Multimodal Protocol for Assessing Metacognition and Self-Regulation in Adults with Learning Difficulties

Specific learning disorders (SLDs) encompass disorders of those who have difficulty learning and using academic skills, exhibiting performance below expectations for their chronological age in the areas of reading, writing, and/or mathematics^1,2. There are different estimations of prevalence rates depending on the age, language and culture analyzed but they are between 5% and 15%^1,3. Within the global category of neurodevelopmental disorders in the Diagnostic and Statistical Manual of Mental Disorders (5^th Ed.)¹, it is also necessary to focus on the incidence of Attention-Deficit/Hyperactivity Disorder (hereinafter ADHD) as it is a common disorder that has given rise to various controversies about how to approach it in recent years. Based on the DSM-5¹, it can be defined as a pattern of persistent behaviors of inattention and/or hyperactivity-impulsivity. Likewise, autism spectrum disorder (hereinafter ASD) is a category in the same manual that includes students who present neurodevelopmental disorders as a result of multifactorial dysfunctions of the central nervous system, which result in qualitative dysfunctions in three fundamental areas of the development of the person: social interaction, communication and interests and behaviors^1,2.

On these lines, a new concept has emerged moving away from the sense of deficit and offering a more positive approach to these disorders to be consistent with current ideas of neurodevelopmental difficulties as highly coexistent and overlapping⁴. From these new models, it is understood that the skills involved in high-level cognitive processes, which allow managing and regulating one's behavior in order to achieve a desired goal, are crucial for self-regulation and, therefore, for activities of daily living, including the academic ones⁵. In the context of adulthood, neurodiversity has evolved to include various types of difficulties, including ADHD and ASD, as well as dyslexia, dyspraxia, and/or dyscalculia. Accordingly, we are approaching this neurodiversity from a broad conception of learning difficulties (LDs). The increase in students with this diversity enrolled in postsecondary education is well documented and is due, in part, to the increase in high school graduation rates for students with disabilities⁶, but at the same time, there is less research about the learning process of these students than necessary⁷.

Each of the disorders approached in isolation involve different deficits and manifestations; however, some commonality can be found within that heterogeneity in terms of LD, such as metacognitive, self-regulatory, and emotional malfunctioning^8,9,10,11. Three fundamental foundations in the literature of learning in general, and LDs in particular, that represent the basis of successful learning and play an essential role in these well-known difficulties at the academic level¹². As well as this, other approaches understand that there could be a certain commonality between deficits in executive functions, such as problems in automatic processing or working memory, that occur in different disorders such as ADHD and reading disorders¹³ or ADHD and ASD⁵. However, there is still work to be done in this field, since not all studies reach the same conclusions about these points in common in relation to executive functions. It could be due to the variations presented by the samples from which the studies are based and the evaluation procedures of the executive functions used in the investigations^5,14.

In educational terms, this diverse mix affects not only the quality of learning, due to the fundamental nature of the affected functions, but also phenomena such as school dropout, change of degree, etc., with economic implications for governments and universities¹⁵. The dropout rate for students with LDs is higher than for students in the general population¹⁶ but also higher than the dropout rates for any other category of psychological disabilities except for those students with emotional disturbances¹⁷. In contrast, the number of students with LDs who are accessing post-compulsory education (vocational training, college, etc.) is increasing¹⁵, specifically in higher education^19,20,21,22. Moreover, one might well assume that there are many more students with LD than those who officially pass through student services and typically make up the prevalence statistics²³.

These difficulties are not always detected during childhood, especially in adults born before these disorders were considered in the regular academic system, and the symptoms of these disorders persist throughout people’s lives and cause difficulties in work, education and personal lives²⁴. Research has shown that although people might overcome some of their difficulties, most continue to exhibit struggles with learning during adulthood and their persistence is still problematic at those higher educational levels²⁵.

Paradoxically, unlike in previous educational levels and earlier ages, there are hardly any evidence-based instruments or evaluation protocols for adults with LDs. Despite the proliferation of diagnostic tools to evaluate LDs during childhood, the availability of valid, reliable instruments and methodologies for the adult population is significantly limited²⁴. A recent literature review about learning disabilities in higher education found that most of the information collected in this regard is done through interviews, and only occasionally are self-report questionnaires used²⁶. Self-report methodology and interviews, although valuable, are not enough to accurately assess metacognitive, self-regulation, and emotional skills processes, in fact, among others, because of the process nature. The importance of scales and interview methodology for measuring those processes is undeniable^27,28, but so too are the associated problems of validity²⁹ and incongruence with other innovative methods of assessment³⁰. An additional problem in the detection of LDs is the bias in the diagnosis of the disorder due to the absence of comprehensive assessment protocols. The fact that professionals do not have a reference protocol based on objective variables is frequently causing many false positive and false negative cases of LDs³¹.

In response to both scarcity of instruments for adults and the need to improve existing methodology, the current study proposes a multimodal evaluation protocol focused on metacognitive, self-regulation, and emotional processes, which make up the basis of the difficulties in adults with LDs. In line with the current literature, we propose a move toward integrative and multichannel measurement^32,33. The assessment is carried out through an analysis of the on-line learning process using several methods, techniques, and sensors (e.g., hypermedia learning environment, virtual reality, eye tracking, facial expressions of emotion, physiological responses, log files, screen recordings of human-machine interactions) and off-line methods (e.g., questionnaires, interviews, and self-report measures). This mixed methodology provides evidence of the deployment of target processes before, during, and after learning that can be triangulated to enhance the understanding of how students learn and where the problem lies, if there is one³⁴.

The evaluation protocol is carried out over two sessions. The sessions can be done in one sitting or may need partial applications depending on the person. The first is focused on the detection or confirmation of LDs and what specific kind of disorder we are facing, and the second is designed to go into the metacognitive, self-regulation, and emotional processes of each individual case in depth.

Session 1 is intended to be a diagnostic or confirmation assessment of the participant’s learning disabilities: SLD, ADHD and/or ASD (high functioning) to determine what type of specific problems the participants have. This assessment is essential for two reasons. 1) Adults with Learning Disabilities rarely have accurate information about their dysfunctional behavior. Some of them suspect that they have a LD but have never been evaluated. Others may have been assessed when they were children but do not have any reports or further information. 2) There may be discrepancies with previous diagnoses (e.g., a previous dyslexia diagnosis as opposed to a current diagnosis of attention deficit and slow processing speed; previous ASD diagnosis in contrast to current limited intellectual ability, etc.). The participant is interviewed, and questionnaires and standardized tests are applied. This session here is carried out by therapists with experience in diagnosing developmental and learning difficulties in the research and clinical context in different offices of a Spanish Psychology Faculty. The session begins with a structured interview that collects biographical information along with the presence of symptoms related to SLDs that are referred to in the DSM-5¹. Following that, the reference intellectual ability test WAIS-IV³⁵ is used in case of exclusion criterion implementation and because it provides very valuable information for learning difficulties from the scales “work memory” and “processing speed”³⁶. Additionally, the PROLEC SE-Revised Test³⁷ is extensively used to evaluate reading disabilities (lexical, semantic and/or syntactic processes of reading), one of the most prevalent and disabling difficulties for learning in current academic contexts, which overlaps with other disorders such as ADHD³⁸. This evaluation collects reading accuracy, speed and fluency along with reading disabilities, and more importantly, in which reading process the failure occurs³⁷ (this test has been evaluated with pre-university students. Currently, there are no tests in Spain that are adapted to the general adult population, so this test was selected because it is the closest to the target population). Then, we screen symptoms of ADHD through the World Health Organization Adult ADHD Self-Report Scale (ASRS)³⁹ and refine the evaluation of this disorder, introducing multimodality with a cutting-edge virtual reality continuous performance test for the evaluation of attentional processes and working memory in adults, the Nesplora Aquarium^31,40. This test is a very useful tool when diagnosing ADHD in adults and adolescents over 16 years old in an ecological scenario, providing objective, reliable data. It evaluates selective and sustained attention, impulsivity, reaction time, auditory and visual attention, perseverance, quality of attentional focus, motor activity, work memory and cost of change of task. Additionally, along with the WAIS-IV³⁵ as a whole for collecting information about the participant’s intellectual ability, we pay special attention to the scales “work memory” and “processing speed” because they are related to learning difficulties and the results of these scales are used in the final decision. Finally, we include the Autism Spectrum Quotient (AQ-Short)⁴¹ in the protocol, the short version of the reliable AQ-Adult from Baron-Cohen, Wheelwright, Skinner, Martin and Clubley⁴².

Session 2 focuses on a multimodal assessment of the participant’s learning process. The key to understanding complex learning lies in understanding the deployment of students’ cognitive, metacognitive, motivational, and affective processes⁴³. To that end, participants work with MetaTutor, where the use of metacognitive and cognitive strategies deployed are observed while they are learning. MetaTutor is a hypermedia learning environment that is designed to detect, model, trace, and foster students’ self-regulated learning while learning different science topic⁴⁴. The design of MetaTutor is based on extensive research by Azevedo and colleagues^43,45,46,47 and belongs to a new trend in the measurement of SRL, the so called third wave, which is characterized by combined use of measurement and advanced learning technologies³³. The use of MetaTutor also provides multimodal trace data, incorporating measures such as, eye tracking, emotional physiological responses (galvanic skin response (GSR) and facial expressions of emotions)⁴⁸, log-data and questionnaires. All these measures are combined to reach a deeper understanding of the participants SRL and metacognition.

Eye tracking provides an understanding of what attracts immediate attention, which target elements are ignored, in which order elements are noticed, or how elements compare to others; electrodermal activity lets us know how emotional arousal changes in response to the environment; facial-emotion-recognition allows the automatic recognition and analysis of facial expressions; and data logging collects and stores the student´s interaction with the learning environment for further analysis. Concerning the questionnaires, the Mini International Personality Item Pool⁴⁹ informs about a range of activities and thoughts that people experience in everyday life assessing each of the five major personality traits (extraversion, agreeableness, conscientiousness, neuroticism and openness). The Connotative Aspects of Epistemological Beliefs⁵⁰ provides information about participants’ beliefs about knowledge. The Rosenberg Self-esteem scale shows how the participants feel about themselves overall⁵¹. The Emotion Regulation Questionnaire⁵² provides information about participants’ emotion regulation. The Achievement Emotions Questionnaire (AEQ)⁵³ informs about emotions typically experienced at university.

In short, assessing LDs during adulthood is particularly difficult. Education and experience allow many adults to compensate for their deficits and later show undifferentiated or masked symptoms, on which scientific knowledge is still scarce. Taking into account the critical research gap that arises, this current work aims to ensure theoretically-driven, empirically-based guidelines for accurate assessment of LDs during adulthood in order to design effective prevention and intervention actions.

To help readers decide whether the method described is appropriate or not, it is necessary to specify that the protocol is not suitable for people with intellectual disabilities because their diagnosis invalidates the diagnosis of learning difficulties. In addition, due to the singularities of the equipment used and the format of showing the learning content, it is still not possible to evaluate people with motor disabilities (upper limbs, neck and/or face), hearing or visual impairment. Nor would it be suitable for participants with severe psychiatric disorders. It would require the use of drugs that could alter information processing or the physiological expression of emotions.