This study is a comparison of 3 learning techniques, Errorless learning, modelling and trial and error, in the relearning of IADL of Alzheimer patients from mild to moderately severe dementia.

Tailored IADL will be chosen for each patient (n=300) and trained in individualized sessions for 6 weeks.

This study focuses on the relationship between learning techniques, IADL and memory processes, in a threefold way:

1. it will determine which of the of the three learning techniques (EL, MR, TE) will improve most the (re)learning of instrumental skills in different dementia stages using a randomized controlled trial;
2. it will explain the role of implicit and explicit memory mechanisms in the (re)learning of IADL tasks; and
3. as a secondary objective, it will explore the possible drug treatment by behavioral intervention interaction effects of the three learning techniques.

Scientific background and rationale Alzheimer dementia (AD) is the most common cause of progressive cognitive deterioration that alters memory and learning to such a degree that it heavily interferes with daily living. Functional autonomy loss is a key feature of AD, as it follows a slow degradation process in cognitive function and in the ability to perform instrumental activities of daily living (IADL), such as managing finance, food preparation or using a dish washer. Normally, learning occurs in an unstructured manner, which consists of guessing and the occurrence of errors during acquisition (Trial and Error, TE). However, there is abundant evidence that reducing errors during learning (Errorless Learning, EL) or increasing the time period between recall attempts (Modeling with Spaced Retrieval, MR) allow even moderate and severe Alzheimer Dementia (AD) patients to (re)learn instrumental activities of daily living (IADL) such as using a new route, an agenda or a cassette/radio player. While these findings are encouraging, we still do not fully understand the memory mechanisms underlying different learning techniques that are crucial in improving IADL tasks (re)learning and remembering in AD patients. Because acetyl cholinesterase inhibitors or memantine may be active moderators of intervention targeting memory improvement, complex intervention using behavioral enrichment training should explore any drug treatment by behavioral intervention interaction effects.

Description of the project methodology

This study focuses on the relationship between learning techniques, IADL and memory processes, in a threefold way:

1. it will determine which of the of the three learning techniques (EL, SR, TE) will improve most the (re)learning of instrumental skills in different dementia stages using a randomized controlled trial;

2. it will explain the role of implicit and explicit memory mechanisms in the (re)learning of IADL tasks.; and

3. it will explore the possible drug treatment by behavioral intervention interaction effects of the three learning techniques.

   • Primary outcome :

Is the (re)learning effectiveness (physical performance) of each procedure and the overall maintain of the autonomy. All intervention are individualized training. The learning procedure comparison will allow us to assess the (re) learning capacities of IADL tasks in AD patients according to severity stages from mild to moderately severe.

• Secondary outcome : Assessing the role of the implicit memory processes over the explicit memory processes in the (re)learning of IADL tasks. Overall effects of the intervention over the patient's autonomy, cognitive functioning, behavioral disturbances, quality of life and careers burden.

Settings: Nursing Homes, Day Care centers, Memory Clinics Locations: Nijmegen, The Netherlands and Nice and Bordeaux agglomerations, France 3 different interventions given in individual sessions at participant's facility.

Each intervention is a 6-week training with a post assessment at 4-week follow-up. Each participant will receive one of the 3 interventions for 2 hours twice a week in individual sessions.

Errorless learning(EL) refers to the use of feedforward instruction (i.e., how to do) before actions to prevent learners from making mistakes.

Modeling with Spaced Retrieval (MR) techniques refers to the modeling of the steps and the increasing time interval between the completion of the task and the rehearsal of the targeted information by the patient.

Trial and Error (TE) refers to the regular unstructured learning and is considered as control condition.

A standardized 1-week training has been developed in French and Dutch to train French and Dutch therapists at each learning techniques.

Inclusion criteria:

• Having a diagnosis of mild to moderately severe Alzheimer Dementia type with a MMSE score between 10 and 26;
• Fulfill the DSM-IV-TR and NINCDS-ADRDA criteria for Alzheimer's dementia type (33;34);
• Aged 60 and older;
• Not able to complete without cue the proposed tasks during the screening interview.
• Having a Social Security System

Exclusion criteria:

• MMSE < 10 or > 26
• Participants with severe deficits in alertness,
• Deemed behavioral disturbances (e.g., such as high NPI irritability symptom as defined with a score of 6 or above out of a maximum score of 12),
• Known medications that could interfere with the intervention (except AD medication, cf AD treatments).