April 8, 2019

expert reaction to brain stimulation to restore working memory in older adults

Research published in Nature Neuroscience describes a technique of noninvasive brain stimulation that resulted in improvement in working memory of performance in adults aged 60-76.

Dr Sven Braeutigam, Magnetoencephalography (MEG) Physicist at the Wellcome Centre for Integrative Neuroimaging, and Senior Researcher, Department of Psychiatry, University of Oxford, said:

“Increasing life expectancy implies, unfortunately, an increasing prevalence of cognitive decline in older adults. The effects of age-related decrease in our ability to think and remember are highly significant both at the individual and societal level, however, neither the biological mechanisms underlying such decrease are fully understood nor are consistent therapeutic approaches available.

“Here, the recent work by Reinhardt and Nguyen could be a real advancement in the debate. The authors employed transcranial electric stimulation targeting a specific neuronal mechanism, known as phase-amplitude coupling (PAC), in order to modulate the behavioural response in older adults performing a change-detection memory task. Critically, during and even after stimulation, task performance in older subjects reaches the level of young participants. Importantly, the observed correlation of task performance and PAC provides strong, arguably the best currently available insight into the neurophysiological foundations of human working memory.

“There is reason, however, to control ones enthusiasm because the authors applied only sham stimulation (a kind of placebo control) in young people, implying that it is currently not known to exactly what extent the effects observed are age specific. Nevertheless, the work by Reinhardt and Nguyen is important early stage work that may well stimulate fruitful research into effective treatments of age-related decline of human cognitive function.”

Dr Sara Imarisio, Head of Research from Alzheimer’s Research UK, said:

“To function properly the brain requires both electrical as well as chemical signals and research is beginning to investigate whether electrical stimulation of certain brain regions can help improve people’s memory and thinking.

“In this study, a non-invasive technique was shown to stimulate certain brain regions and improve memory function in cognitively healthy people. While this is a well-conducted study involving more people than other studies of its type, it did not look at people with dementia.

“What makes this study intriguing is that it tests a non-invasive approach unlike previous research in this area, which has often studied direct brain stimulation through invasive implants. Although this lays the ground work for future tests, further research is vital before we’ll know if this approach will provide long-lasting benefit for people with dementia.

“With few treatment options available for people with dementia, we must continue to explore every avenue for developing new treatments. Harnessing the latest technologies to develop new interventions to improve the lives of people living with dementia continues to be an important area of research for Alzheimer’s Research UK.”

Dr James Pickett, Head of Research at Alzheimer’s Society, said:

“We can’t cure, prevent or even slow down dementia so it’s vital we explore all possible areas for treatments.

“This study is interesting because it suggests this non-invasive technique using electrical stimulation may improve the working memory of older people. However, this research didn’t look at whether this might also be helpful for people with dementia.

“Altering and correcting the circuitry of the brain with technology is a new exciting avenue of research for dementia. Deep-brain stimulation, a surgical procedure used in Parkinson’s, is proof of principle that this approach may one day be fruitful for dementia. 

“Alzheimer’s Society is committed to spending at least £150m on dementia research in the next decade, which will include looking for innovative technological approaches with the potential to one day be part of the way we treat dementia.”

Dr Nir Grossman, Fellow, UK DRI at Imperial, said:

“The paper provides important evidence of the association between a type of working memory and oscillatory activities in the brain that involve synchronization within/between brain sites and within/between frequency bands.

“The inference that this kind of brain stimulation improved working memory performance in older participants however may not be fully supported by the data as participants always completed the task under the stimulation condition after completing the task without stimulation so there is potential improvement due to learning/training on the task that was not well controlled for.

““In addition, future studies should control more systematically for the effect of somatosensory activation, due to the stimulation of the skin underneath the electrodes (the authors state that there was no pain sensation but did not work to exclude sensation more generally), which has the potential to improve working memory performance via modulation of brain state such as vigilance and arousal.

“Nevertheless, the results that emerge from the study raise interesting questions about the brain activity that supports working memory and its mechanism of action. Regardless of the exact brain activity and mechanism involved, the sustained improvement in working memory that was observed after the short stimulation period may potentially elude to future research and rehabilitation potential.”

Prof Matt Jones, Professor of Neuroscience and Director of Bristol Neuroscience, said:

“This study builds on an extensive body of evidence from rodents, monkeys and humans showing that coordinated, rhythmic interactions allow brain regions to tune-in to one another and share information stored in memory.  The premise of the work is that these interactions may be impaired by aging.

“Experiments combined a simple test of short-term memory with non-invasive recordings of brain activity using scalp EEG electrodes and non-invasive electrical stimulation, again using electrode arrays placed on the scalp.  A well-designed series of tests were used to study the effects of age and transcranial stimulation.  Experimenters and subjects were blinded to stimulation conditions in any given session; of course, it is unlikely that experimenters were blind to subject age.

“The main behavioural test used is very simple: subjects were required to spot the difference between pictures presented 3s apart.  Nevertheless, subjects in their 60s and 70s were significantly slower and less accurate (averaging 8/10 trials correct) than subjects in their 20s (averaging 9/10).  This result appears solid: it corroborates previous studies, is based on groups of 40 participants-per-age, and is repeated across experiments.

“Memory impairments correlated with reduced rhythmic tuning between 2 cortical regions well-known to be central to memory processing, namely temporal and frontal cortex.  Again, this broadly corroborates previous studies – though it is not clear here whether communication failure only occurred during trials when subjects made mistakes.

“Previous work in young adults has shown that various flavours of rhythmic, electrical stimulation via scalp electrodes can modulate memory in this kind of task.  A similar approach was used here, but the stimulation was carefully focused on temporal and frontal cortex, plus parameters were tuned to individual participants’ brain waves, which were assessed prior to the main experiment. Stimulation was applied throughout a 25min testing session, normalising behaviour and brain activity in both aged participants and inaccurate younger subjects; these beneficial effects endured for at least 50min post-stimulation.

“One surprising aspect is that continuous stimulation consistently improved task performance whereas, under normal conditions, frontotemporal interactions are highly dynamic and only emerge when required.  This makes it harder to tell whether the stimulation made learning of pictures more accurate, holding of pictures in memory more robust, or recall of pictures more reliable.  Would stimulation have the same effect if only applied for the 3s between picture presentations, for example?  Would it also work if subjects had to remember multiple pictures in each trial?

“In a real-world scenario, using stimulation to continuously encourage or modify frontotemporal interactions might be expected to induce impairment by artificially “locking” networks of brain regions together.  However, the authors make no claims that such an approach would be viable, and their conclusions acknowledge that further work is required to understand mechanisms of age-dependent impairment and of the stimulation’s effects.

“The authors also acknowledge the importance of further work to understand whether “fixing” short-term memory might have knock-on benefits for other aspects of cognition.  In fact, there is evidence that frontotemporal interactions are also abnormal in conditions such as schizophrenia – should this type of stimulation also be brought to bear in psychiatry?

“Overall, the study does present frontotemporal interactions as an important target, either for further refinement of brain stimulation to augment memory, or for drugs that may mitigate against the effects of aging or brain disorders on cognition.”

Dr Vladimir Litvak, Associate Professor, Wellcome Centre for Human Neuroimaging, University College London (UCL), said:

“This is an interesting and exciting study showing that non-invasive brain stimulation can improve the performance in a task testing working memory in healthy older adults (62-75 years old) and bring it to a range comparable with younger people (20-29 years old). The stimulation was done in a very specific way by targeting two different brain areas involved in memory and synchronising their activity and the stimulation frequency was tuned individually for each subject based on the frequency of their natural brain rhythms. The authors showed that all these ingredients are essential for the procedure to work and also demonstrated a whole pattern of results in their analysis of EEG recordings from the subjects that seem consistent with their interpretation of how their stimulation works.  The press release is a fair summary of the findings. Some points I would suggest to consider when reporting on this study are as follows:

(1) The findings in the field of non-invasive electrical brain stimulation are notoriously difficult to reproduce (see e.g. https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0175635 ). There were even questions raised as to whether externally applied electrical current can stimulate the brain at all (see https://buzsakilab.com/wp/wp-content/uploads/2017/07/Underwoood-Cadaver-Science-16.pdf). The authors were able to reproduce their findings with two separate subject groups which is encouraging but it remains to be seen whether the method works in other people’s hands.

(2) The type of memory the authors were studying is working memory rather than episodic memory. So it’s not about remembering past events but about e.g. keeping the digits of a telephone number in mind while you are writing it down. The same kind of brain rhythms also play a role in episodic memory, so it is possible that a similar stimulation protocol could improve episodic memory as well. But this has not been shown.

(3) The older subjects in the study were healthy and did not have any memory impairments. They scored in the normal range in a routinely used dementia screening test. Thus, any differences they had with younger people were not of clinical significance and there is no evidence at the moment that the same procedure could help patients with dementia who are likely to have brain atrophy.

(4) Also there is no evidence that the kind of memory improvement that was achieved by the stimulation would make any difference for the subjects’ functioning in everyday life. 

(5) The stimulation effect lasted at least an hour after the stimulation ended but the authors did not test for longer periods, so it is not presently known how long it really lasts. To use this kind of method as treatment, the effect would have to be sustained between stimulation sessions.”

Prof Dorothy Bishop, Professor of Developmental Neuropsychology, University of Oxford, said:

“This study reports use of brain stimulation to improve performance on a memory task in adults aged between 60 and 76 years.

“It’s worth starting by saying what this research doesn’t show. It is a proof-of-concept study conducted on neurologically normal volunteers doing a lab-based experiment, and it would be premature to extrapolate the findings to everyday functioning in individuals with clinically significant memory problems. There is no indication that any beneficial effects of stimulation persist beyond the experimental session. Considerably more research would need to be done before concluding that this method had clinical application.

“The idea behind the research is certainly interesting: that memory in older people can be improved by using a particular kind of electrical stimulation designed to strengthen connections between brain regions. However, the research is essentially exploratory, with a relatively unconstrained set of measures that allow considerable flexibility in data analysis, and – presumably due to length limits in the journal – data from confirmatory experiments are presented very briefly. The next step is to confirm the results in a pre-registered study where the parameters of stimulation and the dependent measures are specified in advance.”

Prof Robert Howard, UCL Professor of Old Age Psychiatry, said:

“While this is an interesting and positive piece of research, suggesting that carefully positioned and timed electrical stimulation of the brain can potentially improve short-term memory function in older people, I would caution against any uncritical assumption that this will translate into clinical benefit. The findings need to be replicated under clinical trial conditions, with larger numbers of participants and with robust blinding of subjects and outcome assessors. The “real World” benefits of any apparent improvements in experimental working memory function associated with the technique will also need to be evaluated together with the impact of any potential adverse effects of brain stimulation. For example, induced improvements in working memory might come at the price of worsening of other areas of cognitive function.”

Prof Tim Griffiths, Professor of Cognitive Neurology, Newcastle University, said:

“The work demonstrates in older subjects a decline in visual working memory performance and an associated decline in low-frequency coupling between frontal and temporal areas and low-to-high frequency coupling within the temporal lobe. The most interesting aspect is that non-invasive low frequency alternating current stimulation improved the performance of the older subjects and these measures of coupling between and within areas.

“The work establishes a plausible mechanism for poorer working memory in older populations based on oscillatory interactions. In terms of the benefit from the manipulation the working memory task was not hard and the accuracy of the young (but not old) participants approached ceiling levels. It would have been hard to show any benefit in the young and they did not undergo stimulation: I am left wondering how specific the effect of stimulation is to the older population. A previous study of younger subjects from the group at the Montreal Neurological Institute showed a benefit from low frequency magnetic stimulation on working memory for sounds.

“An uncoupling deficit could explain loss of working memory ability as a function of normal ageing. However, I do not agree with authors’ claim that working memory deficits are central to Alzheimer’s disease in which these emerge only when the disorder is moderate to severe. In terms of the intervention, it would be very interesting to see how long the benefit might last beyond the 50 minutes studied here.”

‘Working memory revived in older adults by synchronizing rhythmic brain circuits’ by Robert Reinhart et al. was published in Nature Neuroscience at 16:00 UK time on Monday 8^th April.

Declared interests

Dr Sven Braeutigam: No declarations of interest.

Prof Matt Jones: No conflicts of interest.

Dr Vladimir Litvak: I provide paid consultancy services to OpenSatori Inc. an early stage start-up in the field of neural data analysis. They are not currently working on brain stimulation or memory enhancement.

Prof Dorothy Bishop: No conflicts of interest.

Prof Robert Howard: No conflicts of interest.

None others received