Skip to main content

Improving Memory Loss in Older Adults with Joel Voss, PhD

As we age, almost all of us have some memory loss. This age-related affliction is normal, but a new Northwestern Medicine study suggests it can be improved with non-invasive brain stimulation that sends electromagnetic pulses into a specific area of the brain. Joel Voss, PhD, an associate professor at Northwestern, led this study, published in the journal Neurology.


Joel Voss

"At baseline ... (the older adults) were worse off than the young adults, significantly worse off compared to the young adults. After receiving their five days of stimulation and coming back a day later, we can no longer statistically tell them apart from the young adults, so it improved their memory to a level where they no longer were age-impaired."

Joel Voss, PhD

  • Associate Professor of Medical Social Sciences
  • Associate Professor of Neurology
  • Associate Professor of Psychiatry and Behavioral Sciences

Episode Summary

Joel Voss, PhD, has been using transcranial magnetic stimulation (TMS) in his lab for the past several years in experiments that have improved memory loss. 

Joel Voss: "The brain is complex. Memory function is complex. We don't fully understand how the brain supports memory. We have some clues. The real question that we're addressing here is not just whether or not we can make people's memory better ... the real question is, can you use something like noninvasive brain stimulation to actually change the function of brain networks involved in memory and does that affect people's memory performance?"

Voss' lab studied young people and found that TMS can improve their memory by targeting specific brain networks. These studies focused on improving what is called episodic memory, a particular kind of memory that the hippocampus area of the brain cares about the most. It is also the kind of memory that declines the most and leads us to forget where we put our car keys or the name of a new neighbor. For his latest study, published in the journal Neurology, Voss chose to focus on adults over the age of 64, to see if the TMS procedure would work the same in older adults as in young adults. 

Joel Voss: "Not surprisingly, when older adults first walk through the door, they don't perform the memory task as well as young adults do, because pretty much all people experience some level of age-related memory weakening, where they have not as good as the ability to remember new things as they did themselves years before and relative to other younger adults, too."

The participants underwent MRIs and a battery of memory tests before receiving TMS. They received 20 minutes of TMS for five days in a row. 

Joel Voss: "After receiving their five days of stimulation and coming back a day later, we can no longer statistically tell them apart from the young adults, so it improved their memory to a level where they no longer were age-impaired, so to speak."

The effect lasted 24 hours and caused the changes in the brain and improved memory performance Voss had achieved in other age groups. He plans to design another study that would extend the number of days participants are exposed to TMS to see if the improved memory effect can last longer. 

Subscribe to Feinberg School of Medicine podcasts here:

Google Play Music

Continuing Medical Education Credit

Physicians who listen to this podcast may claim continuing medical education credit after listening to an episode of this program.

Target Audience

Academic/Research, Multiple specialties

Learning Objectives

At the conclusion of this activity, participants will be able to:

  1. Identify the research interests and initiatives of Feinberg faculty.
  2. Discuss new updates in clinical and translational research.
Accreditation Statement

The Northwestern University Feinberg School of Medicine is accredited by the Accreditation Council for Continuing Medical Education (ACCME) to provide continuing medical education for physicians.

Credit Designation Statement

The Northwestern University Feinberg School of Medicine designates this Enduring Material for a maximum of 0.5 AMA PRA Category 1 Credit(s)™. Physicians should claim only the credit commensurate with the extent of their participation in the activity.

Disclosure Statement

Joel Voss, PhD, has nothing to disclose. Course director, Robert Rosa, MD, has nothing to disclose. Planning committee member, Erin Spain, has nothing to disclose. Feinberg School of Medicine's CME Leadership and Staff have nothing to disclose: Clara J. Schroedl, MD, Medical Director of CME, Sheryl Corey, Manager of CME, Jennifer Banys, Senior Program Administrator, Allison McCollum, Senior Program Coordinator, and Rhea Alexis Banks, Administrative Assistant 2.

Claim your credit

Erin Spain: This is Breakthrough a podcast from Northwestern University Feinberg School of Medicine. I'm Erin Spain, executive editor of the Breakthroughs newsletter. As we age, almost all of us have some memory loss. This age related affliction is normal, but a new Northwestern Medicine study suggests it can be improved with noninvasive brain stimulation. This sends electromagnetic pulses into specific areas of the brain. Joel Voss and associate professor at Northwestern led this study, published in the Journal of Neurology and he's here to share the results. His lab focuses on understanding the mechanisms of learning and memory and is working on several new treatments for memory impairment. Thank you so much for being here today.

Joel Voss: Thanks for having me.

Erin Spain: In this study, you used a method called transcranial magnetic stimulation or TMS to improve memory loss. What is TMS and how we're using it in your lab? You've used it quite a bit in the past?

Joel Voss: Yeah, we've been using it for the last several years. It's a large electromagnet and so it looks sort of like a little magic wand. They call it a butterfly coil or a figure eight. It's an electromagnet, so it receives an electrical signal from a big electrical pulse generator and there're a bunch of windings around this magnet and it pretty much just generates a very high intensity magnetic field that can be turned on and off very rapidly, like within a millisecond or so. What that does, when you have a magnetic field that varies in time, it can induce an electrical current at a distance and a wire. And so if you were to actually take this wand and hold it up to like a wire that you have a volt meter hooked up to and you pop the machine on and off real quick, there's nothing connecting the two, it can be a little bit of a distance, but the magnetic field will actually make an electrical current in that wire that you can measure with your volt meter. It's just a basic physical property and what we do like many labs around the world is we essentially aim this at the brain. In the brain there are sort of wires, the axons of neurons, which are kind of the, you know, the business of movers in players and the brain and so we aim it at those and it induces electrical activity in those neurons. It's a method to be able to produce activation in brain areas that you target non-invasively through the outside. It doesn't pass any electrical current through anyone's head. It actually creates the electrical current at a distance.

Erin Spain: When did you start using this as a, as a student before you were in the lab working with a study participants? When was your first time using this technology?

Joel Voss: Well, I started playing around with it when I was a postdoc at the University of Illinois, at the Beckman Institute, there was a TMS machine there that nobody was using, so I had some ideas about, you know, it's used in cognitive neuroscience research quite frequently as kind of a tool and so I started playing around with it there and kind of got used to it. When I came here to Northwestern, there was sort of a serendipitous opportunity to start applying it in a more rigorous way in our research and I took advantage of that and we started using it then somewhere about 2012.

Erin Spain: It's about bigger than your hand, the size of it and it goes up against the side of your head and it kind of makes a noise when it's activated, right?

Joel Voss: Yeah, a little clicking noise and that's the current turning on and off through the machine essentially. So each pulse creates a little click and depending on the particular experiment that we're running there'll be a different pulse train, by which I mean a different frequency or intensity. A rhythm really. The one used in this particular study was a 20 hertz rhythm, so it's 20 beats per second. It sounds like a really fast drum roll, essentially going on for a couple of seconds. In this study we did it for about a half an hour almost of the individuals who participated.

Erin Spain: And they experience no pain, as you said, maybe they feel just a little?

Joel Voss: Not for this particular study, in this particular target. It depends on how close you get the device to the various sensors that can pick up this electrical activity. Some parts of the head have muscles or cranial nerves or peripheral nerve, other kinds of things that if you stimulate them electrically are somewhat painful. So depending on where you go on someone's head, it can cause some pain and discomfort, but for this particular study, and for most of our studies, we try to avoid those regions and kind of tweak it so that it's not unpleasant for people. So mainly you just hear a popping sound and feel almost like a little tapping sensation on your scalp and that's about it.

Erin Spain: Let's talk a little bit about this study. You've been using this device in younger people for a while and you've shown some memory improvement. This is the first time you've looked at older adults over the age of 64. Tell me what you want it to achieve in this study.

Joel Voss: The brain is complex. Memory function is complex. We don't fully understand how the brain supports memory. We have some clues. The real question that we're addressing here is not just whether or not we can make people's memory better. I mean that's kind of the main question, but the real question is just can you use something like noninvasive brain stimulation to actually change the function of brain networks involved in memory and does that affect people's memory performance? That's really the fundamental question, which really seems as though it should be something that scientists and in their dark laboratories know the answer to, right? We should be able to manipulate memories however we want, but in fact, that's not true and really the basic question is just it possible to use a technology like this to reliably and in a predictable way, change brain function and memory ability.

Erin Spain: So when looking at this question, what steps did you take with each of these participants to prepare for this study? You took a look at their brains and some functional MRI. Tell me what went into setting this study up.

Joel Voss: This particular study is unique in its design and I'll kind of get into that, but you know, one of the first major challenges is just actually getting older people in the door. You can imagine this particular study involves a couple of weeks of activities, daily activities and so we had to find people who are willing to come into the laboratory every single day for, you know, a couple of week period and undergo MRI as well as brain stimulation using TMS and memory testing. Even undergoing an MRI you can't have metal on your body or other kinds of, you know, no pacemakers or anything like that for our particular parameters that we're using in the MRI scanner. We had to first identify a lot of people, bring them in and then for everyone in the study, we'd kind of do a baseline assessment where they get an MRI scan. We do a particular method called functional MRI where we tuned the MRI scanner to measure an indirect measure of brain activity. We use that to identify what we call the hippocampal cortical network and that is what we're aiming our stimulator at. So we use that baseline scan to say, "Where are we going to stimulate in this person's brain - this individual's brain"?

Erin Spain: Each individual's a little different?

Joel Voss: Yes, everybody's face is a little different and everybody's brain and head shape and everything else is a little bit different too. And so in order to reliably target a particular brain network, we need to get an individual measure of that brain network in each person using MRI. We do that and then we also give them some baseline memory testing, so we see how well they're doing at baseline, as well as a variety of other cognitive functions, like a tension and executive planning abilities and reaction time speeds and things like that. They get a whole battery of tests. After that they go home and we do some computer work and determine where we're going to be stimulating them and kind of plan out the whole study and whatnot and then they come in for the stimulation regimen. There are two conditions, which means two different ways that stimulation is applied in an active way and in a sham way for this particular study. In the active condition they are getting full intensity stimulation to the spot that we determined. It's kind of fun actually how we calibrate that intensity. In order to make sure that the device is kind of getting into their brain and affecting neural activity, we actually calibrate it off of affecting their motor cortex. So, you know, your body movements are all controlled by a very stereotyped pattern part of your brain, so there's one little spot that controls your thumb movements and another spot that controls your finger movements. You know, every major muscle group has a particular set of neurons that control it. And so we pretty much know where those are in each person's brain. That's actually pretty stereotyped across people. So we go into an area that controls their thumb and we pop the device at that thumb spot and we make their thumbs move around.

Erin Spain: Oh, come on! How do they react to that?

Joel Voss: Well, it's a little odd feeling. You know, it's kind of like, when you go to the doctor's office and you get your reflexes checked, you know, so you kind of involuntarily move your knee around it's the same sort of thing. Although, there's nothing actually touching you. So it's a little bit of a ghostly kind of experience, but you know, it's not too unsettling.

Erin Spain: It shows that the device is working.

Joel Voss: Definitely. We make sure that we can actually affect the brain and that's actually a nice feature and one of the reasons why we as a laboratory use TMS, transcranial magnetic stimulation, because we can verify and have a pretty high level of confidence that we're actually doing something to the brain tissue that we hope to target. Then the people go through five consecutive days of daily sessions of active stimulation. We go to this area, it's in their parietal lobe, so kind of like back behind your ear is where we're putting the device and they get stimulated using this high frequency drum roll, like pattern that turns on and off throughout 20 minutes while they sit around and watch nature videos to give them something to do. I don't actually think it necessarily does anything, but it gives people something to do and makes it so that they kind of can remain still without trying to talk and move their head around too much and things like that, so they do that every day for about 20 minutes. When it's all done, they go home. We wait for a full day, then they come back, put it back in the MRI scanner, test their memory again and that's what we use to assess change essentially. How much did your memory ability and your brain activity change as a function of having gotten this week of stimulation compared to what, how you were at baseline when you came in. There's also a sham condition. They do all this, then they go home for several weeks and they come back and they do it all again, but this time the machine, the volume is turned down essentially so that it's not actually affecting brain activity. We measure their memory and brain activity before and after that too so we can compare how much did active stimulation affect your memory and brain activity versus sham stimulation affecting your brain activity. People do that and we call a counterbalanced order, which just means that half the people in the study did act of stimulation first, then they do the sham and the other half of the people in the study do sham first and then active to make sure that it's not just some kind of practice effect or something going on that we are mis-identifying as the effects of brain stimulation.

Erin Spain: And the results were pretty exciting. You were able to improve the memory of older adults with age related memory loss to the level of younger adults that you've looked at in the past and you were actually able to make some changes, see some changes in the brain. Tell me about that.

Joel Voss: I like you said, we've also done similar kinds of studies and lots of different groups of young people and so we had some data of young people performing the exact same memory task. Not surprisingly, when older adults first walk through the door, they don't perform the memory task as well as young adults do because pretty much all people experience some level of age related memory weakening where they have not as good as the ability to remember new things as they did themselves years before and relative to other younger adults too. At baseline, so to speak, they were worse off than the young adults and significantly worse off compared to the young adults. After receiving their five days of stimulation and coming back a day later, we can no longer statistically tell them apart from the young adults, so it improved their memory to a level where they no longer were age impaired, so to speak. That is exciting. I mean it's pretty amazing actually when you think about it that that is possible to achieve, but from a scientific perspective, perhaps just as interesting is the way in which their memory changed and the way in which their brain activity changed along with that memory change. Memory is not a single thing. It's actually kind of a multifaceted ability and we're focusing on one particular aspect of memory that's called episodic memory. Think about it for a second. You can use your memory to solve all kinds of problems. So, what's the capital of France?

Erin Spain: Paris.

Joel Voss: Yeah. Perfect. There you go. You probably didn't think about any particular event in your life when you learned that information. That's just a fact. So that's kind of semantic memory. That's different than if I were to say, what did you have for breakfast today?

Erin Spain: My coffee ... my Starbucks.

Joel Voss: Yeah, there you go. Well maybe that's semantic too, because maybe you do it every day. But if you think about the specifics you can and you can relive those specifics and that's episodic memory and that's a particular kind of memory that the hippocampus, this area of the brain that we're targeting its network, that's the kind of memory that it cares about the most. It supports that. People show, compared to other kinds of memory, the most market decline and episodic memory ability with age and that's associated with a reduction and kind of functional and structural measures of the hippocampus as they get older. It's kind of the part of the brain that thins out with aging and it becomes slightly less functional. That kind of memory declines the most. What we did for the memory test is we have it set up so that people have to memorize novel associations. So for instance, in one portion of the task they see a set of items, so like objects, like keys or pictures of a pencil or a shoe or other random pictures we can find. So they see these objects and they have to memorize the objects, but they also have to memorize something associated with the objects. In one condition of the experiment, it's what we call a paired associate where they have to memorize that a different object goes with that object. Right? So like the pencil goes with the water bottle or something like that. Then in another condition of the experiment, they have to memorize a different kind of association that, where the object goes. So, they're looking at these things on a computer screen and maybe the pencil goes in the upper left and maybe the water bottle goes in the bottom right or something like that. Then later on we test their memory using a two-step procedure. So first off, we say here's a pencil picture. Did you see that? Yes or no? And they see some things that they did see originally that they did study. They also see some new things that they did not study. So maybe they studied a pencil but they did not originally see a car, I don't know, some picture or something and now they have to say yes or no to each one of those things. We can look at their accuracy at just being able to say, yes, I saw that object before, during this experiment. For the second step, for all the old objects, the ones they did study, they have to say, well, what was its associate? What was the other object it was associated with? Or where was it located on the screen? Even though it seems as though those are two fundamentally different kinds of things to remember, what they have in common is that the hippocampus cares about that. The hippocampus is the portion of the brain that's really the most critically responsible for building those kinds of associations. For the most part, it doesn't care about what the nature of the associations. It doesn't care if it's an object to a location. Where did I put my car keys? Or if it's a one thing with another, what's this person I just met to name that I heard a few minutes ago? The hippocampus does all that kind of stuff.

Erin Spain: How long is that battery of tests from when they first see the first object until they're tested on it?

Joel Voss: Not that long. They memorize about 84 of these things. It's a pretty big memory load. You've got to memorize 84 different pairings and then they wait. I think the average study tests delay is on the order of about 20 minutes or so. It's not that long, but it's a very good metric of how well your hippocampus is functioning. Your ability to do that predicts how well you're able to remember things over much longer delays because they're both good metrics of how well your hippocampus is operating. We have a very specific hypothesis. We think that stimulation should affect one type of memory. Your ability to put an object with its associate regardless of whether it's a location or whether it's a paired associate shouldn't care. Stimulation should effect that. It should not affect your ability to just recognize the objects in general because the hippocampus doesn't actually really care about that so much. Within the same experiment and even within the same memory test, we have a hypothesis where we should affect one type of memory if stimulation is doing what we think it's doing and not the other kind of memory and that's exactly what we found. And so there was no effect whatsoever - this is almost the same thing that we found with younger people in previous studies as well. It doesn't affect these kind of general, non-specific, non-hippocampal dependent forms of memory. It only is affecting the hippocampus dependent component of the memory task. That's an important validation that we're affecting the brain in the way that we hypothesize we would affect it. Their brain activity as well reflects this kind of specificity and the ability to predict what's going to happen in a similar kind of way. There are lots of different areas of the brain. There are lots of different functions that they support and we're only aiming at one network, which is the hippocampus and the regions that are kind of closely connected to it, i.e. the regions that participate with the hippocampus during memory, encoding and retrieval. It's a restricted set of regions that can be identified with MRI and separated from regions elsewhere. For this particular study, we picked a network of the hippocampus, a set of these regions, that have been shown in previous studies to be affected by aging. They're less well connected, they talk to each other less with aging and we had a control network, so a set of brain regions that really don't have as much to do with memory. They're involved in more executive planning and self-control and regulation and things like that. They're also affected by age, but they just don't have anything to do with memory. The effect of stimulation was to change the activity in the memory related brain regions, the hippocampal cortical network that we targeted, and not in that other control network or other areas of the brain. The reason why this is scientifically important, is that at some level we're kind of calling our shots and validating that the method that we're using is having the predicted effect on the brain that we hope it should be able to achieve.

Erin Spain: And it could open doors to looking at other areas of the brain and being this specific to test other hypothesis.

Joel Voss: Perhaps it could, it certainly will open the doors for us to be able to test other questions about memory. There's still as big a big open question about whether this same approach would be able to be used for other brain regions and networks. So at some level the stimulation we're doing is tailored to this network. We define it anatomically. We're also using stimulation patterns, so frequencies that we think should be relevant to this particular network. This network is also kind of unique that it has a hippocampus in it and this is a potentially critical thing. This is an area of the brain that we know is plastic. It has neuroplasticity, the ability to change in response in kind of a very pronounced way compared to a lot of other brain regions and other networks that don't have such a plasticity center might not be so amenable to changing in response to stimulation. It's kind of an open question, but being a memory laboratory, we're focusing on that.

Erin Spain: Yes. Well, and some people may have heard of TMS. The procedure was cleared by the FDA about 10 years ago to treat depression. But the way you're using it is different and actually maybe we don't really know how it's working in the brain to treat depression. Is that right?

Joel Voss: Yeah. Well, I mean at some level we don't understand in detail how it's working to achieve even the effects that we saw. You know, we were validating that it caused a predicted effect. We don't exactly know how. Figuring that out would require detailed kind of mechanistic studies with nonhuman animals where you can actually see what's happening in more detail. But, you're right, this procedure is FDA approved for treatment of depression. It's not the same procedure that we're doing. It's the same device ,TMS, but it's being applied in a, in a very different way for depression. It turns out that with depression, the areas that are most effective for treatment are in the frontal Cortex targeting a different network. A network of regions that are sort of related to depression. Interestingly enough, that network also includes a potential kind of plasticity center, the Amygdala, which, which could kind of explain why it might be effective in that condition as well.

Erin Spain: Well how long do the results typically lasts with the folks who are in your study because they come for a week and they do this testing and then you get the results back. Does it last for a few weeks, a few months? Do you know?

Joel Voss: Yeah, for this particular study, the primary outcome was 24 hours. It's there to demonstrate our ability to engage the target effectively. Did we cause the kinds of changes in the brain and performance that we would expect? We also did a delayed test, so people came back a week later and they were still elevated from where they were at baseline after receiving stimulation. But it was only weakly statistically different from sham at that point. Essentially the effect of stimulation became weaker over the course of that week and it was no longer very reliable at the one week follow up point. So it definitely decays after receiving stimulation. Interestingly, and this is something we're following up in a study right now in depression, likewise after receiving only a few days of stimulation, they don't really see much of a lasting antidepressant effect - the typical FDA approved treatment for depression involves five straight weeks of daily stimulation sessions. They see you lasting antidepressant effects for months afterwards. And so, you know, maybe simple minded ideas that if we stimulate for longer, we might get effects that last for longer as is the case for depression and so we're testing that out, right?

Erin Spain: Much longer study. Yes.

Joel Voss: That does take longer. Right now we have a study where older adults are coming into the lab for upwards of three weeks to see if we get a longer time course of effects afterwards.

Erin Spain: What is it that motivates you to continue going down this path looking at the different populations and getting deeper and deeper into memory loss and learning about that.

Joel Voss: Part of my research in the past was working with people with Amnesia, so debilitating memory impairment. You know, memory impairment is a sad and incredibly burdensome neurological condition to have. Right now, there are really no very effective treatments for it. To think that there might be something here with further development is a very strong motivating factor.

Erin Spain: Well, we're excited to see what's coming next. You know, maybe someday we'll all be able to go get our TMS procedures and go take a quiz or a test, or remember where our keys are.

Joel Voss: We can only hope.

Erin Spain: Thank you so much Joel Voss for coming today and sharing your research and we look forward to seeing what comes next.

Joel Voss: Thank you for having me.

Erin Spain: A note for physicians who listen to this program, you can now claim continuing medical education credit just by listening. Go to our website, and search CME.