Exploring why “memory loss” is a misleading descriptor of people living with dementia and can lead to dysfunctional care

Explicit memory

Explicit memory or declarative memory (Squire, 1994) is the conscious recollection, either by recall or recognition, of recent or remote events. Numerous cognitive assessment tools aim to measure explicit memory specifically and serve as valuable screening tools (Mondini et al., 2023), for retrieval from memory via recall is clearly of great importance in everyday life. In many of these commonly employed outcome measures (e.g., MMSE, MoCA, Boston Naming Test) used in drug studies as well as assessment “tools” used in nursing homes and hospitals, recall is the most frequently tapped method of retrieval from memory, whether it’s the recall of answers to “orientation” questions or the recall of the names of objects, or the immediate and/or delayed repetition of spoken words or sentences. In each case, there is one and only one correct answer to the question asked. When people who are living with dementia repeatedly ask the same questions that have been answered already, there is a failure on their part to recall having asked the question and having had that question answered. Thus, in most cases, it is a failure of recall that leads to the description of “memory loss”. Interestingly enough, retrieval via recall even in what might be termed normal ageing, becomes far more difficult than is retrieval via recognition (Rhodes et al., 2019).

For example, word-finding problems are failures to recall a particular word. Failure to recall does not necessarily mean a failure to remember the word, or “memory loss”, or the erasure of the word from the person’s “mental lexicon”. The following example is illustrative (see Sabat, 1991). Dr M had been diagnosed with Alzheimer’s disease (AD) two and a half years before this conversational extract and had experienced memory problems for seven years prior to that. Here, she exhibits failure to recall a particular word; duration of pauses is noted in brackets.

Dr M: I happened to have a visit to my uh [2.4 seconds] uh [12.5 seconds], the person who takes care of me when I have any problem physically.

SRS: A dentist?

Dr M: No, not in this case.

SRS: It’s not a mailman?

Dr M: No, it’s a female person (laughs)

SRS: A physician?

Dr M: Yes!

Here, we see that (a) Dr M failed to recall the word, “physician”, but accurately described that person’s vocation via a circumlocution, and (b) she recognized the word she sought when given a choice of words, while simultaneously making the nuanced discrimination between two types of health care professionals, only one of which was the type to whom she was referring. In order to recognize correctly the word she sought, she must have been able to remember it in the first place. Thus, we note that failure to recall does not necessarily mean failure to remember and that failure to recall itself does not mean “memory loss”.

Another example of this phenomenon (Sabat, 2017) can be found in the following interaction between the above author and a person who had been diagnosed with dementia two years earlier and had an MMSE score of 22/30. A few minutes after finishing a dinner meal that consisted of baked salmon, vegetables, and salad, Mr U and the aforementioned author had the following conversational exchange:

Mr U: See, now I don’t even know what I ate for dinner.

SRS: I don’t know about that. Let’s see. Did you have chicken?

Mr U: No, I think it was fish.

SRS: That’s exactly right. You had salmon.

Mr U: (To his wife and adult daughter) When is this guy coming back here again?

In this exchange, Mr U at first had no conscious recall of exactly what he had for dinner. He did, however, recall without any prompt that he had eaten dinner. When given a prompt about another possible but incorrect choice, however, he replied with the rather less than definitive answer, “No, I think it was fish”, which was correct. Again, the lack of immediate recall about the details of what he ate, the fine-grained episodic memory thereof, did not mean that he had “memory loss” at all. In addition, there was a wide range of possibilities regarding what, exactly, he had for dinner, but when there was no pressure on him to provide the one and only correct response, he did evince a “feeling” that it was fish and he was correct. Therefore, he did make a memory of exactly what he ate, but required a cue or another way of gaining access to that memory. In both these cases, the dysfunction, to one or another degree, is in explicit memory. There are, however, other memory systems that do not entail, or tap into, explicit memory systems and these are known as implicit memory systems.

Implicit memory

Implicit memory is inferred from a change in a person’s performance or action that occurs as a result of a previous experience that the person may deny, or not recall, having had (Cubelli & Della Sala, 2020). Specifically, the person may not be able to recall consciously this or that previous experience, but his or her actions will reflect, or at least imply the existence of, a memory of that experience nonetheless (Howard, 1991; Roediger, 1990). Implicit memory is another component of memory systems reflective of learned behaviors that are important for activities of daily life (Amodio & Ratner, 2011; Cubelli & Della Sala, 2020).

The difference between explicit and implicit memory is illustrated by the Word-stem Completion Task as presented to persons diagnosed with AD (Wit et al., 2022). The person in question is presented with a list of words to be studied, including the word, “defend” and is subsequently tested in two different ways on his or her memory of the words on that list. One method of testing taps the person’s explicit memory, wherein the person is given the word stem, “def____” and is asked to fill in the blank so as to make a word that was on the list that he or she studied previously. A different method of testing taps the person’s implicit memory, wherein the person is asked to fill in the blank so as to make the first word that comes to mind. There are clear and crucial differences in what the person is asked to do in each method because in the latter method, (a) there is no mention of the list of words that the person was asked to study previously, and (b) there isn’t one and only one correct response to saying “the first word that comes to mind”.

In the method that tests explicit memory, it is common for the person living with AD to respond by asking, “What list?” By questioning the person in this way, the person with AD exhibits a dysfunction in recall, a type of explicit memory wherein he or she seems to have no memory of having previously studied the list of words in question. If, however, instead of asking the person to fill in the blank to make a word from the list he or she studied previously, the person is asked to fill in the blank to make “the first word that comes to mind”, the person fills in the blank with “end” so as to make the word, “defend” which was on the list that he or she does not recall having studied. The correct answer the person provided here was not the result of a lucky guess, as there is a plethora of words that begin with “def”. These findings were demonstrated in people diagnosed with AD in the early stages more than thirty years ago (Morris & Kopelman, 1986) and in the mild to moderate stages as well (Grosse et al., 1990; Knopman, 1991; Partridge et al., 1990; Randolph et al., 1995; Russo & Spinnler, 1994).

Implicit memory has been demonstrated in people living with AD in ways other than word-stem completion tasks. For example, Mitchell (1988) found a “savings score” for people with AD when it took them less time to name a picture upon seeing that picture a second time than it took when they viewed the picture initially even though they did not recall having seen the picture previously. Thus, the previous experience had a facilitative effect on the subsequent experience, an effect that could not have occurred in the absence of some sort of memory ability. The savings score that Mitchell found in people with AD was equivalent to that of younger and older adults deemed healthy. The effect seen here is often referred to as a “priming effect” and priming of verbal information can have long-term effects even in people diagnosed with amnesia (McAndrews et al., 1987). Implicit memory in people with AD has been shown also in tasks such as the Gollin Incomplete Pictures task (Gollin, 1960), wherein the person is shown pictures of objects and asked to identify them in various degrees of completeness, beginning with the least complete and ending with the most complete. With practice, people with AD learned to identify the pictures at lower levels of completeness than they did without practice even though they had no explicit memory of, did not recall, having practiced the task (Gabrieli, et al., 1993). Similar findings regarding reading speed (Monti, et al., 1994), and music processing (Deason et al., 2019) have been reported.

Other preserved memory abilities

Explicit episodic memory is the type of memory most affected by the disease process of AD, but implicit memory types are relatively spared and can contribute to familiarity that informs recognition (Deason et al., 2019). Explicit and implicit memory are only two of many types of memory. Not only are there several types of memory systems (i.e., short-term; working; long-term; declarative/explicit; nondeclarative/implicit; autobiographical; episodic; emotional, etc.), but also different ways in which they interact in the daily milieu of human existence (Poldrack et al., 2001). Furthermore, each type of memory is associated with multiple brain regions rather than restricted to a single area. Although the hippocampus is a main “memory hub” (Schwabe & Bolam, 2017), and one quite vulnerable to the degeneration associated with dementia, it is not the only brain region responsible for memory. Several other brain structures contribute to memory function that are not immediately affected by dementia-related deterioration. What’s more, some regions may even compensate by upregulating their activity (Cope et al., 2022). Understanding the ways in which several forms of memory and consciousness are preserved in persons with dementia can improve quality of life and foster meaningful communication. A concerted effort to employ strategies that maximize use of these preserved functions is essential for successful support.

Owing to considerable neural plasticity and dynamic brain network functions, variability in deficit severity may be observed throughout the disease process (Mashour et al., 2019). To caregivers, this might appear as symptoms that wax and wane depending on a myriad of factors (e.g., amount of sleep, food ingestion, exercise, etc.) (Griffin et al., 2022; Ramirez et al., 2023). These are commonly characterized as “good” or “bad” days by caregivers (Rockwood et al., 2014). Remarkably, there have been observations of significant lucidity despite an advanced stage of disease (Griffin et al., 2022; Mashour et al., 2019; Ramirez et al., 2023). The salient concept is that the severity of memory problems, or likewise, the ability to retrieve information from memory, fluctuates greatly. Indeed, persons with dementia have memory difficulties, but to say they have memory loss is clinically inaccurate. Furthermore, the memory difficulties that manifest can vary by dementia type. For example, a person with AD may have difficulty with episodic memory of events and personal experiences, but maintain procedural memory (learned skills, habits) (Kudlicka et al., 2019), while other forms of dementia, such as posterior cortical atrophy, affect memory quite sparingly and instead result in visual deficits (Crutch et al., 2012; DeTure & Dickson, 2019). Regardless of the type of dementia, it has long been known, but rarely discussed, that persons with dementia retain considerable cognitive capacities and are capable of new learning (Kudlicka et al., 2019). For example, studies examining cognitive plasticity and learning potential in persons with MCI and AD have demonstrated improved performance in visual memory, verbal learning, and executive function in all participants (Backman, 1992; Fernandez-Ballesteros et al., 2006), challenging the enduring notion that persons with dementia lose their memory but are also incapable of creating new memories.

In addition to the preservation of implicit memory in general, persons with dementia maintain other components under this classification, including emotional memory, emotional communications, a variety of skills, and ability to learn. Persons with dementia maintain gaze with their partners, retain emotional traces of film clips, and can have accentuated responses to social contagion (Fredericks et al., 2018). Additionally, though some brain networks deteriorate, those that become heightened can at times account for psychosocial changes (Cope et al., 2022; Warren, 2022). These networks correspond to heightened emotions and implicit memory and, when negative, are often mistaken for the behavioral and psychological symptoms of dementia, rather than meaningful attempts to communicate unmet needs (Warren, 2022). Furthermore, research supports that even during cognitive decline, persons with cognitive impairments can continue to learn new skills and thus create new memories (Sabat, 2018).

Considering the evidence, dementia is a syndrome of memory retrieval difficulties rather than loss and characterizing it as such can have far reaching consequences for persons with dementia, especially how they are treated and the fate of their quality of life.

Examples from everyday life

The preceding experimental findings regarding the implicit memory abilities of people living with dementia have indeed been mirrored in their everyday lives. The case of Mr C, diagnosed with mild to moderate AD, and had always done the landscaping around the house in which he lived with his wife, is illustrative (Sabat, 2018). Although he was accustomed to mowing the lawn and trimming the shrubs, his wife became anxious about his using the lawnmower even though he never had any accidents. Thus, she put a lock on the shed in the back garden where the lawnmower and other gardening tools were kept. Mr C, finding the door of the shed now locked, broke the lock, retrieved the lawnmower, and mowed the lawn without incident. Subsequently and without any discussion with her husband, Mrs C arranged with John, the oldest of their three adult children who lived in the area, to take the lawnmower to his house. When the grass needed cutting again, Mr C went to the shed, found the lawnmower missing, and reported to his wife that it had been stolen. Mrs C then explained that John agreed to take the lawnmower to his house because she and John agreed that it might be dangerous for Mr C to continue using the lawnmower. Five days later, the three adult children came for dinner with their parents. Mr C greeted the two younger children warmly, but when John arrived, Mr C uncharacteristically ignored him completely. When Mrs C asked him if he was angry with John, Mr C said, “Yes”. When asked why he was angry, Mr C said, “I don’t know”.

Thus, Mr C had no explicit memory of, could not recall, the details of why he was angry with John, but he did, in fact, make a memory of John having been involved with essentially disenfranchising him regarding his use of the lawnmower and how offensive that was to him. If one did not know about implicit memory and if one assumed that Mr C had “memory loss”, one would easily assimilate Mr C’s anger into a story line that draws on the pathology focused narrative of dementia, and interpret his anger as an example of irrational hostility rather than the appropriate righteous indignation. The pathology-based interpretation in turn would then fuel the family members’ beliefs that Mr C’s AD “was growing more severe” because Mr C had never before exhibited “irrational hostility”. It is curious that simply because his family did not understand why Mr C was angry, they interpreted his anger as being “irrational”. That is, because it was not understandable to them, it was Mr C who was being irrational. The reason, of course, that they did not understand why Mr C was angry was based upon their belief that he had “memory loss”. He therefore could not have remembered anything about having had his job of landscaping taken away from him peremptorily, without including him in any discussion of that decision, and therefore could not be angry for “any good reason”. Hence, his anger was deemed “irrational”.

As well, a person living with dementia can demonstrate intact implicit memory of an interaction that has positive emotional content (unlike Mr C’s experience above), as can be seen in the following interaction involving a woman living with dementia who was a daily participant at an Adult Day Center (Sabat, 2018). Mrs L was diagnosed with AD four years prior to the following interaction. The lunch meal was ending and staff were clearing the tables of the participants’ lunch trays. The first author approached Mrs L, who was a lively person with an engaging personality, and asked her, “Mrs L, would you do me a favor?” She smiled broadly and replied, “Sure”. The first author said, “Would you go over to Mr A (another participant seated at one of the tables), pick up his tray and empty it in the trash can, please?” Mrs L said, “Of course” and proceeded to do so. Two days later, the first author returned to the Day Center. The lunch meal was nearing an end and the staff were clearing the tables. Mrs L approached the first author and said, without any prompting, “Do you have something for me to do?” The first author replied, “Like what?” Mrs L responded, “I don’t know.”

When the first author asked Mrs L if she would do him a favor, it was the first time he had ever made such a request of her. Therefore, when she asked him two days later if he had something for her to do, we have an example of her having made a long-term memory of his request, albeit demonstrated implicitly. She did not recall, did not have an explicit memory of, exactly what the first author had asked her to do. It was rare that Mrs L was asked by anyone to do him or her “a favor” at the Day Center so we may surmise that such a request made a positive memory in her in that she was able to help another person.

Another example (Sabat, 2017) of the intact implicit memory of a person living with dementia is that of Mrs D, a member of a support group for people living with AD. The first author co-led that group, wanted to meet with Mrs D to interview her, and spoke with her by phone to arrange to meet at the church where the support group met. The first author suggested meeting at the church the next day at 10:00 a.m., and Mrs D agreed. They continued to chat for a few minutes when Mrs D interrupted the conversation and asked, “So what time are we going to meet?” The first author replied, “How about 10:00 a.m. tomorrow at the church?” and Mrs D agreed. They continued to chat and Mrs D interrupted again by asking, “Don’t you think we should decide what time to meet?” and the first author replied, “How about 10:00 a.m. tomorrow at the church?” and Mrs D agreed. This pattern repeated itself another three times, with the first author answering each repetition of the question as if it were the first time Mrs D was asking. When Mrs D asked yet again, “So what time are we going to meet?”, the first author said, “Take a guess” and Mrs D replied, “Ten o’clock” and the first author said, “That’s correct.” Mrs D’s correct response was hardly a lucky guess because there were many other possible meeting times on the hour, the half hour, the quarter hour, and so on.

Mrs D did not have “memory loss” for at least two reasons. First, she made an explicit memory (recalled) that the two people were going to meet, as her question about what time to meet implied that she recalled that they had agreed to meet in the first place. In other words, she never said, “Why are we having this conversation?” Second, even though she did not have an explicit memory of, could not consciously recall, the exact details of the time agreed upon, she made a memory of that information that could be accessed none the less by asking her to, “Take a guess” about the agreed upon time, which is analogous to the situation in the word-stem completion task when the person with dementia is asked to complete the word stem with the first word that comes to mind. The analogy holds because in both cases, there cannot be one and only one correct answer, whereas in the case of tapping explicit memory (recall) of the meeting time, or the word that was on the list studied, there is in fact one and only one correct answer.

The incorrect assumption that people living with dementia have “memory loss” can easily lead to two forms of dysfunctional social treatment that were already alluded to, but unnamed, in the example of Mr C: Malignant Positioning and Malignant Social Psychology. We now turn to these two social situations that can have powerful effects on the lives of people with dementia as well as caregivers.