Does TMS Cause Memory Loss? What the Research Shows

Transcranial magnetic stimulation does not cause memory loss according to the available clinical evidence. Across studies testing memory before and after treatment courses, researchers consistently find that TMS leaves memory intact and, in many cases, measurably improves it. The concern is understandable because TMS is sometimes mentioned alongside electroconvulsive therapy, which can affect memory, but the two work very differently in the brain. The real story behind TMS and cognition turns out to be more interesting than a simple safety reassurance.

What Cognitive Testing After TMS Treatment Actually Shows

The most direct way to answer this question is to look at studies that gave patients memory and cognitive tests both before and after a full course of TMS. A review of safety evidence across neuropsychological domains concluded that significant cognitive impairment from a course of TMS is unlikely.¹ Rather than showing decline, one study of depression patients found that right-sided TMS was associated with gains of more than one standard deviation in language and visuospatial function, and over two standard deviations of improvement in verbal episodic memory compared to left-sided stimulation. Those improvements appeared to be independent of depression relief itself.²

This pattern holds across different patient populations. In a study of Parkinson’s disease patients with depression, those who received TMS showed improved performance on tests of executive function and cognitive flexibility, with no adverse effects on any neuropsychological measure in the testing battery.³ The cognitive gains were comparable to those seen with the antidepressant fluoxetine.

Why People Confuse TMS With ECT

A big reason patients worry about TMS and memory is that they’ve heard stories about electroconvulsive therapy causing confusion and memory gaps. ECT deliberately induces a seizure under anesthesia and can temporarily disrupt the formation of new memories, especially around the time of treatment. TMS, by contrast, uses focused magnetic pulses to stimulate specific brain areas without sedation, without triggering a seizure, and without the systemic brain effects that ECT produces.

Head-to-head comparisons bear this out. A randomized study comparing the short-term cognitive effects of ECT versus TMS in depressed patients found that scores on tests of attention, processing speed, and executive function were significantly better in the TMS group at the end of treatment.⁴ If you’re choosing between the two treatments and memory preservation is a priority, that distinction matters.

How TMS Can Actually Improve Working Memory

Not only does TMS appear safe for memory, a growing body of research suggests it can enhance certain types of it. Working memory, the mental workspace you use to hold and manipulate information in the moment, has been one of the best-studied targets. A systematic review and meta-analysis found that TMS aimed at the dorsolateral prefrontal cortex significantly improved working memory performance across multiple measures.⁵

Individual experiments confirm this. When healthy volunteers received high-frequency TMS over the left dorsolateral prefrontal cortex, their performance improved on both verbal digit-span tasks and visuospatial working memory tasks.⁶ The effect was specific to the stimulation site and frequency, not just a general arousal boost. A study examining theta burst stimulation at different intensities observed an inverted-U relationship where a moderate intensity produced the largest changes in brain activity linked to working memory, though behavioral differences between intensity conditions were not significant in the healthy participants tested.⁷

Targeting the Hippocampal Network for Episodic Memory

Episodic memory, your ability to recall specific events and experiences, depends heavily on the hippocampus and its connections to the cortex. The hippocampus sits deep in the brain where magnetic pulses can’t reach it directly, but researchers found a workaround: stimulating a spot on the parietal cortex that is functionally connected to the hippocampal network.

A large meta-analysis of studies using this hippocampal-targeted approach found a robust positive effect on episodic memory retrieval, with the improvement well above conventional thresholds for a meaningful effect size.⁸ A replication study confirmed that this approach increased episodic memory performance one day after the final stimulation session, though the benefit did not persist at one week.⁹ That durability gap is one of the open questions in the field.

In a small study of Alzheimer’s disease patients, gamma-frequency TMS targeted at the precuneus produced improvements in immediate word recall in three out of four participants. Three months after treatment, broader gains were still apparent in immediate and delayed recall, attention, and verbal fluency.¹⁰ This is very preliminary work with a tiny sample, but it illustrates where the research is heading.

TMS in Alzheimer’s Disease and Mild Cognitive Impairment

If TMS actually harmed memory, the last population you’d want to test it in would be people already losing theirs. Yet researchers have been studying TMS extensively in Alzheimer’s disease and mild cognitive impairment precisely because the evidence points the other direction. A systematic review and meta-analysis covering these populations found large effect sizes favoring active TMS over sham stimulation on standard measures of global cognition.¹¹ Out of 143 studies reviewed, only two reported seizures as adverse events, three of which were judged unrelated to TMS and one resolved with coil repositioning.

Another meta-analysis found a medium-to-large overall effect size favoring active TMS over sham for cognitive improvement in MCI and Alzheimer’s patients. The analysis showed that high-frequency stimulation of the left dorsolateral prefrontal cortex and low-frequency stimulation of the right side both improved memory, and the benefits of a treatment course could persist for four to twelve weeks.¹² A separate trial reported that TMS improved overall cognitive scores, memory, and language in Alzheimer’s patients, with the strongest effects in those at the mild stage of the disease.¹³

In patients with mild cognitive impairment specifically, a recent trial found that TMS significantly increased memory quotient scores and improved both long-term and short-term memory compared to a control group.¹⁴ Immediate memory was the one subtype that didn’t show significant improvement, hinting that TMS may preferentially strengthen consolidation and retrieval rather than the initial encoding of information.

Why Stimulation Settings Matter So Much

The answer to “does TMS affect memory” depends heavily on how the stimulation is delivered. Frequency, intensity, brain target, and timing all influence the outcome, and some combinations can actually impair performance in experimental settings.

An early controlled study found that high-frequency TMS applied to the left dorsolateral prefrontal cortex during a memory task temporarily reduced episodic verbal memory scores.¹⁵ This is worth understanding in context: the stimulation was delivered while participants were actively performing the memory task, effectively disrupting the brain region’s function in real time. That’s quite different from the therapeutic use of TMS, where stimulation sessions and the cognitive demands of daily life are separated. A study targeting the parietal cortex during a working memory task found that only 5 Hz stimulation improved reaction times, while 1 Hz and 20 Hz had no effect, showing how narrowly tuned the frequency-response relationship can be.¹⁶

Animal research underscores how much intensity matters. In aging mice, low-intensity stimulation activated growth factors that support synaptic connections and reversed spatial memory problems. High-intensity stimulation did the opposite, damaging synaptic structures and worsening learning and memory.¹⁷ A more recent mouse study found that continuous 40 Hz stimulation at extremely low intensity impaired exploratory behavior and spatial memory, while 10 Hz stimulation did not have the same harmful effect.¹⁸ These animal findings don’t translate directly to human clinical protocols, but they reinforce that the parameters used in a TMS session aren’t arbitrary. The protocols approved for clinical depression treatment have been optimized over decades and sit in a range that research consistently shows is cognitively safe.

When Memory Complaints Are Really About Depression

There’s an important wrinkle that many patients and even some clinicians miss. People undergoing TMS for depression often report feeling mentally “foggy” or forgetful. But depression itself is one of the most powerful drivers of subjective memory complaints. Research has shown that improvements in depression and fatigue are associated with fewer perceived cognitive problems, even when objective test scores haven’t changed.¹⁹

A study examining what predicts subjective cognitive complaints found that depression and rumination were significant predictors, while actual objective memory performance was not.²⁰ In practical terms, this means a patient in the middle of a TMS treatment course who still feels depressed may report memory difficulties that have nothing to do with the stimulation. As their depression lifts, those subjective complaints tend to resolve. Patients who are partway through treatment and still symptomatic are in the worst spot for this kind of misattribution.

How TMS Reshapes Brain Network Activity

TMS doesn’t just zap one spot in the brain and leave everything else unchanged. It alters the way networks of brain regions communicate with each other, which helps explain both its therapeutic effects and its cognitive safety profile. Research on the default mode network, which is overactive in depression and plays a role in self-referential thought and memory, found that TMS reduced abnormal hyperconnectivity within this network. It also shifted the relationship between the default mode network and the central executive network, the system you rely on for focused attention and problem solving.²¹

A study of theta-burst TMS confirmed that stimulating the left dorsolateral prefrontal cortex reduced functional connectivity between that region and the default mode network.²² In depression, the default mode network tends to be excessively coupled with executive regions, which may contribute to rumination and difficulty concentrating. By normalizing that connectivity, TMS may simultaneously relieve depression and free up cognitive resources. This could partly explain why many patients report thinking more clearly after treatment, not less.

Side Effects That Actually Occur

The common side effects of TMS are physical, not cognitive. In a large clinical trial of theta-burst stimulation for depression involving over 200 participants receiving active treatment, about two-thirds reported headaches, roughly one in ten had dizziness, and smaller numbers experienced fatigue or nausea.²³ Scalp discomfort at the stimulation site is also common, especially in the first few sessions, and tends to diminish as patients get accustomed to the sensation.

Seizure is the most serious potential adverse event, but it is rare. A literature review estimated the overall risk at less than one percent, comparable to the seizure risk associated with most commonly prescribed psychiatric medications. When seizures do occur during TMS, they are typically self-limiting and resolve without lasting effects.²⁴ International safety guidelines have been in place since the 1990s to keep stimulation parameters within safe bounds, and those guidelines were informed partly by the handful of seizures that occurred in TMS’s early years before standardized protocols existed.²⁵

TMS Safety in Adolescents

Parents considering TMS for a teenager with depression understandably worry about effects on a developing brain. The evidence so far is reassuring. A mixed-methods study of adolescents receiving TMS for depression found that the treatment was safe and well-tolerated, with adverse event frequency and type no different from what adults experience.²⁶ A randomized, double-blind, sham-controlled pilot trial of low-frequency TMS in adolescents with first-episode major depression found no significant differences in cognitive battery scores between the active and sham groups, and side effects like fatigue and headache were similar in both arms.²⁷ The adolescent evidence base is still smaller than the adult one, but nothing in it points toward memory harm.

The Unresolved Question of How Long Benefits Last

If TMS can improve memory, a natural follow-up is how long those gains stick around. The answer is encouraging but incomplete. In Alzheimer’s and MCI populations, treatment effects on cognitive scores have been documented lasting four to twelve weeks after a course of sessions.¹² Hippocampal-network-targeted TMS produced episodic memory improvements that were present one day after the final session but had faded by one week.⁹

Researchers are still figuring out the optimal number of sessions, maintenance schedules, and which patient characteristics predict a longer-lasting response. Some of the variability likely comes down to the same parameter sensitivity that determines whether stimulation helps or hinders in the first place. For now, the durability question is more practically relevant for people seeking cognitive enhancement than for people worried about harm: the memory changes from clinical TMS protocols are generally positive, and when those positive changes fade, they fade back to baseline rather than below it.

One Cautionary Finding Worth Knowing About

A randomized controlled study in treatment-resistant depression tested verbal memory before and after TMS and found no significant change in the treatment group. Interestingly, the sham group actually showed improvement in instant recall and long-term recall, likely from practice effects on the test itself.²⁸ This is sometimes cited by people searching for evidence that TMS “did nothing for memory” or even held it back relative to sham. But the finding is better read as showing that TMS didn’t improve verbal memory in that particular protocol and patient sample, while also not impairing it. The absence of improvement is not the presence of harm, and practice effects in sham groups are a well-known complication in cognitive testing research.

This study is a useful reminder that TMS doesn’t universally enhance every type of memory in every patient. The cognitive effects depend on which brain region is targeted, what frequency and intensity are used, and what cognitive domain you measure. Memory is not one monolithic function. The literature broadly shows that clinical TMS protocols spare memory at worst and improve aspects of it in many cases, but no honest reading of the evidence would promise blanket cognitive enhancement from every treatment course.