

For those like me that forget names seconds after hearing them and struggle to remember passwords, it is natural to believe that we would be much better off if we had a memory like Shereshevsky’s. However, such an extraordinary memory came at a high price. According to Alexander Luria, the Soviet psychologist who studied Shereshevsky, he grew so burdened by his vast recollections that, in a desperate attempt to purge them, he wrote them down on slips of paper he then burned1,2. One possible reason for this desperation is that everyday occurrences and objects could involuntarily prompt him to recall specific memories in excruciating detail. Given that he had within his mind an immense number of recollections, this may have made his life an exhausting parade of recall after recall, as others with similar memory conditions have reported4,5.

One might assume that despite negative effects like these, perfect (or near-perfect) memory would otherwise enhance one’s life. This, however, does not appear to be the case. Rather, the inability to forget actually harms some of our most powerful brain functions – the ability to learn effectively from the world and our ability to make good decisions under changing circumstances6,7.

The role of forgetting in learning

To illustrate why forgetting may be beneficial for learning, let us consider an example. Imagine you are stranded in the wilderness and the only source of food you have are various kinds of berries found around the area. However, some of these berries are poisonous, so not all are safe to eat (Figure 1A). A month ago, you read a guidebook that identified over 100 varieties of berries as either poisonous or non-poisonous. While those with perfect memory could remember the safety status of each and every variety in the guidebook, those with an average memory would be unable to do so. Instead, they need to learn the few essential features that distinguish poisonous from non-poisonous berries (Figure 1A). In this simplified, fictitious example, all blue-colored and/or clustered berries are safe to eat.

Figure 1. A) The difference in how people with average memory and perfect memory learn which berries are safe or unsafe to eat. People with perfect memory are able to memorize whether each and every berry is safe, unlike those with average memory who are forced to learn the core features that indicate safety. B) Generalization to unknown berries. When faced with unknown berries, those who were forced to learn the core features indicative of safety are able to infer which are safe. Those with perfect memory are unable to make this determination because they did not learn these core, generalizable features.

Whether someone memorizes the safety of every variety individually or the essential features across varieties, committing this knowledge to memory is one form of learning. In the brain, learning is facilitated by adjusting the strength of the connections between groups of neurons8. These adjustments make it possible for our brains to associate specific berries and/or certain core features with their safety. At the same time learning causes these adjustments to be made, there are mechanisms within the brain that work to weaken or undo them6,7. For simplicity, it is helpful to think of these as mechanisms of forgetting.

The effect of these forgetting mechanisms is that, in the average brain, only the adjustments associated with the most important or most rehearsed knowledge are retained. Knowledge that is seldom encountered or that does not seem relevant might cause some initial weak learning that is quickly undone by forgetting6,7. In our berry example, the average brain would only be able to learn that there are certain core features (blue-colored and/or clustered) shared by many safe varieties6. To a growing number of neuroscientists, the existence of these opposing learning and forgetting mechanisms and their effects suggest a possible source of super-human memory. Namely, one possibility is that the forgetting mechanisms of people like Shereshevsky are much weaker than that of the average brain6. Neurologically, this means that learning would seldom, if ever, be reversed by forgetting for even the most trivial of knowledge. In our berry example, this would manifest as being able to perfectly learn whether each and every individual variety is safe to eat after one or maybe a few reviews.

Again, this may seem like a huge advantage for those with perfect memory. If all of the berries you can find while stranded were in the guidebook, then that indeed may be the case. However, imagine now that many of the berries you find were not in the guidebook. In this scenario, a perfect recollection of all the varieties of berries does not help. Rather, only those that learned the essential features shared by all safe varieties (blue-colored and/or clustered) would be able to determine whether each unknown variety is safe (Figure 1B). Thus, those that were forced to learn these essential features because they could not just memorize everything in the guidebook are much better off.

This exact scenario reflects a fundamental aspect of the world we live in – we are often required to make judgments about things we have not seen before. While these things might closely resemble other things we have experience with, it is often the case that no two things are exactly the same. By forcing us to learn the essential features and general trends that hold across a variety of similar, but different things (like all of the different types of berries that are safe to eat), a healthy balance between learning and forgetting allows us to make good judgments about things we have never seen before. This process of learning the essential features of a group of things and using this knowledge to understand similar, but never before seen things is called generalization9,10. Perhaps unsurprisingly, it was reported that Shereshevsky struggled with this process2,6. Because he could learn individual instances to such an incredible degree, he was never forced to learn the kinds of trends or features that held true across many instances, and thus was unable to generalize his knowledge to new ones.

The role of forgetting in flexible decision-making

Now let’s imagine you have survived this ordeal. For some odd reason, this does not deter you from exploring the outdoors and you prepare to venture to a region in a different part of the world. Upon reading a guidebook to prepare for your excursion, you find that although this new region has the exact same species of berries, the ones that are safe to eat are different due to environmental differences (Figure 2). Regardless of whether one has memorized the safety status of each variety individually or has learned the essential features shared by all safe species, this poses a difficult challenge. Since some species you previously associated with being safe are now unsafe and vice versa, you need to undo these now outdated associations and form the new correct ones. This process of undoing the original association and forming the opposite one is called reversal learning11. Since it involves learning information that directly conflicts with old knowledge, it is a common measure of mental flexibility.

Figure 2. The difference in how those with average memory and those with perfect memory adapt to a change in which berries are safe to eat. In people with average memory, mechanisms of forgetting eliminate the outdated knowledge so that the new appropriate associations can be made. These mechanisms of forgetting are weaker in those with perfect memory and thus fail to eliminate the outdated knowledge, which means the updated associations cannot be learned.

In the brain, reversal learning relies on the neurological mechanisms of forgetting that we explored earlier6,12–14. Blocking these mechanisms in the brains of flies and mice inhibit their ability to learn new, updated information that conflicts with old, outdated information12–14. Crucially, neuroscientists noted that forgetting was necessary for the process of eliminating old, outdated information. If this purging cannot be done, then it becomes difficult, if not impossible to learn any new information that directly conflicts with the old6,13.To see why, we return to the idea that our brains learn information by adjusting the strength of connections between groups of neurons. After these adjustments are made, the information is essentially stored as the pattern of connections that have been formed. Just as conflicting information is logically incompatible with the old information, the pattern of connections that correspond to the conflicting information is physically incompatible with the pattern corresponding to the old information. Thus, the old pattern of connections must be removed by forgetting before the new pattern is put into place by learning.  In our berry scenario, the inability to forget (perfect memory) would cause one to struggle to unlearn the outdated associations between berry variety and safety. This would in turn prevent the learning of which berries were safe to eat in the new region.

This example highlights another important aspect of our lives – the world around us often changes, rendering some of our knowledge outdated or even outright misleading. Indeed, relying on outdated knowledge in the scenario we just discussed could prove fatal. Therefore, forgetting appears to be essential for allowing us to update outdated information so we can flexibly adapt our decisions to a changing environment. Although it is not reported whether Shereshevsky had difficulty adapting to changing circumstances, it seems plausible to believe so given his apparent desperation to forget things1,2.

__________________________________________________

When hearing about the story of Solomon Shereshevsky, it is easy to consider it an example of what perfect memory looks like. However, the evidence and examples we have explored suggest that rather than representing perfection, memory of this kind reflects a disorder of forgetting. Instead, the hallmark of a good memory system seems to be a healthy balance between learning and forgetting6,7. As we have discussed, the inability to forget leads to difficulty in learning generalizable knowledge and adjusting to changing circumstances. On the flip side, too much forgetting (or not enough learning to counteract forgetting) leads to the inability to form any new knowledge to begin with. Altogether, these ideas suggest that the goal of our memory is not to perfectly remember everything, but rather to help us take the best actions possible6.

While the vast majority of us will never be able to remember virtually everything like Shereshevsky, now you know why that may not be the blessing it sounds like. Hopefully that brings you comfort the next time you lock yourself out of your Facebook account because you cannot remember your password.

References

1. Solomon Shereshevsky. In: Wikipedia. ; 2023. Accessed January 3, 2024. https://en.wikipedia.org/w/index.php?title=Solomon_Shereshevsky&oldid=1190039596

2. Johnson R. The Mystery of S., the Man with an Impossible Memory. New Yorker. Published online August 12, 2017. Accessed January 1, 2024. https://www.newyorker.com/books/page-turner/the-mystery-of-s-the-man-with-an-impossible-memory

3. Solomon V. Shereshevsky: The great Russian mnemonist. Cortex. 2013;49(8):2260-2263. doi:10.1016/j.cortex.2013.05.007

4. McRobbie LR. Total recall: the people who never forget. The Guardian. https://www.theguardian.com/science/2017/feb/08/total-recall-the-people-who-never-forget. Published February 8, 2017. Accessed January 5, 2024.

5. Zeliger O. Like it was yesterday. PennNeuroKnow. Published October 24, 2023. Accessed January 5, 2024. https://pennneuroknow.com/2023/10/24/like-it-was-yesterday/

6. Richards BA, Frankland PW. The Persistence and Transience of Memory. Neuron. 2017;94(6):1071-1084. doi:10.1016/j.neuron.2017.04.037

7. Hardt O, Nader K, Nadel L. Decay happens: the role of active forgetting in memory. Trends Cogn Sci. 2013;17(3):111-120. doi:10.1016/j.tics.2013.01.001

8. Citri A, Malenka RC. Synaptic Plasticity: Multiple Forms, Functions, and Mechanisms. Neuropsychopharmacology. 2008;33(1):18-41. doi:10.1038/sj.npp.1301559

9. Generalization (learning). In: Wikipedia. ; 2023. Accessed August 14, 2023. https://en.wikipedia.org/w/index.php?title=Generalization_(learning)&oldid=1146657538

10. Gluck MA. Learning and Memory: From Brain to Behavior. Worth Publishers, Incorporated; 2010.

11. Transfer of training – Stimulus Predifferentiation | Britannica. Accessed January 5, 2024. https://www.britannica.com/topic/transfer-of-training/Stimulus-predifferentiation

12. Dong T, He J, Wang S, Wang L, Cheng Y, Zhong Y. Inability to activate Rac1-dependent forgetting contributes to behavioral inflexibility in mutants of multiple autism-risk genes. Proc Natl Acad Sci. 2016;113(27):7644-7649. doi:10.1073/pnas.1602152113

13. Epp JR, Mera RS, Köhler S, Josselyn SA, Frankland PW. Neurogenesis-mediated forgetting minimizes proactive interference. Nat Commun. 2016;7. doi:10.1038/ncomms10838

14. Shuai Y, Lu B, Hu Y, Wang L, Sun K, Zhong Y. Forgetting Is Regulated through Rac Activity in Drosophila. Cell. 2010;140(4):579-589. doi:10.1016/j.cell.2009.12.044

Cover photo by Tumisu on Pixabay.

Figures 1 and 2 made by Jacob Parker using PowerPoint

One thought on “Why forgetting might actually be a good thing

Add yours

Leave a comment

Website Powered by WordPress.com.

Up ↑