Science AMA series: Hi, I’m Amy Smith, graduate student at Tufts University. I study stress and memory, and recently published a paper in Science showing that taking practice tests leads to better memory recall, especially under stress, than traditional “studying.” Since it’s almost exam time, AMA!


Hi Reddit,

I’m Amy Smith, and I work in Tufts’ Cognitive Aging and Memory Lab with my advisor, Ayanna Thomas. We recently published a paper in Science showing that retrieval practice, a strategy where a person takes practice tests to learn material, can protect memory against the negative effects of stress. The traditional studying strategy of re-reading material over and over doesn’t have the same benefits.

This runs counter to more of a decade of research saying that stress always impairs memory. As you prepare for your exams or any other high-stakes situation where you need to have good memory, our findings suggest that it’s more important to focus on how you study, not how much you study.

You can read the paper here:

And a write up (and short video) of our study here:

Ayanna Thomas, co-author and PI of the lab, is here to answer questions as well!

A big thanks to Reddit and all of the question-askers for this AMA! I'm signing off for now, but will answer more questions in the next few days.

Can you please discuss techniques for long term (not just passing a test one time) maintenance of information?


Hi, GlobbyDoodle! According to a vast amount of research in Cognitive Psychology, taking practice tests (aka: retrieval practice) is actually one of the best techniques for keeping information in your long-lasting memory. There is more to it than just "taking practice tests" though. The best kind of practice test is one in which you force yourself to freely recall information. For example, forcing yourself to write down as much information as you can remember on a blank sheet of paper. Multiple-choice tests are not as effective, because they give you a crutch to lean on (the answer is right there in front of you!). Also, taking these free-recall practice tests is most effective if you let a substantial amount of time pass between each practice test. For example, if you're studying for an exam, it would be best to start studying several weeks in advance and test your memory every few days leading up to the exam.

Another piece of advice regarding effective study techniques: always try to relate new information to things you already know. If you just met a woman named Carol and you want to remember her name, think of other women you know named Carol. Think about Christmas Carols, or maybe you have a friend who lives on Carol Street. The more you make meaning out of new information, the more likely you are to remember it in the future.

Thanks for your question--hope this helps!

Can you please discuss techniques for long term (not just passing a test one time) maintenance of information?


Ayanna -

Deeply engaging in material will result in long-term retention. In fact, a lot of our research is focused on long-term retention. Taking practice tests during learning is one of the best techniques for improving long-term retention.

Hi Amy! Is there any truth behind the myth where if you chew gum while studying and chew the same kind of gum while taking the exam, you'll remember it better? Beyond a placebo effect of course


Hi there! The findings on this are mixed. One study did find that the best memory performance occurred when participants chewed gum while studying AND while testing. However, a few studies have tried to replicate the effect and failed. There's a lot to be said about placebo effects though--if chewing gum during studying and testing feels like it helps you concentrate or remember information, I'd say go for it. It's probably not going to hurt, unless you have TMJ :)

Hi Amy! Is there any truth behind the myth where if you chew gum while studying and chew the same kind of gum while taking the exam, you'll remember it better? Beyond a placebo effect of course


Ayanna - You are getting at a phenomenon known as encoding specificity. Basically, the more similar the context in which you learned information is to the context in which you retrieve information, the more likely you will be to successfully retrieve information. That said, something like chewing gum or using the same pen probably has little noticeable effect.

Hello Amy, From your work, would you recommend less time studying through exams and more time taking care of ones personal well being? Or is it beneficial to 'crunch' and work super hard and deal with the stress when Exam season is over?


Ayanna -

So our work does not directly speak to well being, but I think that we can agree that personal well being is more important that exams.

Research does suggest that "crunching" is actually not all that beneficial for long-term retention. Of course students work hard during exam season, but I think that some of that end-of-the-semester burden could be reduced if students did not crunch but rather spread practice of the material across the semester. You may still be stressed during exam season, but at least you will have stable and robust memories to rely upon.

Hello Amy, From your work, would you recommend less time studying through exams and more time taking care of ones personal well being? Or is it beneficial to 'crunch' and work super hard and deal with the stress when Exam season is over?


Here is a link to a recent CBC article that might help with this question:

I personally try to find a balance between studying hard and taking time for myself (exercising, sleeping well, etc.). I find this is easiest to accomplish if I start work projects several weeks in advance and don't leave studying/writing to the last minute.

After a decade of chronic pain, several surgeries requiring general anesthesia, and years of medications of many kinds I know that my memory is worse than it was before this horse shit started. I imagine the stress of bearing such a constant state of assault on my ability not to scream or whimper in public plays a role.

Is there anything I can do to help improve my memory or return it to its former glory?


Thanks for your question, MainAccount! And sorry to hear about your misfortunes. There are always techniques you can use to strengthen your memory! Even people in their 80's and 90's can improve their memory performance (I also study aging). There is a lot of good advice out there, but I'll name a few of my favorites. First, learning new things is highly beneficial for strengthening memory and other aspects of cognitive function. Always challenge yourself to master new skills (for example, cooking authentic Indian food) and acquire new knowledge (for example, learn about a historical era that you are unfamiliar with). Second, take good care of your body. Eat healthy, get as much exercise as you can, and try to sleep 8 hours a night. Healthy food and exercise promote brain function, and sleep is a crucial time for brain repair. Our memory works best when those three needs are met.

Can you speak to how effective state-dependent learning is?

Do does caffeine have a positive effect on your efficiency to commit short term to long term memory or is it merely a stimulant? How about control substances such as methylphenidate?


There is evidence that memory is facilitated by a match between the state and/or context you are in during learning and at retrieval. That said, I think these sorts of effects are less important in the context of education, because there are so many factors in addition to state congruency that will impact memory.

The literature regarding caffeine on memory encoding (which would be the transfer between short and long term memory) is mixed. Caffeine does affect some cognitive processes, such as attention and vigilance, but more research is necessary to understand whether caffeine affects memory. We're actually doing this research presently.

Sorry, cannot help with the methylphenidate question.

Have you studied how this applies to subjects like math and engineering, where it's not about memory necessarily, but problem solving?


Thanks for your question, soapfrog. I think the top four comments below do an excellent job of answering this. Problem solving relies heavily on memory, and thus taking practice tests prior to a test in math or engineering would still be beneficial. I was a theoretical math major as an undergrad, and I used retrieval practice to memorize theorems and proofs. By having those in memory, I was able to more flexibly prove the novel theorems I encountered on the exam. Essentially, practice testing helps you memorize the steps to solving a problem, which helps you solve novel problems in the future.

I have not personally studied whether practice testing is helpful for STEM learning, but a few other researchers have and indeed it is helpful!

Have you studied how this applies to subjects like math and engineering, where it's not about memory necessarily, but problem solving?


Ayanna -

There are studies that have investigated the value of retrieval practice for map learning and spatial skills. However, there is little research examining the effects of stress on learning and retrieval of this kind of information. Researchers have investigated how math anxiety impairs performance. I think a really interesting question is whether that impairment could be reduced if we influence the way the information is learned.

Hi Amy, have you used your findings in your own studying in grad school?


Oh yes--all the time--and in undergrad, too. I'm currently preparing to give a presentation on my research next week. I prepare by reading over my speech notes, pushing them aside, and then forcing myself to give the speech without looking at my notes. Then I look at my notes to see where I went off-track, and then practice the speech again. The first practice is always terrible. By the third or fourth time, the speech is so engrained in my memory I don't forget it for months.

For tests in undergrad, I was a huge fan of flashcards. Flash cards are a wonderful tool for retrieval practice, and you can even have them made for you online now (as previous comments on this thread have suggested).

Thanks for your question :)

Hi, thanks for doing this. I used to tutor kids for the ACT. I found that, in my experience, simply having them take as many practice tests as possible was not anywhere close to as effective as breaking down problems slowly and making sure understood the why, the how and general test taking strategy. Can you talk about the differences between learning "words and images" in this manner and solving more complicated problems?


Hello, ASonnetOfIceAndFire. I completely agree with you--taking practice tests is useless if students do not understand the material to begin with. I believe the utility of practice tests comes in at the point when material has been learned and understood, and the student now has to make sure that the material is stored in long-term memory for a future test. For example, memorizing the Pythagorean Theorem without understanding it will likely not help you figure out when you need to use it on a test. But understanding the Pythagorean Theorem without memorizing it will also not help you on the test. You need both.

We used words and images as stimuli in our study because we wanted to conform to the methods used by previous stress-and-memory researchers, with the goal of comparing our findings directly to theirs. However, in the general retrieval practice literature, researchers have found that taking practice tests is beneficial for strengthening memory for a wide variety of topics and materials. As examples, practice tests support memory for maps and complex written texts. Whether retrieval practice continues to have these benefits with more complex materials when stress is present is TBD :)

Do you think your findings apply regardless of learning style and the strength of a person's ability to perform rote memorization? Also, I'd be interested to know if there are performance differences between people who find tests really stressful vs. those who are more relaxed about them. Thanks!


Hi LudovicoSpecs! Your question about learning style is a good one--and a tough one to answer. Some people simply memorize information better than others. However, all healthy, non-brain-damaged humans are capable of memorizing vast amounts of information. Some of us just need to work harder at that than others (myself included). Taking multiple practice tests is a very effective way to do this, but again, some people may just need more practice.

Regarding your second question, the beauty of this experiment is that we found that learning by taking practice tests helped people remember information when they were under stress. In fact, their memory was so good it was as if stress hadn't even been present. I think these findings are especially relevant for people who experience test-related stress and anxiety. If you study using the conventional method of re-reading your notes, you will not remember information during that stressful test as well as if you study using a combination of reading your notes and taking practice tests.

Does it matter whether or not the practice test is something the student chooses to do rather than being assigned to do?


Hello, mepcotterell! I'd say no. It's most crucial that the student puts in effort on the practice test. If the student voluntarily chooses to take the practice test, they are probably already highly motivated and ready to put in effort. If an instructor assigns a practice test, chances are the test is being graded and the student is thus motivated to put in effort. Effort is so crucial because researchers have found that the more cognitive effort a practice test requires, the more beneficial it is for long-term memory.

I completely agree. Ive noticed that I tend to do better in classes that do many quizes and tests. Though I am curious if you found any data showing certain types of test questions are better for learning than others. Like are fill in the blank questions are better than multiple choice?


Ayanna - The more you have to work to retrieve the information, the better long-term retention. In that sense, I would suggest that recalling information is going to be the best way to test, because it provides few (if any) specific cues for the student to recognize.

I have a friend in my class that has to take tests alone because he fells stressed and pressured and forgets everything he studied when there's people around him. Is there an explanation to this situation or something like that?


Ayanna -

Hi! Your friend's situation is one of the motivating reasons for Amy and I starting this research. Students are increasingly feeling stress and anxiety and that is negatively impacting their ability to remember information. I am not certain why this phenomenon is increasing in frequency. There are probably a number of contributing factors. However, Amy and I believe that our line of research is a possible avenue to combat these negative effects of stress.

Have you studied this phenomenon across the sexes and ages? Do you postulate it's a hormone produced under stress that contributes to better recall? Long term, do you think producing a synthetic hormone for students to take would improve exam results?


HiTechCity--excellent questions. The human stress hormone - cortisol - that is released during stress is actually detrimental to memory. It's thought to be the reason why over a dozen studies have found that stress impairs memory recall. There are a few prescription drugs that can block the effects of cortisol, but currently they are only prescribed to Heroin addicts (see buprenorphine) or depressed individuals (see CRF antagonists). In the future, I'd love to see these drugs available to the public in small, safe doses.

There are a TON of sex and age differences in the stress-and-memory literature. I'll name the big ones. Men tend to have a stronger physiological response to stress than women. Women who take oral contraceptives tend to have an even smaller stress response than normal. Because of this, many studies have only found negative effects of stress on memory in men. In our experiment, we also found sex differences, with women remembering more information than men.

As for aging, young adults tend to have a stronger physiological response to stress than older adults (65+). Thus, older adults do not always show memory impairment under stress. We are currently conducting a study in our lab with older adults to follow-up on some of this work. More to come later :)

Wouldn't a practice test only help if the real test asks the exact same questions?


Ayanna -

Absolutely not. Taking practice tests does not simply preparing you to answer those test questions. Taking tests is a form of learning material that leads to robust and interconnected memories. Testing prior to new learning improves new learning. One kind of test also transfers to new but related material.

Hi! I haven't had a chance to read the paper yet, but was there any specific test format that had the greatest positive effect? For example, many multiple choice versus few long form problems.


Hi ARC157! We didn't examine different test formats in our paper, but many other researchers have. The consensus is that tests of free recall (remembering as much information as possible with no prompt--the "blank sheet of paper" approach) yield the best long-term memory. Multiple choice tests are not as effective, because they give you a crutch to lean on (the answer is right there, all you need to do is recognize the right one). Essentially, the more difficult the memory test feels, the better it is for long-term memory.

Hello, I am 24 and have terrible memory. I've always had it and some days is worse than others. I would think that maybe I have a learning disability but I couldn't be since ever since I can remember I have all As in school-high school and graduated with honors with a BS in biochemistry. My learning memory is pretty good, but jee, I forget simple things like my keys, phone wallet all the time to the point that it messes with my time management skills and I have to give me extra hours to get anywhere because I know I'll probably forget something and will have to go back home. I worry that I will forget important things or steps in my job where if I mess up one step I could be ruining batches that cost a lot of money and it messes with my self esteem because i don't trust myself to remember if i did something for sure. Is there any way to improve this type of memory?


Ayanna - I wish I had a good answer for you. It is hard to determine whether you have an isolated memory problem or have attention or some other executive functioning problem. If you really feel like this is negatively affecting your ability, I would suggest neuropsychological assessment. Assessment can hone in one exactly what the issues are. From there, methods for combating these negative effects can be implemented.

How long do you recommend for a person to study at one time before you take a break?


Ayanna -

This is a tricky question and I do not think there is a one-size-fits-all response. I think a good rule of thumb is to stop studying when you start to mind wander to task-irrelevant thoughts. Take a break. Maybe exercise a little, and the come back to the work.

Hello Amy, how do you think this study could impact the education community from primary school up to graduate studies?


Many students experience test-related anxiety and stress, and nearly everyone experiences a stress response when they have to give an oral presentation in class. Sometimes, students in these stressful situations have difficulty remembering crucial information. I believe that this line of research has the potential to inform students about study techniques that they can use to strengthen their memory against stress.

Do you hypothesize that practice tests induce distinct neural networks to be formed compared to other methods of study that can be assessed and quantified/localized using functional MRI?


Yes, that is exactly how I think retrieval practice works, actually. I believe that the act of retrieving information creates distinct neural routes by which that memory can be accessed in the future. Each retrieval attempt creates new routes, resulting in many different pathways by which the same memory can be retrieved. Neuroimaging evidence supports this. One study found that, relative to the conventional method of studying (re-reading), retrieval practice increased hippocampal connectivity with other brain regions.

Thanks for your question :)

Hey Amy! I'm an undergrad at Tufts. Do you have any open lab positions? 😊


Hi jendbwis! Are you looking for a Research Assistant position or a graduate student position? If you're serious, get in touch with Professor (Ayanna) Thomas :)

Amy, what was different about your paper? We've known about the benefits of the testing effect and retrieval practice for years now.


Hi alnumero! Excellent question. You're right--the benefits of retrieval practice have been demonstrated hundreds of times now. In this experiment, we specifically used retrieval practice to strengthen memory against the deleterious effects of stress. When people use the conventional form of studying (re-reading notes) and later get stressed out, they experience memory impairment. However, when we had participants use retrieval practice to study, they no longer showed this stress-related memory impairment. Essentially, our study is different because we are the first researchers to show that stress doesn't uniformly impair memory. Rather, it only impairs memory when participants use ineffective study strategies. Retrieval practice just happens to be one of the effective ones! Does that make sense?

Can stress ever help the human brain work better? I play chess and someone told me that times games help you think better than non-timed, because your brain makes better decisions when you're under a time limit. Is there any truth to this?


Hi Tallm! Yes, actually. There is a wealth of evidence showing that stress can enhance cognition. When you are stressed, you experience a surge of epinephrine and norepinephrine. These hormones tend to have excitatory effects on the brain, which can result in quicker thinking, better focus, increased alertness/awakeness, and even enhanced memory in some cases.

However, intense stress responses can also trigger the release of the hormone cortisol. About thirty minutes after the onset of stress, cortisol binds to memory-related brain regions and can impair neural processing. Thus, stress may enhance memory immediately after the onset of stress when only norepinephrine and epinephrine are present, but can impair memory later on when cortisol is present.

Is this true for long term knowledge or simply short-term test taking?


Long-term knowledge for sure! Retrieval practice is an excellent strategy for strengthening long-term memory. Relative to the conventional method of studying (re-reading notes), retrieval practice has been shown to result in much better long-term memory even six months after initially learning and "retrieval practicing" information.

Additional Assets


This article and its reviews are distributed under the terms of the Creative Commons Attribution 4.0 International License, which permits unrestricted use, distribution, and redistribution in any medium, provided that the original author and source are credited.