2.4.1, 2.4.2 - Memory support for dementia
Consider the contemporary key question: how can psychologists’ understanding of memory help people living with dementia? Apply memory models, different memory types and reconstructive processes to everyday support, then judge the evidence, practical constraints and social consequences.
Memory difficulties and support goals
Dementia can disrupt everyday remembering, but the pattern and severity of difficulties vary between individuals and conditions.
Dementia
A syndrome involving cognitive and behavioural difficulties that interfere with everyday functioning. Memory difficulty is common, but it is not the only possible difficulty.
A person may repeat a question, forget an appointment or lose the steps of a familiar activity. These behaviours need not have the same cause. Forgetting an appointment can involve forming a lasting memory or remembering to act later; losing track halfway through an instruction may involve temporary storage and attention. A useful support plan begins with the person’s actual difficulty, preferences and remaining abilities, rather than a generic assumption that all memory is absent.
The multi-store model distinguishes brief retention from long-term storage and highlights attention and rehearsal. The working memory model adds separate verbal and visuospatial resources and limited executive coordination. Different forms of LTM also matter: memory for a recent event may be weaker than familiar factual knowledge or a practised skill. H.M.’s skill-learning dissociation shows that impairment can be selective, though surgical amnesia is not the same condition as dementia. Sebastián and Hernández-Gil’s older-group comparison likewise cautions against treating a forward digit-span score as an index of every affected function.
For society, the aim is meaningful participation, independence and wellbeing alongside appropriate care. An intervention that helps someone find their appointment information may be useful without increasing their unaided digit span or reversing a disease process. Relevant outcomes include whether the person can accomplish a chosen activity, distress, quality of life and the demands on carers. A laboratory score is one possible measure, not the sole definition of help.
Using memory theory in daily life
Memory theory helps explain why changing an activity or its surroundings can reduce the demands on impaired processes.
Consider an original illustrative case: Amira enjoys preparing tea but loses track when several instructions are given at once, and forgets where appointment information is kept. Giving one step at a time reduces concurrent demands on verbal storage and executive coordination. A clear visual cue placed by the relevant object can make the next step available without keeping an entire spoken sequence active. It should be simple enough to understand and should not compete with confusing background material. This is an application of the working memory model, not proof that a particular component is damaged in Amira.
An external calendar, labelled storage place or agreed reminder can reduce reliance on unaided prospective memory, remembering to carry out an intention later. A familiar routine and repeated practice may draw on comparatively preserved skill knowledge, but effectiveness must be assessed rather than assumed. The long-term-memory distinction encourages a strengths-based approach: a person who struggles to recall yesterday’s conversation may still participate in familiar activities. It does not justify leaving a risky activity unsupervised solely because it was once familiar.
Reconstructive memory suggests that retrieval depends partly on cues, schemas and prior experience. Personally meaningful photographs, music or objects can cue conversation about past experience and support identity. They should reflect the person’s own history, language and preferences. Cues can also shape what someone reports, so a suggested recollection is not automatically an accurate record. Leading questions and pressure to produce a “correct” memory can create distress or introduce errors. Familiar material may also recall upsetting events; participation and responses should be respected.
Connect each support to its demand: simplify information to reduce working-memory load, provide cues to reduce unaided retrieval demands, and draw on preserved abilities without assuming they are universal.
Interventions and evidence
An attractive theoretical explanation needs outcome evidence before a service can confidently offer it as an effective intervention.
Cognitive stimulation uses varied activities and discussion, often in groups. Reminiscence uses past experiences as material for engagement. Cognitive rehabilitation works towards the person’s chosen functional goals through strategies and environmental adjustments. These differ from cognitive training, repeated practice on standard cognitive tasks. NICE NG97, recommendations 1.4.1–1.4.7, recommends tailored wellbeing activities and group cognitive stimulation for mild to moderate dementia, and advises considering reminiscence and rehabilitation or occupational therapy. It does not recommend cognitive training as treatment for mild to moderate Alzheimer’s disease. A lesson about memory mechanisms should not silently turn “practise memory tasks” into a promise of clinical benefit.
The GREAT randomised trial provides a relevant example beyond the prescribed studies. It allocated 475 people with early-stage Alzheimer’s, vascular or mixed dementia to goal-oriented rehabilitation plus usual care or usual care alone. Participants had an involved family carer. Rehabilitation involved ten weekly sessions and four later maintenance sessions; follow-up assessments were conducted by researchers blinded to allocation. Participant and carer ratings showed improved attainment of the targeted goals at three months, maintained at nine months. General cognitive and other secondary outcomes did not show significant group differences.
Random allocation makes a treatment explanation stronger than merely observing that motivated service users improve. Blinded assessors reduce expectation effects in measurement. However, participants and carers knew about therapy and supplied ratings, so expectations could affect reports. Goals were personalised and targeted in the intervention; improvement on them need not transfer to every activity. Participants had mild-to-moderate impairment and carer support, limiting generalisation to more severe dementia or people without similar support. The warranted conclusion is useful improvement in targeted functioning, not a cure or general restoration of memory capacity.
A person-centred social judgement
The key question concerns how society can use knowledge responsibly, including where support has costs and limits.
Services should choose goals with the person, make communication accessible and review outcomes. A culturally unfamiliar cue, an unreadable display or an app that someone cannot use may increase demands instead of helping. Training carers to simplify instructions can be inexpensive in equipment but takes time and consistency. Specialist rehabilitation and group programmes require staffing, transport and accessible spaces; unequal access may widen disadvantage. Outcome evidence should therefore be considered alongside feasibility and who is represented in the research.
Autonomy and safety can conflict. A reminder chosen by the person can support independence, while surveillance or compulsory routines can become social control if imposed without proportionate justification. Appropriate consent, attention to preferences, privacy of stored personal information and respect for changing capacity matter. Families and professionals need support too, but convenience for a service should not be treated as the person’s wellbeing. Research and communication about dementia are socially sensitive because claims can influence stigma, expectations and treatment of individuals.
A supported answer to the key question is that psychology can help identify the demand, use preserved strengths and design compensatory support, with evidence for some interventions and selected outcomes. Memory models provide useful explanations but do not replace assessment, clinical expertise or the person’s own priorities. The most defensible judgement specifies what improved, for whom, and under what conditions, rather than claiming understanding alone repairs dementia.