While smartphones and other digital assistants have been found to help people with mild memory impairment, their use by those with greater impairment has been less successful. However, a training program developed at the Baycrest Centre for Geriatric Care has been using the power of implicit memory to help impaired individuals master new skills.

The study involved 10 outpatients, aged 18 to 55 (average age 44), who had moderate-to-severe memory impairment, the result of non-neurodegenerative conditions including ruptured aneurysm, stroke, tumor, epilepsy, closed-head injury, or anoxia after a heart attack. They all reported difficulty in day-to-day functioning.

Participants were trained in the basic functions of either a smartphone or another personal digital assistant (PDA) device, using an errorless training method that tapped into their preserved implicit /procedural memory. In this method, cues are progressively faded in such a way as to ensure there is enough information to prompt the correct response. The fading of the cues was based on the trainer’s observation of the patient’s behavior.

Participants were given several one-hour training sessions to learn calendaring skills such as inputting appointments and reminders. Each application was broken down into its component steps, and each step was given its own score in terms of how much support was needed. Support could either comprise a full explanation and demonstration; full explanation plus pointing to the next step; simply pointing to the next step; simply confirming a correct query; no support. The hour-long sessions occurred twice a week (with one exception, who only received one session a week). Training continued until the individual reached criterion-level performance (98% correct over a single session). On average, this took about 8 sessions, but as a general rule, those with relatively focal impairment tended to be substantially quicker than those with more extensive cognitive impairment.

After this first training phase, participants took their devices home, where they extended their use of the device through new applications mastered using the same protocol. These new tasks were carefully scaffolded to enable progressively more difficult tasks to be learned.

To assess performance, participants were given a schedule of 10 phone calls to complete over a two-week period at different times of the day. Additionally, family members kept a log of whether real-life tasks were successfully completed or not, and both participants and family members completed several questionnaires: one rating a list of common memory mistakes on a frequency-of-occurrence scale, another measuring confidence in dealing with various memory-demanding scenarios, and a third examining the participant's ability to use the device.

All 10 individuals showed improvement in day-to-day memory functioning after taking the training, and this improvement continued when the patients were followed up three to eight months later. Specifically, prospective memory (memory for future events) improved, and patient confidence in dealing with memory-demanding situations increased. Some patients also reported broadening their use of their device to include non-prospective memory tasks (e.g. entering names and/or photos of new acquaintances, or entering details of conversations).

It should be noted that these patients were some time past their injury, which was on average some 3 ½ years earlier (ranging from 10 months to over 25 years). Accordingly, they had all been through standard rehabilitation training, and already used many memory strategies. Questioning about strategy use prior to the training revealed that six participants used more memory strategies than they had before their injury, three hadn’t changed their strategy use, and one used fewer. Strategies included: calendars, lists, reminders from others, notebooks, day planner, placing items in prominent places, writing a note, relying on routines, alarms, organizing information, saying something out loud in order to remember it, mental elaboration, concentrating hard, mental retracing, computer software, spaced repetition, creating acronyms, alphabetic retrieval search.

The purpose of this small study, which built on an earlier study involving only two patients, was to demonstrate the generalizability of the training method to a larger number of individuals with moderate-to-severe memory impairment. Hopefully, it will also reassure such individuals, who tend not to use electronic memory aids, that these are a useful tool that they can, with the right training, learn to use successfully.

When a middle-aged woman loses her memory after sex, it naturally makes the headlines. Many might equate this sort of headline to “Man marries alien”, but this is an example of a rare condition — temporary, you will be relieved to hear — known as transient global amnesia. Such abrupt, localized loss of autobiographical memory is usually preceded by strenuous physical activity or stressful events. It generally occurs in middle-aged or older adults, but has been known to occur in younger people. In those cases, there may be a history of migraine or head trauma.

Following an earlier study in which 29 of 41 TGA patients were found to have small lesions in the CA1 region of the hippocampus, scanning of another 16 TGA patients has revealed 14 had these same lesions. It seems likely that all the patients had such lesions, but because they are very small and don’t last long, they’re easy to miss. The lesion is best seen after 24-72 hours, but is gone after 5-6 days.

At the start of one of these attacks, memory for the first 30 years of life was significantly impaired, but still much better than memory for the years after that. There was a clear temporal gradient, with memory increasingly worse for events closer in time. There was no difference between events in the previous year and events in the previous five years, but a clear jump at that five-year point.

The exact location of the lesions was significant: when the lesion was in the anterior part of the region, memory for recent events was more impaired.

The hippocampus is known to be crucially involved in episodic memory (memory for events), and an integral part of the network for autobiographical memory. In recent years, it has come to be thought that such memories are only hosted temporarily by the hippocampus, and over a few years come to be permanently lodged in the neocortex (the standard consolidation model). Evidence from a number of studies of this change at the five-year mark has been taken as support for this theory. According to this, then, older memories should be safe from hippocampal damage.

An opposing theory, however, is that the hippocampus continues to be involved in such memories, with both the neocortex and the hippocampus involved in putting together reconsolidated memories (the multiple trace model). According to this model, each retrieval of an episodic memory creates a new version in the hippocampus. The more versions, the better protected a memory will be from any damage to the hippocampus.

The findings from this study show that while there is indeed a significant difference between older and more recent memories, the CA1 region of the hippocampus continues to be crucial for retrieving older memories, and for our sense of self-continuity.

Interestingly, some studies have also found a difference between the left and right hemispheres, with the right hippocampus showing a temporal gradient and the left hippocampus showing constant activation across all time periods. Such a hemisphere difference was not found in the present study. The researchers suggest that the reason may lie in the age of the participants (average age was 68), reflecting a reduction in hemispheric asymmetry with age.

There’s another message in this study. In these cases of TGA, memory function is restored within 24 hours (and generally sooner, within 6-10 hours). This shows how fast the brain can repair damage. Similarly, the fact that such tiny lesions have temporary effects so much more dramatic than the more lasting effects of larger lesions, is also a tribute to the plasticity of the brain.

The findings are consistent with findings of a preferential degeneration of CA1 neurons in the early stages of Alzheimer's disease, and suggest a target for treatment.

A study involving five patients with severe amnesia due to damage in the hippocampus, resulting in a condition comparable to Alzheimer's, has found that memory tests given 5-10 minutes after sad and happy film clips showed little (if any) memory of the details, but the generated emotion lasted for 20 to 30 minutes afterward. Interestingly, normal controls also felt happy for about the same length of time, but the impact of sad scenes was shorter. The findings challenge the idea that by minimizing a specific memory of past trauma, associated sadness will also decrease. Indeed, it may be that forgetting the details of unhappy events prolongs the effects. The findings also point to the need for care in dealing with those with impaired memory — don’t assume that any induced emotion will vanish as quickly as their memory of it.

[471] Feinstein JS, Duff MC, Tranel D. Sustained experience of emotion after loss of memory in patients with amnesia. Proceedings of the National Academy of Sciences [Internet]. 2010 ;107(17):7674 - 7679. Available from:

Older news items (pre-2010) brought over from the old website

Hypnosis study sheds insight on amnesia

An intriguing study investigating brain activity of hypnotically induced forgetting may shed light on amnesia. Researchers showed volunteers a documentary depicting a day in the life of a young woman, followed a week later with a brain scan while they were put into a hypnotic state. They were given a posthypnotic suggestion to forget the movie, and a reversibility cue that would restore the memory. When their recall of the movie was later tested, those susceptible to posthypnotic amnesia showed reduced recall. Brain scans revealed different brain activity patterns between those susceptible and those who were not. For the susceptible, activity in some brain regions was suppressed during memory suppression, while activity in other regions increased. But when the posthypnotic suggestion was reversed, the susceptible group showed recovery of activity in suppressed regions. The findings suggest that suppression was exerted at early stages of the retrieval process, specifically, an executive pre-retrieval monitoring process that produces an early decision on whether to proceed or not on retrieval. The researchers suggest that some forms of amnesia may be a consequence of this ‘preretrieval memory abort’ mechanism.

Mendelsohn, A., Chalamish, Y., Solomonovich, A. & Dudai, Y. 2008. Mesmerizing Memories: Brain Substrates of Episodic Memory Suppression in Posthypnotic Amnesia. Neuron, 57, 159-170.

Statins associated with rare cases of temporary amnesia

Two recent studies have documented cases of amnesia and other nervous-system side effects after taking statins, the cholesterol-lowering drugs being prescribed to millions of people at risk of heart disease. It is emphasized that this is a rare problem, but given the vast numbers of people taking statins, it might still add up to a significant number of problems.

[2389] Wagstaff LR, Mitton MW, Arvik BML, Doraiswamy MP. Statin-Associated Memory Loss: Analysis of 60 Case Reports and Review of the Literature. Pharmacotherapy [Internet]. 2003 ;23(7):871 - 880. Available from: