Memory training with the method of loci for children and adolescents with ADHD—A feasibility study

Abstract

The aim of this study was to investigate if training with the memory technique Method of Loci (MoL) is feasible for children and adolescents with ADHD. Twelve children (aged 9–17 years) with ADHD participated. Training with MoL was done using a mobile application, memorizing a sequence of 20–80 pictures, intended to be carried out five times per week for 4 weeks. Feasibility was assessed with pre- and post-intervention ratings, and with interviews after the training. Qualitative data were analyzed with content analysis. Those who trained with MoL performed better on memory test and reported fewer ADHD symptoms after completing the training, as compared to their baseline levels. All of these children would recommend the training to peers but the duration of training varied considerably. The participants and their parents reported that the MoL training was easy and fun to use, although lack of motivation, distractions in every-day life, and lack of routines created challenges. We conclude that training with MoL was considered feasible by most of the participants. Future research should try to make the intervention more acceptable by motivating the participants and limiting potential distractions and involving larger study groups and controls to study the efficacy of the training.

Keywords:

• ADHD
• memory training
• method of loci
• neurodevelopmental conditions

Introduction

Attention-deficit/hyperactivity disorder (ADHD) is a neurodevelopmental condition, usually diagnosed in childhood and often lasting into adulthood. ADHD is characterized by symptoms of inattention and/or hyperactivity/impulsivity, which are present in multiple settings (APA, 2013). ADHD affects ∼5–7% of children and adolescents (Polanczyk et al., 2014), who often display several functional impairments in such areas as relationships with friends and family and functioning in schools (Loe & Feldman, 2007; Nijmeijer et al., 2008). Further, ADHD is associated with poor academic performance, high rates of grade retention, and lower grade point average (Galera et al., 2009; Kent et al., 2011), hence decreasing their chances of successful social adjustment over time. While there is some evidence that ADHD symptoms may diminish with increasing age (Das et al., 2014), research suggests that the prevalence of ADHD in adult populations is still significant, around 2.5% (Song et al., 2021), with symptoms impacting negatively on well-being across a number of different domains.

A substantial degree of continuous psychosocial impairment associated with ADHD symptoms suggests a need for early interventions that are both acceptable and effective to minimize the potential negative outcomes. Even though pharmacological treatments for ADHD have proven to effectively reduce its core symptoms (Cortese, 2020), a significant proportion of affected children does not respond to currently accessible medications or experience unpleasant side effects (Cortese, 2020), and/or are cautious toward medication because of their concerns about the unknown long-term effects (Berger et al., 2008). In addition, to experience effect, the ADHD medications may require continuous use, while cognitive and behavioral interventions can teach children skills that will continue to benefit them as they grow up. As a result, the need for non-pharmacological interventions have also been highlighted (Coghill et al., 2021; Cortese, 2020; NICE, 2019).

Alterations in cognitive processes that control goal-oriented behavior, also referred to as executive functioning (EF), are hypothesized to underlie symptoms associated with ADHD (Barkley, 1997). Core cognitive processes that make up EF include inhibition, cognitive flexibility, and working memory (WM), but also planning, verbal fluency, and processing speed (Erostarbe-Perez et al., 2022). Deficits in WM have been associated with ADHD (Doyle, 2006; Kasper et al., 2012; Kofler et al., 2019; Willcutt et al., 2005). WM is the function of retaining and processing information and is a necessary mechanism for complex tasks, such as learning, comprehension, and reasoning (Baddeley, 2010). Hence, deficits in WM have been associated with reading disabilities (Swanson et al., 2009), lower mathematical skills (Peng et al., 2016), but also anxiety (Moran, 2016). WM is also of importance for social perception (Phillips et al., 2007), and accordingly, WM deficits have been linked to social problems in children with ADHD (Kofler et al., 2011). Distinguishing between working (central executive) and memory (storage/rehearsal) deficits have been suggested to be critical for treatment development, with a majority of children with ADHD having a deficit in the working component of WM, as opposed to the less impaired memory component (Rapport et al., 2013).

Interventions aiming to improve memory performance have been introduced in the literature as a potential intervention for ADHD (Gropper et al., 2014; Rapport et al., 2013). Such interventions are often characterized as either process-based memory training or strategy-based memory training (Schubert et al., 2014). Most previous studies targeting memory have focused on the process-based memory training, which targets basic processing capacities and is based on the assumption that the effect of practice and learning in one task may transfer to another task if there is a process overlapping between the tasks (Schubert et al., 2014). Thus, intensive training in WM should transfer improvements to other cognitive functions, such as attention, as a function of underlying overlapping neural networks (Klingberg, 2010). The putative effects of memory interventions, such as computer-based WM training have however been questioned in comprehensive meta-analyses. Previous interventions have not been able to show the evidence for transfer effects to other neuropsychological processes, such as inhibitory or attentional control (Cortese et al., 2015; Rapport et al., 2013), verbal ability, word decoding, arithmetic (Melby-Lervag & Hulme, 2013), and their results could not be generalized to important areas of everyday functioning, such as reading and academic achievement (Cortese et al., 2015; Melby-Lervag & Hulme, 2013; Rapport et al., 2013). Further, computer-based WM training interventions may be time-consuming, expensive, and often not incorporated in a more naturalistic, complex, and diverse environment (Moreau & Conway, 2014; Sonuga-Barke et al., 2014).

In contrast to process-based memory training, in strategy-based training studies the participants are taught to use a strategy and are encouraged to use and refine this specific strategy in practice, and as such improve the performance in the given task (Schubert et al., 2014). Strategy-based interventions have been suggested to serve as an effective tool for decreasing the cognitive load placed on the processing and encoding of information, and as opposed to the process-based interventions, it is easier to isolate and test their underlying mechanisms (Chan et al., 2019). Method of Loci (MoL) is a strategy-based memory training, stemming back to ancient Greece and Rome (Bower, 1970), used by public speakers to memorize the sequence of main topics in an entire speech. The information was translated into easily visualized things that they imagined placing at specific places (loci) along a familiar path. Whilst performing a speech, the speaker recollected the content by “walking” the same imaginary path. In modern days, the MoL technique is widely used by contestants at memory competitions and has also been useful in the educational settings (Twomey & Kroneisen, 2021). The effects of MoL training have been assessed in adult populations, giving support for MoL as a mnemonic device resulting in improving memory recall (Twomey & Kroneisen, 2021), durable and longer-lasting memories (Wagner et al., 2021) and self-reported benefited memory in everyday life (Sandberg et al., 2021). In clinical contexts, practice in MoL has been limited to elderly individuals, with a reported reduction of cognitive dysfunction (de Tournay-Jette et al., 2012; Saleh et al., 2015). To the best of our knowledge, no study has explored the feasibility and effectiveness of MoL training among individuals with ADHD.

Although MoL is commonly used for improving WM performance, the technique makes an active use of the items (locations) that are already stored in the long-term memory by pairing one or more to-be-remembered items with the familiar visuo-spatial landmarks already established (Li et al., 2021). As such, it has been argued that MoL expands on WM training, also relying on spatial processing and episodic memory (Wagner et al., 2021). The use of long-term memory structures when encoding has been referred to as the long-term working memory (Ericsson & Kintsch, 1995). It has been argued that MoL makes use of the key aspects that are thought to affect memory, including the spatial representations that help to organize a conceptual knowledge, schema-like knowledge structures that promote the transfer from WM into a long-term storage, and relatively bizarre associations between the items to be remembered and the loci in a personally relevant context, which triggers neural mechanisms related to novelty (Wagner et al., 2021).

Given the importance of memory in a daily life functioning and considering that WM problems are commonly reported in ADHD patients, it is important to evaluate the effects of interventions aimed at strengthening memory in this group. The aim of the present study was to investigate whether a strategy-based training with MoL may be feasible for young people with ADHD. Hypothetically, training with MoL could be particularly beneficial for this group of patients since it actively uses the schema-like knowledge structures already stored in the long-term memory (the loci), which is believed not to be impaired in ADHD, hence helping the patients to alleviate the deficits in the more problematic areas of functioning by using their strengths in other areas.

Materials and methods

Memory training

The intervention was based on the paradigm of MoL, as a strategy-based training technique. The participants were first asked to write down 40 landmarks (called hooks) where they could mentally place the items to remember as pictures, using an environment they were familiar with, e.g., their home. For the training, the participants were recommended to practice at a calm and quiet place, where they would not be disturbed. They were instructed to practice five times a week, 10–15 min every time, over a period of four weeks. The training was conducted using an application on a smartphone where pictures were presented, two pictures at a time. The pictures were of different objects (e.g., animals and food). Using their imagination, the participants then mentally placed two pictures on the first hook, and then moved on to the next two pictures, which should be placed on the second hook, and so forth. Starting with 20 pictures, participants could gradually, at their own pace, increase the number of pictures to memorize up to 80. After memorizing the pictures, the participants also practiced memorizing them backward to further consolidate the list to be remembered. The ability to remember both frontwards and backward is a common feature in MoL (Bower, 1970). The information about the number of training sessions was recorded by a research assistant from the application. For more information about MoL see Higbee (2001), and a more detailed description of the memory training used in our study is presented in Supplementary Appendix 1.

Procedure and participants

The study group was recruited via flyers in the waiting room at the child and adolescent psychiatric (CAP) outpatient unit in Uppsala, Sweden. For those who showed interest to participate in the study, more information was provided over the telephone, where also the patients’ eligibility for the study was evaluated. If proven eligible, a date for the first session was booked. The inclusion criterion was a diagnosis of ADHD (including inattentive or hyperactive type, as well as ADHD unspecified). The exclusion criteria involved having recently changed or planning to change the ADHD medication during the course of the study, severe depression, suicidality, psychosis, bipolar disorder, intellectual disability, organic brain injury, or current substance abuse. During the first meeting and before the intervention, the information about the child/adolescent’s clinical diagnoses, previous and concurrent treatments, demographic data, and parental diagnoses, as well as self- and parent ratings of ADHD symptoms and functional impairment were collected. In addition, a memory test was conducted. The child/adolescent and their parent/s were then introduced to the memory training and the application was downloaded on their mobile phone. The study was conducted between 2020 and 2022.

In total, 15 children and adolescents expressed their interest to participate, of which one did not provide an informed consent and was excluded. In addition, two individuals and their parents signed an informed consent form, but did not participate in any interviews and data collection and were also excluded. Hence, the final study sample consisted of 12 children and adolescents [Age range = 9–17; M = 13.75; SD = 2.18; 5 (41.7%) girls] with a clinical diagnosis of ADHD (combined type, n = 9; predominantly inattentive type, n = 3), established according to ICD-10 after a thorough evaluation by a neuropsychiatric team, including child psychiatrist and child psychologist, although not validated separately for this study. Of those, 5 (41.7%) individuals had a comorbid clinical diagnosis (one had separation anxiety disorder, one had mild depressive disorder, one had autism, and two had dyslexia); 9 (75.0%) individuals had an ongoing ADHD medication. Four (33.3%) participants had at least one parent with the similar problems. Seven (58.3%) participants had at least one parent with a university education. After 4 weeks of memory training, a follow-up meeting was conducted, and the participants were interviewed about their experiences and any potential changes in/adherence to their medication during the entire period, a memory test was conducted, and their ratings of symptoms and functional impairment were collected. Written informed consent was obtained from all participants and the study was approved by the Ethical Review Board of Uppsala University, Dnr. 2018/052, 2019-04528.

Measures

As a memory test, 10 randomly selected pictures of different objects to face down were put in front of a child who was instructed to open the first picture, memorize it, and then turn it upside down again before looking at the next picture. The child was instructed to remember as many pictures as possible (10 in total). After having looked at all pictures he/she was instructed to describe the pictures underneath to a research assistant. A total score (number of pictures remembered) was recorded. The memory test was unstandardized and had not been used previously.

The ADHD symptoms were assessed both pre- and post-intervention with self- and parent-ratings using the Adult ADHD Self-Report Scale for Adolescents (ASRS-A) (Sonnby et al., 2015). The questionnaire consists of 18 items corresponding to the diagnostic criteria for ADHD in the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV). Each item was rated using a Likert-scale ranging from never (0) to very often (4), with a total score ranging between 0 and 72, with higher scores indicating more symptoms. The ASRS-A has shown high internal consistency and concurrent validity in both general and clinical populations of adolescents 12–17 years of age (Sjölander et al., 2016; Sonnby et al., 2015). In the present study, the Cronbach alpha was 0.89 for the parent-rated and 0.91 for the child-rated scale pre-intervention, and 0.91 and 0.89, respectively, post-intervention.

Functional impairment was assessed pre- and post-intervention with the Child Sheehan Disability Scale (CSDS) (Sheehan et al., 1996; Soler et al., 2021), consisting of three items for adolescent self-assessment and five items for parental assessment. The self-rating scale assesses the impact of troubles and feelings in three areas (school, social activities, and home). The parental rating scale assesses the impact of the adolescents’ troubles and feelings in five areas (school, social activities, home, parents’ work, and parents’ social activities). The impairment in each area is measured on an 11-point scale from 0 (not at all) to 10 (very much), with higher scores indicating more impairment. The Swedish version of CSDS has, in a sample of adolescents seeking help for psychiatric problems, shown good internal consistency, with the items loading on one general component (Soler et al., 2021). In the present study, we used the total impairment score, including all areas of impairment within the scale, with a total score ranging between 0–30 for self-ratings and 0–50 for parent ratings. The Cronbach alpha was 0.83 for the parent-rated and 0.73 for the child-rated CSDS pre-intervention values, and 0.75 and 0.85, respectively, for the post-intervention ratings.

Potential benefits from the intervention were assessed using a self- and parent-rating questionnaire after the intervention. The participants and their parents were asked if the participants had improved in relation to memory, concentration, and control of problems associated with ADHD. The answers options were “I do not know,” “not true,” “partly true,” and “completely true.” The participants also rated their perceived benefit from the training, from no benefit (1) to very great benefit (5). In the self-rated version, two additional items were added inquiring whether the child/adolescent would recommend the training to their peers (yes/no) and if they perceived the number of training sessions as sufficient (too few, just enough, or too many). If a participant decided to end their participation, they were also asked about the reason for terminating the intervention.

Interviews

Semi-structured interviews were conducted after the intervention. The interview guide included open-ended questions. The participants (and parents) were asked: What was difficult with the training? What was good about the training? Was there something you would like to change with the training? Was there any more information you would have needed before the training?

Statistical analyses

To evaluate the feasibility of cognitive training with MoL for children and adolescents with ADHD both quantitative and qualitative data were used. Descriptive statistics are presented as numbers, means, and standard deviations. Changes in the pre- and post-measurements (memory test, ADHD symptoms, and functional impairment) were assessed with Wilcoxon signed rank test. The qualitative data were analyzed with an inductive method based on the process described by Elo and Kyngäs (Elo & Kyngas, 2008). The data from the first nine participants were analyzed first, followed by the data from the last three participants, to explore data saturation. The written comments were read by two researchers (SK and JI) separately, several times. Next, the contents were discussed by the two researchers in relation to the aims of the study. Then, the meaning units, defined as one or more sentences or just part of a sentence carrying a meaning related to the patients' and parents’ experience with the intervention, were identified from the written comments. Meaning units with a similar meaning or concerning a similar theme were grouped. The categories were then divided into subcategories based on the differences in meaning. This analysis was conducted by both researchers together and continued until all categories and subcategories were considered to be separate from each other. In the second step, to determine the data saturation, two researchers (EO and JI) read the comments from the last three participants separately and then discussed if the comments fit into the existing categories or if new categories were needed.

Results

MoL training and symptoms ratings

Of the 12 enrolled individuals, three participants did not start MoL training in the application at all, whereas one participant only did the training three times and did not provide any information on the rating scales. These four participants were considered as drop-outs. The rest of the participants (n = 8, 66.7%) trained with the mobile application between 8 and 32 times (M = 17.78; SD = 10.07). Those who dropped out did not differ from those who continued the training neither on the baseline measures of symptoms and functional impairment nor in sex and age. The main reasons for the drop out were lack of time and forgetting to do the training. Those who trained with MoL performed significantly better on the memory test (greater number of pictures remembered) after completing the training (Table 1). In addition, parents and children reported fewer ADHD symptoms after the training, while no differences were found in the ratings of functional impairment (see Table 1). All children who completed ASRS-A were 14 years of age or older. One child stopped taking the ADHD medication during the study period.

Table 1. Comparisons between pre- and post-measures of memory test, ADHD symptoms, and functional impairment using Wilcoxon signed rank test.

	Pre-training	Post-training	Statistics
Memory test^a	4.86 (3.67)	7.57 (3.51)	Z = 2.04; p = 0.041
ADHD symptoms
Parent rated^a	38.31 (12.20)	25.43 (9.98)	Z = 2.20; p = 0.028
Child rated^b	34.00 (14.24)	24.17 (12.11)	Z = 2.20; p = 0.028
Functional impairment
Parent rated^a	15.71 (10.98)	15.14 (10.62)	Z = 0.00; p = 1.000
Child rated^a	12.29 (5.31)	8.43 (7.44)	Z = 1.10; p = 0.271

ADHD symptoms was measured with the Adult ADHD Self-Report Scale for Adolescents, Functional impairment with the Child Sheehan Disability Scale.

^an = 7 raters.

^bn = 6 raters.

Ratings of acceptability

As shown in Table 2, of those who had trained with MoL, most children and parents reported true or partly true for improved memory, ability to concentrate and ability to handle problems related to ADHD. All of the children would recommend the training to their peers, and both participants and their parents indicated some perceived benefit from the training.

Table 2. Child and parent ratings of acceptability regarding memory training with the method of loci.

Questions	Response options
He/she can remember things better	Don’t know	Not true	Partly true	Completely true
Child rated	1	0	4	2
Parent rated	1	0	4	2
He/she can concentrate better	Don’t know	Not true	partly true	Completely true
Child rated	2	1	3	1
Parent rated	1	1	5	0
He/she can better control the problems associated with ADHD	Don’t know	Not true	Partly true	Completely true
Child rated	1	2	3	1
Parent rated	0	1	5	1
How much of a benefit do you think the treatment had for you/your child?^a
Child rated mean (SD)	3.57 (1.13)
Parent rated mean (SD)	3.86 (0.90)
Would recommend the treatment to a friend	Don’t know	Not true	True
Child rated		0	7
What do you think of the number of training sessions?	Too few	Too many	Just enough
Child rated	0	2	4

^aOn a scale from 1 (no benefit) to 5 (very great benefit).

Qualitative results

From the study population, all families were interviewed after the intervention. The qualitative analysis resulted in four categories and 14 subcategories, which are presented below with quotes and sentences that reflect each subcategory in Table 3.

Table 3. Categories and subcategories in the qualitative analyses.

Category	Subcategory
Challenges with the training	Remembering pictures and hooks
	Remember to do the training
	Distractions
	Lack of routines
	Lack of motivation
Benefits with the training	Better memory
	Fewer symptoms
	Academic benefits
Experience of the training	Usability and fun to use
Suggested improvements	Better pictures
	Number of pictures
	Purpose and usefulness of the technique
	Number of training sessions
	Have reminders

Challenges with the training

A common reported challenge was remembering pictures and hooks, including remembering the pictures in the correct order. One of the parents commented “she did not remember the order in which her hooks came. I needed to support her by asking leading questions, such as ‘what comes after it?’ and give her clues.” Others had a hard time to remember to do the training: “(it was hard) to remember it, to do it every day.”

In the subcategory distractions, it was mentioned that the children could be distracted due to external factors, such as “everyday events.” One of the parents commented “(that it was) difficult to get the (necessary) amount of training” and that “other things got in the way, such as driving test and math test in school.” One child commented that it was hard if “something happened (before training) or if I went to the computer first, then I got stuck there.”

Some participants mentioned a lack of routines which made it harder for them to train: “if the routine for training was broken, it was difficult to do it” and that it was “difficult to get a continuity with the training.”

Lack of motivation to do the training created challenges: “did not want to sit in the room. Said no, not now. Could not deal with it.” One of the participants commented, “If I don’t feel motivated, I can’t do anything.”

Benefits with the training

The majority of children commented that they have got a better memory from the training: “can remember more things more easily, if I do something at home, homework, tests. Something they said in the class was easier to remember,” “funny pictures, actually made me get a little better memory, easier to remember what I was going to do, fun things that were easy to remember,” and “I learned a lot, mainly memorizing pictures, remembering things better and remembering what to take with me.”

In the subcategory fewer symptom, one of the parents reported perceiving their child as being more calm and less prone to conflicts. One parent commented: “he is much calmer now than he was before. (He has) a test tomorrow, but is calm. When his dad study with him, he is calmer, talks more. Describes more in-depth about things.”

Some commented on the academic benefits that they experienced with the technique. One of the children commented:” yes, with concepts and stuff. As in science and physics and stuff, helps you with how to think. Trying to remember things in pictures and with the hooks,” and another that he “learned a new way to study.” One mother commented that “I have reminded her that she can use the technique when she has to memorize birds for school. I think the training helped and we will continue with this.”

Experience of the training

In the subcategory Usability and fun to use, several children reported enjoying the training and thought that it was fun and easy to use. One family commented that it was “quite simple, could do it myself, even if you did not want to, you could. Easy to maneuver,” and several mentioned that they would like to continue with the application.

Suggested improvements

A few children thought that there was a need for better pictures, that the pictures should be updated and be more fun: “the pictures were a bit boring.” Others commented that they liked the pictures.

One child reported that it would be good if one could choose a more specific number of pictures to practice on. One commented that it was “a relatively small assortment of images, it became difficult when an image from a previous round came up again.”

In the subcategory purpose and usefulness of the technique, some thought it would be easier for the children to train with MoL if the training became a part of the school education: “Imagine if you could do this during the school hours. That would have made it easier, I think” and one participant wished “that it could be integrated with school education.” One comment suggested it should be “clearer what the purpose of the training was.”

Number of training sessions was discussed and one comment suggested to “reduce the number of training sessions,” and that “three times per week would be more appropriate.”

Some commented on the need to have reminders to facilitate the training. One mentioned the need for an “alarm that reminded me (to do the training),” and another child said that “I wrote down (the hooks) on a paper, and had it with me. To have it written down is a tip to others.”

Discussion

This is the first study examining the feasibility of a cognitive training with MoL for children and adolescents with ADHD. Two-thirds of the study population have completed the training with a mobile application. Although all children indicated that they would recommend the training to their peers, ratings of the benefits from the training were mixed and the duration of training varied considerably. Those who participated in the intervention, performed better on the memory test after the intervention, when compared to their baseline level, indicating that the MoL training produced some positive results. Similarly, the rating scales indicated benefits from the training, with a decrease in both self- and parent-rated ADHD symptoms. In the interviews, the participants and their parents reported that the MoL training was easy and fun to use, although lack of motivation, distractions in every-day life, and forgetting to do the training created challenges.

The study suggests that a strategy-based memory training may represent a feasible alternative for young patients with ADHD, not only in terms of their performance on the memory tests but also in improving their ADHD symptoms. However, given the small sample size and the non-randomized design in a real-life setting, the findings need to be interpreted with caution and to be replicated in larger samples. The WM interventions have been suggested as a potential intervention for ADHD (Cortese et al., 2015; Gropper et al., 2014; Rapport et al., 2013), and especially process-based memory training, such as computer-based Cogmed Working Memory Training, have been studied, resulting in improvements in verbal and nonverbal working memory storage among children with ADHD (Chacko et al., 2014). Unfortunately, benefits regarding transfer effects were limited, with a lack of generalizable effects in other areas, such as academic achievements (Cortese et al., 2015; Melby-Lervag & Hulme, 2013; Rapport et al., 2013; Sonuga-Barke et al., 2014).

No previous study has investigated a strategy-based MoL training for children and adolescents with ADHD. Since ADHD has primarily been associated with deficits within the working component of WM, while memory, i.e., phonological and visuospatial storage/rehearsal processes, has been suggested to be minimally involved in ADHD symptoms and associated impairments (Rapport et al., 2013), the implementation of already stored items (loci) in long-term memory in the memory training may be particularly helpful to alleviate deficits in the more problematic areas of functioning. Further, including aspects, such as visuospatial processing, prior knowledge, and novelty may also be beneficial for this group. According to the long-term working memory hypothesis, new information is more rapidly stored in the long-term memory through an elaboration of long-term memory patterns and schemas and the use of retrieval structures (Ericsson & Kintsch, 1995), which is why the use of schema-like structures as in MoL could hypothetically promote a direct transfer from WM into long-term storage. As such, it is suggested that the use of long-term episodic memory patterns to meet the need for temporary storage during complex processing, may help to circumvent the usual limits of the WM in individuals with ADHD. Training with MoL may strengthen the interaction between working- and long-term memory, facilitating both the encoding and extraction of what to remember. According to the embedded process model, WM is not seen as a system separate from long-term memory, but rather as its activated subset that is in the current focus of attention (Schweppe & Rummer, 2014), and potentially, techniques, such as MoL can strengthen the ability to activate or bring into focus of attention the items stored in long-term memory.

A potential advantage of our study includes an active role taken by the participants and their parents, where the intervention was conducted at home and did not require any substantial resources. Strategy-based training has also been reported to produce a more far-transfer effect in other areas, such as a novel problem-solving task (Chan et al., 2019), indicating potential generalizable effects. It was mentioned in the training that the participants had difficulties to remember to do the training, that they were often distracted by other things, and that the need for establishing steady routines was emphasized. Both participants and their parents named that at times a lack of motivation was perceived as one of the shortages of the study, requiring an extra support/guidance from the parents and a need for more stimulating/exciting materials/tasks. Suggestions for changes included having reminders, such as alarms, using better pictures, having the opportunity to choose the number of pictures to practice on, highlighting the purpose and usefulness of the training (e.g., educational gains), and decreasing the number of training sessions.

The relatively large drop-out rates and the reported problems with motivation may indicate that the training could be challenging for many ADHD patients in child and adolescent psychiatric outpatient settings. It should be mentioned however, that of those participants who started the training, only one did not complete it. It also proved difficult to recruit the participants. Some of those who showed an interest decided to not participate when more information about the training was given, the reason being that it sounded too demanding. Dropout is not uncommon in the treatment studies but may represent a bigger problem in such populations, in particular, because of the difficulties associated with ADHD symptoms, e.g., with planning and other cognitive processes that control goal-oriented behavior (Barkley, 1997; Doyle, 2006; Willcutt et al., 2005). Impairments in the executive functions associated with the diagnosis can also make it challenging to train memory deficits, which was also noticeable in this study.

This study had several limitations that need to be addressed. This is a feasibility study, and we did not include a control group, and therefore, no definitive conclusions about either effect of the treatment or the group can be made. Rather, the aim of the study was to explore how the children and adolescents with ADHD would perceive the MoL training if it can be practically administered and be useful for patients with ADHD in a child and adolescent psychiatric setting. While the pictures for the memory test were randomly selected from a relatively large set of 80 pictures, it is possible that some of the improvement in memory could be related to the practice effect, which however was also a part of the MoL training itself. In order to investigate any quantitative effect of training with MoL on memory and symptoms, a randomized control trial would be an asset, preferably using an active control treatment. The results were evaluated based on the self- and parent-reported data, which could be vulnerable to bias. From the application on the smartphone, we only collected data on how many times they trained with the application, and more detailed information, for example, on the number of pictures memorized forward and backward could have provided additional knowledge. The researchers who informed about the MoL training also conducted the interviews which could also result in a bias; further, one of the researchers who performed the analyses also led the intervention and the previous knowledge of the participants could potentially influence the interpretation of the text. However, the use of two persons in the analysis, where one had not met the families, could help the researchers to stay close to the text and increase credibility (Graneheim & Lundman, 2004). It may also be so that the parents with a university education (2/3 of the participants) could more clearly perceive a need for intervention and possibly felt more motivated to continue with it. The study had relatively few participants and a larger sample could have led to additional information. However, we analyzed the data from the interviews in two steps to assess the data saturation, and the categories largely remained unchanged, although two subcategories were added.

Conclusions

The results of this cost-effective study based in a real-life clinical setting were somewhat mixed. The participants and their parents generally expressed their enthusiasm over the study and there was a significant improvement in the memory test and in the symptoms ratings. At the same time, even though the memory technique was considered acceptable and beneficial by most of the participants, the recruiting difficulties, high number of drop outs, lack of motivation and the every-day life distractions indicate that the training may be challenging for some ADHD patients in a child and adolescent psychiatric outpatient setting. However, a majority of those who started the study were able to complete it, the ratings showed promising results, and the future research should involve larger groups, and include a case-control design, to study the efficacy of the training. Also, the engagement from the parents and the memory training design may be of particular practical importance in this population.

Acknowledgement

We are grateful to all the families who participated in this study and to the child and adolescent psychiatric units for their contributions. We are also grateful to Sofia Lantz, Maria Mårtensson and Stefan Wallin for their valuable contribution to the study.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This research was financially supported by funds from the Uppsala University Hospital Research Fund (ALF) and Uppsala university hospital. The funders have no influence over the design, data collection, analysis, interpretation or in writing the manuscript of this study.