Why do we remember information that we attach significance to better than information we repeat?


Levels of Processing Effect

, explained.

What is the levels of processing effect?

The levels of processing effect is based on the idea that the way in which information is encoded affects how well it is remembered. The levels of processing model counters the idea that mere repetition helps us retain information long-term. Instead, it suggests that information that is encoded on a deeper level, through meaningful association, is easier to remember.

Where this bias occurs

When we are studying for a test, we might find it really hard to remember certain terms that are unfamiliar. If we try to remember the word by repeating it, we are only processing it at a shallow level. However, if we associate those terms with other information or imagery, it is encoded at a deep level and is easier to remember.

That is why a popular studying technique is to create songs or sequences that binds individual terms to other things. If you are studying for a Biology exam and are trying to remember the order of taxa (Domain, Kingdom, Phylum, Class, Order, Family, Genus, Species), simply remembering the words in that order might prove to be difficult. Alternatively, if you remember a mnemonic, a pattern of letters that helps you remember something, like “Dear King Phillip Came Over For Good Soup”, you are more likely to recall the order of taxa during the exam. 

Debias Your Organization

Most of us work & live in environments that aren’t optimized for solid decision-making. We work with organizations of all kinds to identify sources of cognitive bias & develop tailored solutions.

Learn about our work

Individual effects

If we are not aware of the levels of processing effect, we may use methods to try to retain information that are less effective than ones that focus on the way that information is encoded.

It is commonly believed that if we repeat something enough, we will remember it. This belief can cause us to use studying techniques like “cramming” where we quickly review information right before an exam. While this may help us remember information in the short-term, it is unlikely to commit information to long-term knowledge. Cramming includes reviewing notes, thereby encoding the information structurally, by the way that it appears. Structural encoding is considered a shallow level of encoding.

The levels of processing effect suggests that in the long-run, repetition is insufficient for long-term memory. Instead, we must encode information semantically, by relating it to other ideas and knowledge. This encodes the information on a deep level and is a more effective studying technique. 

The levels of processing effect is not only relevant to studying. We are asked to recall information daily, sometimes in a form as simple as remembering the name of someone we met. Simply repeating their name could lead to embarrassing moments of forgetfulness, since repetition alone does not commit information to lasting memory. Instead, if we are to associate someone’s name with something more meaningful, like a hobby that they share with us, encoding occurs at a deeper level and we are more likely to easily remember their name. 

Systemic effects

Although individually, we can use tactics that activate the levels of processing effect, often, how information is presented is not under our control. Formative years of our lives are spent learning in school and college, and without knowledge of the levels of processing effect, institutions may not be as effective as they could be in ensuring their students absorb the information into knowledge.

Although shallow encoding might retain information long enough for students to pass an exam, they are unlikely to remember that information years down the line, when they need it for a job. One of the skills educational institutions are supposed to teach us is how to effectively learn so that we are prepared to continue learning throughout our lives. Schools should therefore ensure that they are teaching their students the best possible ways of retaining information, which according to the levels of processing effect, is to encode information on a deep level. 

Why it happens

There are multiple ways in which information can be encoded in our brains which will then affect our ability to recall it later. The levels of processing effect predicts that when information is encoded on a deep level, it is easier to remember than when it is encoded on a shallow level. That is because if it is processed on a deep level, it is more meaningful to us. 

A shallow level of encoding occurs when we process information, like words, structurally, by the way they appear, or phonemically, by the way they sound.1 An example of structural encoding would be to remember the color of a word, and an example of phonemic encoding would be to think about what the word rhymes with.

A deep level of encoding occurs when we process information semantically, by associating it with emotion, idea or previous knowledge.1 An example of semantic encoding would be to figure out where a word fits in a sentence.

Why it is important

If memory recollection is affected by the way information is encoded, it is important that we focus on ways to improve how we encode information. Knowing about the levels of processing effect can help us improve the way that we teach, learn and study. Awareness of the effect can therefore lead to stronger learning outcomes, which is useful both in school and work settings.

The levels of processing effect demonstrates that if we associate new information with something meaningful, we are more likely to remember it. This knowledge can inform teachers or bosses that they should frame their teaching to be more specific and relevant to their students and employees.

Imagine, for example, that Raj is in an economics class and is finding it difficult to remember what “opportunity cost” is. Opportunity costs are what someone misses out on when they make one choice over an alternative.2 However, that definition has little meaning to Raj. His teacher, Miss Maitland, sees that he is struggling, and knows that Raj is really passionate about cars. To help Raj encode the term “opportunity cost” semantically, Miss Maitland asks him what his favorite two cars are. Raj says that he likes the Audi A5 and the BMW X3. Miss Maitland explains to Raj that if he were to purchase the Audi A5, the opportunity cost would be missing out on the BMW X3. 

After this explanation, Raj has encoded the term semantically by associating it to something that is meaningful to him. Due to the levels of processing effect, Raj would be more likely to remember the term “opportunity cost” at a later date.

How to activate it

Since the levels of processing effect is a cognitive bias that can lead to desirable outcomes, we should try to activate it rather than avoid it. For information to be remembered long-term, it should be encoded on a deep level, which can be achieved by encoding it semantically.

As the example of remembering the order of taxa shows, one way to encode information semantically is by linking it with other pieces of information. The mnemonic “Dear King Phillip Came Over For Good Soup” links each of the taxa to one another, and the order of the taxa to a fun rhyme. Another way to encode information semantically is by relating it to something that is meaningful to us, as demonstrated by the example where Raj’s teacher linked the economic term “opportunity cost” to his passion for cars. 

Other ways of making information meaningful to us can include trying to explain it in our own words, researching topics by ourselves rather than just listening to teachers, or finding real-life examples of a theory instead of reading about it in a textbook.

Information is also better retained when we attach an emotion to it. Returning to the example of remembering someone’s name, we can try to attach an emotion to someone’s name in order to remember it. For example, if we meet a boy named Ryan while we are swimming, we can try to attach their name to the happy feeling we felt while swimming. We would be more likely to remember Ryan’s name later on than if we had simply repeated his name.

Where it all started

The levels of processing effect was first proposed as a model by Fergus Craik and Robert Lockhart in 1972.3 The two pioneers in behavioral science proposed the model as a counter to the dominant theory on memory, the multi-store model of memory, and it quickly became very influential for cognitive psychology.

The multi-store model of memory suggested that memory functioned in a linear fashion, moving from sensory memory to short-term memory to long-term memory. This model proposed that to move information from short-term memory to long-term memory, it had to be repeated.

Craik and Lockhart believed that the multi-store model was insufficient for explaining how memory works. In their alternative model, they suggested that information did not enter memory in a linear fashion, but instead, that the way that information was processed in the brain corresponded to a different level of depth.3 Based on their model, merely repeating information would not commit it to long-term memory. Instead, information has to be encoded in a meaningful way to be committed to our long-term memory. 

Shallow levels of encoding occur when information is encoded structurally, by the way it appears, or phonemically, by the way it sounds. Repetition of this information just helps us retain it for a short period of time, instead of moving it to our long-term memory. 

Deep levels of encoding occur when information is encoded semantically, by relating it to other words, ideas, or previous knowledge. Craik and Lockhart referred to this as elaborative rehearsal, differing from the multi-store model of memory that believed any kind of rehearsal would help retain information long-term.

Example 1 - Mental Health

People with the mental health disorder schizophrenia have impaired memory function when it comes to their episodic memory. Episodic memory is a kind of long-term memory that can recollect everyday events with reference to place, time and emotions. 

Along with a team of researchers, John D. Ragland, professor in the Department of Psychiatry and Behavioral Sciences, wanted to examine whether patients with schizophrenia could improve their memory with organizational strategies that were based on the levels of processing effect.4

For shallow encoding, patients were asked if words appeared in upper or lower case, a structural form of encoding. For deep encoding, patients were asked if words were concrete or abstract, a semantic form of encoding. Next, 40 of the words that a patient had seen were mixed with 20 new words, and patients were asked whether the word was a new or old stimuli. The time it took for patients to make each decision was also recorded.

The researchers found that patients were both faster and more accurate in their recollection of words that had been encoded on a deep level. The researchers concluded that when patients with schizophrenia do not have to consider how to organize data themselves, and are helped through the levels of processing effect, their episodic memory improved.

This study provides support for the levels of processing effect over the multi-store model of memory. Patients saw each word only once yet were able to remember some words better than others. Those words were ones that had been encoded on a deep level, suggesting memory recollection is affected by elaboration rehearsal rather than rehearsal alone.

Example 2 - False memories

While there has been ample research conducted into how encoding on a deep or shallow level affects how well information is retained, psychologist Maria Soledad Beato and her team wanted to investigate whether the levels of processing effect would still be true for false memories.5

False memories are memories that we are led to believe happened even though they didn’t. This is sometimes achieved in studies through suggestibility, through giving participants plausible information. Participants may be presented with a number of words like “moon”, “darkness” and “stars”, which sometimes causes them to falsely remember an unpresented word like “night” in a subsequent recollection task. The word “night” can also be referred to as a critical lure. 

In her experiment, Beato showed participants groups of words that had corresponding critical lures. Participants were asked to either create a mental image of the word, which is a form of semantic encoding, or asked whether the word contained an ‘o’, a form of structural encoding. 

Beato found that participants remember words that had been encoded on a deep level 85.4% of the time compared to 71.2% of the time when words had been encoded on a shallow level. However, the levels of processing effect did not make a significant difference on the number of critical lures that participants falsely remembered.

This study provides evidence for the levels of processing effect, and also shows that it only affects our memory recall for information we’ve actually processed, rather than creating false traces, even though semantic encoding asks us to relate new information to other information.


What it is

The levels of processing effect is the effect that different forms of encoding have on memory. When information is encoded semantically, through association with other information, it is easier to recall at a later date than when information is encoded structurally or phonetically, by the way it appears or sounds. 

Why it happens

The levels of processing effect occurs because our brains find it easier to recall information that is encoded on a deep level. Semantic encoding is considered a deep level because it associates new information with existing emotions, ideas and knowledge, thereby creating a larger memory trace. 

Example 1 – Mental health

Patients with schizophrenia find it more difficult to correctly recall everyday events. In presenting information to patients with schizophrenia in a way that asks them to encode it deeply, like by asking whether a word is concrete or abstract, patients do not have to waste effort trying to come up with their own organizational strategies. This allows patients to focus on encoding the information which improves their episodic memory.

Example 2 – False memories

Even though encoding on a deep level asks us to associate new information with pre-existing ideas and knowledge, researchers found that deep encoding does not create more false memories than shallow encoding. It does, however, help us to better remember true memories. 

How to activate it

In order to benefit from the levels of processing effect, we should encode information on a deep level. Semantic encoding is a deep level of encoding and can include relating new information to other pieces of knowledge, to mnemonics, to our passions or to our emotions.


  1. Mcleod, S. (2007). Levels of processing. Simply Psychology. Retrieved July 27, 2020, from https://www.simplypsychology.org/levelsofprocessing.html
  2. Investopedia. (2003, November 24). Understanding opportunity cost. Retrieved July 27, 2020, from https://www.investopedia.com/terms/o/opportunitycost.asp
  3. Craik, F. I., & Lockhart, R. S. (1972). Levels of processing: A framework for memory research. Journal of Verbal Learning and Verbal Behavior11(6), 671-684. https://doi.org/10.1016/s0022-5371(72)80001-x
  4. Ragland, J. D., Gur, R. C., Valdez, J. N., Loughead, J., Elliott, M., Kohler, C., Kanes, S., Siegel, S. J., Moelter, S. T., & Gur, R. E. (2005). Levels-of-Processing effect on Frontotemporal function in schizophrenia during word encoding and recognition. American Journal of Psychiatry162(10), 1840-1848. https://doi.org/10.1176/appi.ajp.162.10.1840
  5. Beato, M. S., Boldini, A., & Cadavid, S. (2012). False memory and level of processing effect. NeuroReport23(13), 804-808. https://doi.org/10.1097/wnr.0b013e32835734de

About the Authors

Dan Pilat's portrait

Dan Pilat

Dan is a Co-Founder and Managing Director at The Decision Lab. He is a bestselling author of Intention - a book he wrote with Wiley on the mindful application of behavioral science in organizations. Dan has a background in organizational decision making, with a BComm in Decision & Information Systems from McGill University. He has worked on enterprise-level behavioral architecture at TD Securities and BMO Capital Markets, where he advised management on the implementation of systems processing billions of dollars per week. Driven by an appetite for the latest in technology, Dan created a course on business intelligence and lectured at McGill University, and has applied behavioral science to topics such as augmented and virtual reality.

Sekoul Krastev's portrait

Dr. Sekoul Krastev

Sekoul is a Co-Founder and Managing Director at The Decision Lab. He is a bestselling author of Intention - a book he wrote with Wiley on the mindful application of behavioral science in organizations. A decision scientist with a PhD in Decision Neuroscience from McGill University, Sekoul's work has been featured in peer-reviewed journals and has been presented at conferences around the world. Sekoul previously advised management on innovation and engagement strategy at The Boston Consulting Group as well as on online media strategy at Google. He has a deep interest in the applications of behavioral science to new technology and has published on these topics in places such as the Huffington Post and Strategy & Business.

Notes illustration

Eager to learn about how behavioral science can help your organization?