Discuss one disadvantage of developing a classification system for mental disorders

2 Diagnosis

Diagnostic classification systems have been constructed to help clinicians make diagnoses. The most commonly used classification system in the United States is the American Psychiatric Association's Diagnostic and Statistical Manual of Mental Disorders, 4th edition (1994, generally referred to as DSM-IV). This classification system contains specific and explicit criteria for making diagnoses (see Mental and Behavioral Disorders, Diagnosis and Classification of).

Clinicians' diagnoses are reliable and moderately valid, but only when they attend to diagnostic criteria. Unfortunately, there is evidence that a significant number of clinicians do not adhere to criteria when making diagnoses. That is, many clinicians may think that they are making diagnoses according to the DSM-IV criteria, but they do not refer to the criteria when making a diagnosis, and examination of their diagnoses reveals that they are not made in accordance with the DSM-IV criteria. This can lead to different types of problems including race bias, gender bias, age bias, and the underdiagnosis or overdiagnosis of some mental disorders. These problems are described below.

The most widely replicated finding for race bias involves the differential diagnosis of schizophrenia and psychotic affective disorders. African-Americans and Puerto Rican Hispanics with bipolar affective disorder (formerly called manic depression) are more likely than Whites with bipolar affective disorder to be misdiagnosed as having schizophrenia. For this reason, Black patients and Puerto Rican Hispanic patients are more likely than White patients to be overmedicated with neuroleptic medications, and their depressive symptoms are more likely to be untreated.

The most widely replicated finding for gender bias involves the differential diagnosis of histrionic personality disorder and antisocial personality disorder. When different groups of mental health professionals have been given identical case histories except for the designation of gender, clinicians have been more likely to diagnose women as having a histrionic personality disorder and men as having an antisocial personality disorder. Histrionic personality disorder is characterized by overly dramatic, attention seeking behaviors (e.g., uncomfortable when not the center of attention), and antisocial personality disorder is characterized by antisocial behaviors (e.g., habitual lying, having no regard for others, showing no remorse after hurting others) (see Personality Disorders).

The most widely replicated finding for age bias involves the differential diagnosis of organic impairment and depressive disorder. Compared to young and middle-aged patients, elderly patients are more likely to be diagnosed as having organic impairment and they are less likely to be diagnosed as having a depressive disorder, even when all of the clients are described by the same case history except for the designation of age. Of course, someone diagnosed as having organic impairment will be less likely to receive psychotherapy and antidepressant medicine.

It should be noted that even when clinicians attend to diagnostic criteria and apply them the same way for different groups of patients (e.g., for African-American and White patients), diagnoses can be biased (Widiger 1998). For example, diagnoses can be biased because diagnostic criteria, not the cognitive processes of clinicians, are biased. Diagnostic criteria are said to be biased if they are more valid for one group than for another (e.g., if diagnostic criteria for a particular disorder are more valid for males than for females). In general, little is known about whether diagnostic criteria are biased.

Research has also described other types of errors. Mental health professionals disagree strongly over whether dissociative identity disorder (formerly called multiple personality disorder) is overdiagnosed or underdiagnosed. There is also a controversy over whether attention-deficit/hyperactivity disorder (ADHD) is overdiagnosed. Diagnoses of ADHD have doubled in frequency in recent years (see Attention-deficit/Hyperactivity Disorder (ADHD)), while diagnoses of dissociative identity disorder have increased 10-fold. Finally, research suggests that clinicians underdiagnose mental disorders in the mentally retarded, they also underdiagnose mental disorders (e.g., major depression) in terminally ill patients, they frequently underdiagnose personality disorders, they underdiagnose substance abuse in psychiatric patients, and they underdiagnose mental disorders in individuals admitted to substance abuse treatment programs.

ORGANIZATIONAL PLAN

Available DCSs focus primarily on the categorizations of symptoms rather than on children's functioning. To address the need to describe child and adolescent functioning with a common nomenclature, the International Classification of Functioning, Disability, and Health (ICF) was developed for clinical practice, research, and policy development across disciplines and service systems.28 Key dimensions of this system include (1) impairments in body functions and in structured activities; (2) activity limitations; and (3) participation, defined as involvement in a life situation. In addition, the system describes environmental factors (e.g., the physical, social, and attitudinal settings in which individuals conduct their lives) and personal factors that affect functioning. A version of the ICF for children and youths (ICF-CY) has been developed.29 The ICF-CY includes more than 100 functional impairments and relevant codes that are applicable to DSM-IV, DSM-PC diagnostic categories, and the DC:03R. An example of the relevant dimensions and codes that are applicable to one disorder (ADHD) is shown in Table 6-6.

5.21.2.2 Diagnostic Features

Formal diagnostic classification systems for substance use disorders have primarily evolved since the mid-1970s and have been developed exclusively for adults, with no distinction made between adolescent and adult users. The first criteria-based classifications appeared in the third edition of the Diagnostic and statistical manual of mental disorders (DSM-III; American Psychiatric Association, 1980), which identified two categories of substance use disorders: substance abuse and substance dependence. Substance abuse was defined as pathological use associated with impairment of social or occupational functioning with a one-month duration, and dependence as the presence of tolerance or withdrawal and either pathological use or impairment in social or occupational functioning due to such use. Prior to the third edition, the DSM viewed substance use problems such as alcoholism and drug dependence as subsets of sociopathic personality disturbance. The designation of a separate category for substance use disorders in DSM-III differentiated substance use from antisocial behavior more generally and recognized the importance of research on the nature of the disorders (Nathan, 1991).

A number of serious limitations of DSM-III were noted, however, including the insensitivity of the criteria in addressing heterogeneity of problems, the lack of theoretical and empirical basis for some criteria, and inconsistencies across criteria in various substance categories (cf. Nathan, 1991; Rounsaville, 1987). In addition, the reliance on tolerance or withdrawal for the dependence diagnosis determined that, in most cases, adolescents could only be diagnosed as abusers, since they rarely exhibit many of the tolerance or withdrawal symptoms for alcohol or other drugs (Hughes et al., 1992; Vingilis & Smart, 1981).

The DSM-III-R (American Psychiatric Association, 1987) was designed to correspond more closely to the International classification of disease, ninth edition (ICD-9), which had effected a conceptual shift in classifying substance use disorders. In keeping with the conceptual shift, the definition of dependence was expanded to include clinically significant behaviors, cognitions, and symptoms indicative of use and no single symptom was required for a diagnosis of dependence. Marked tolerance and withdrawal symptoms remained in the list of criteria, but other criteria—such as spending large amounts of time on activities to acquire substances, symptoms of intoxication when expected to fulfill major role obligations at work, school, or home, and giving up important activities because of substance use—made it easier for adolescents to be diagnosed with dependence. The changes also allowed for more variability in patterns of use.

By default, abuse was diagnosed when sufficient criteria for dependence were not met. Thus, abuse was conceptualized as a milder or predependence stage in the cluster of disorders. However, some research suggests that abuse and dependence disorders have distinct courses (Hasin, Grant, & Endicott, 1990), questioning the notion of a presumed continuum of severity.

The revision of the system for DSM-IV (American Psychiatric Association, 1994) was intended to reflect empirical research more closely and to increase compatibility with the ICD-10 system. Accordingly, three essential changes were made to the classification of dependence: the criteria regarding the inability to fulfill major role obligations and the criteria for duration were dropped, and subtypes of tolerance and withdrawal symptoms were added. However, adolescents with substance use problems commonly experience difficulty in meeting role obligations. With regard to educational expectations, truancy, attending classes while under the influence of alcohol or drugs, school failure, and premature dropout are common. Similarly, failures to meet role obligations are frequently observed in family contexts. Thus, the elimination of the role obligation criterion will probably reduce the number of adolescents who are diagnosed as dependent.

The criteria for substance abuse in DSM-IV were expanded, requiring the presence of clinically significant impairment or distress, including the failure to fulfill major role obligations and recurrent substance-related legal problems, as part of the maladaptive pattern of substance use. The addition of the role obligation criteria to this category could facilitate diagnostic classification. Nevertheless, adolescents who show a maladaptive pattern of use must now also exhibit at least minimal impairment or distress. This moves classification of abuse away from a topographical definition of substance use, such as quantity, frequency, or patterning of use, toward a definition based on dysfunction associated with use, and essentially raises the bar for adolescents to meet criteria, since adolescent behaviors tend to be more detectable than distress resulting from the behaviors.

The most disturbing aspect of the evolution of classification systems for substance use is the complete lack of evidence either to support or reject the applicability of the systems with adolescents. To illustrate, consider the nature of the prevailing diagnostic systems. The DSM and ICD systems are categorical rather than dimensional, classifying substance abuse on the basis of clustering of symptoms as opposed to viewing symptoms as occurring along a continuum. Within these systems, then, it is possible for adolescents to exhibit high quantity and frequency of use, yet not meet criteria for a substance abuse disorder if they manage to avoid negative consequences or other signs of impairment considered part of the discrete cluster of symptoms that currently define abuse. The confirmation of the reliability and validity of the current approach for adolescents or, alternatively, the formulation of an age-specific substance abuse definition and diagnostic criteria appropriate for adolescents could play a key role in the development and evaluation of prevention and treatment programs.

III Current Dilemmas in Psychiatric Diagnosis

Currently used diagnostic classification systems for psychiatric disorder, such as the Diagnostic and Statistical Manual of Mental Disorders IV (DSM-IV; Schaffer, 1996) and the International Classification of Disease 10 (ICD-10; http://www.who.int/classifications/icd/en/), are known to have a certain degree of reliability. Therefore, a patient presenting with the same symptoms in one hospital is likely to be given the same diagnosis in another hospital if the same classification system is used. However, there is an apparent increase in the prevalence of schizophrenia when ICD-10 criteria are used for diagnosis, compared with the use of DSM-IV (Cheniaux et al., 2009). This may be because DSM-IV and ICD-10 have developed to include a modern compendium of mental disorders that can be reliably diagnosed based on signs and symptoms, but have not been validated (Spitzer et al., 1980; Pierre, 2008; Keller et al., 2011). It is not likely that specific symptoms are linked to a defined natural disease entity. It is well known that patients with neurological, traumatic, infectious, and metabolic disorders can present with symptoms indistinguishable to symptoms of schizophrenia (Yolken et al., 2009; Lovatt et al., 2010; Scaglia, 2010). In addition, some subjects are known to have feigned symptoms of schizophrenia and other mental disorders (Bagby et al., 1997) for reasons such as gaining access to disability payments, social housing, and other benefits.

Most psychiatrists agree that the current construct of schizophrenia is an umbrella term for a complex chimera of etiologies that happens to present with similar symptoms, in the same way that most acute infectious disorders present with fever (Tsuang, 1975). Misdiagnosis is thus a common occurrence in psychiatric practice. For example, Gonzalez-Pinto et al. (1998) found that 31% of bipolar patients were diagnosed with schizophrenia. Follette and Houts (1996) went further in their criticism challenging the fundamental assumptions or theoretical underpinnings of current classifications systems. They pointed out that there is no method to validate current diagnostic concepts with externally validated measures which are independent of the concept itself.

A further potential factor for misdiagnosis and inconsistency is that clinicians do not usually use classification systems to establish a psychiatric diagnosis. Instead, they mostly apply heuristic unstructured interviews. This means their diagnosis may be based on experience and personal views, instead of matching the guidelines or criteria of the diagnostic system. This can be associated with systematic errors in judgment based on misconception and experience, which may rely on selective memory. There has also been a failure to address the problem of false positives in diagnoses of mental disorders (Wakefield, 2010). A study published by Strakowski et al. (2003) investigated the influence of ethnicity on patient diagnosis. This study found that clinicians tend to overdiagnose schizophrenia in African Americans. The bias was removed when examiners were provided with ethnicity-blinded transcripts of otherwise identical patient interviews.

Abstract

Until now, hypersexuality has not found entry into the common diagnostic classification systems. However it is a frequently discussed phenomenon consisting of excessive sexual appetite that is maladaptive for the individual. Initial studies investigated the neurobiological underpinnings of hypersexuality, but current literature is still insufficient to draw unequivocal conclusions. In the present review, we summarize and discuss findings from various perspectives: neuroimaging and lesion studies, studies on other neurological disorders that are sometimes accompanied by hypersexuality, neuropharmacological evidence, genetic as well as animal studies. Taken together, the evidence seems to imply that alterations in the frontal lobe, amygdala, hippocampus, hypothalamus, septum, and brain regions that process reward play a prominent role in the emergence of hypersexuality. Genetic studies and neuropharmacological treatment approaches point at an involvement of the dopaminergic system.

Diagnostic and Statistical Manual (DSM-IV TR) Axis I Disorders

The DSM-IV is well established and recognized as a diagnostic classification system in clinical settings for the purpose of diagnosing behavioral disorders in children. It is divided into five axes for the assessment of multiple domains of information, known as a multi-axial diagnosis. The DSM classification system is challenging for this population because it only offers a small number of child psychiatric disorder categories and lacks developmentally sensitive adaptations. The system generally does not incorporate constructs and criteria that characterize disorder in younger children as viewed and described by intervening professionals outside of the mental health field. It also lacks an integrated emphasis on contextual factors influencing developmental psychopathology (attachment, relationships, behavior). Note: Other diagnoses which are not specific to children but may diagnostically apply to children include anxiety, mood disorders, eating disorders, somatoform disorders, and substance use disorders (DSM IV, 2000; Wolraich et al., 2008).

Another widely recognized and accepted classification of mental and behavioral disorders used to make pediatric diagnoses is the International Classification of Diseases, 10th Edition (ICD-10). The ICD-10 is divided into ten categories of pediatric behavioral and emotional disorders.

2.4 Assessment Instruments and Diagnostic Classification

Population studies and methods-related epidemiological work have been instrumental in the improvement of diagnostic classification systems for mental disorders. Reliable symptom and diagnostic assessment instruments of mental disorders have been created for use in epidemiology and clinical research. This work has not only significantly influenced the content and structure of clinical instruments: Structured Clinical Interview for DSM-IV (SCID) (First et al. 1997); Schedules for Clinical Assessment in Neuropsychiatry (SCAN) (Brugha et al. 1999b) and non-clinical tools: Composite International Diagnostic Instrument (CIDI) (WHO 1990), but also played an important role in the revision processes of diagnostic classification systems (DSM-IV and ICD-10).

Yet these conceptual models of mental disorders are not, and have never been, a paragon of elegance, nor have they resulted in sufficiently neat and crisp classification systems that match basic research findings, and clinical management and decision-making. The introduction of these operationalized and descriptive manuals have resulted in greater diagnostic reliability and consistency in the use of diagnostic terms around the world. In particular, they have been a key prerequisite for epidemiological progress. However, major problems (i.e., thresholds, overlap, and comorbidity), which remain a source of significant dissatisfaction and controversy, will require extensive future work.

At the center of this agenda is the need for convincing clinical and nosological validation in terms of prognostic value and stability, family and genetic findings, and laboratory findings for almost all mental disorders, allowing a sharper genotypical and phenotypical classification. Current diagnostic classification manuals (DSM-IV and ICD-10) deliberately do not contain mutually exclusive diagnostic categories in order to simulate research inquiries into diagnostic boundaries and thresholds—a valuable target for epidemiological research. Consensus is lacking on how to tailor appropriate psychopathological assessment instruments that are able to address such threshold issues appropriately. Further, despite the substantial scientific exploration and examination that went into instruments like the CIDI (WHO 1990) and the SCAN (Brugha et al. 1999a), some basic problems of reliability and validity inherent in the assessment of some mental disorder are yet unresolved.

At the center of discussion is no longer the traditional question of whether to go for categorical or dimensional, but rather to what degree and for which psychological conditions ‘clinical judgment and probing’ should be regarded as a mandatory core element. Empirical evidence needs to be gathered to determine in which diagnostic domains and clinical instruments are superior to fully standardized instruments, such as the CIDI, which try to identify explicitly the latent variables behind the vagueness of clinical judgment. Progress in the resolution of this issue will also offer ultimately more appropriate strategies in resolving the ‘gold standard’ question of the optimal strategy for validating epidemiological instruments (Brugha et al. 1999a; Wittchen et al. 1999c) (see Nosology in Psychiatry; Differential Diagnosis in Psychiatry)

6.01.2.6.2 Controversy in Diagnosis: Dimensional Versus Diagnostic Approaches

Although the primary purpose of a diagnostic interview is to establish a diagnosis, there is considerable controversy surrounding current diagnostic classification systems because of their categorical nature (Haslam, 2003; Helzer et al., 2006). A categorical approach to mental disorders relies on a list of diagnostic criteria to determine the presence or absence of clinical symptoms and disorders (Kraemer et al., 2004). However, there is debate about whether mental disorders can truly be defined into distinct diagnostic categories. The DSM-5 mentions several limitations of a categorical approach to mental disorders including difficulty delineating one disorder from another, high rates of comorbidity, frequency of disorders not meeting full criteria resulting in other specified and unspecified diagnoses, and lack of treatment specificity (APA, 2013). Instead, a growing body of literature states that it may be more helpful to conceptualize mental disorders as dimensional constructs (e.g., Cuthbert, 2014; Krueger and Bezdjian, 2009). A dimensional approach to mental disorders would place clinical symptoms on a continuum of severity and/or frequency as opposed to being fully present or absent. The dimensional approach does not indicate a concrete threshold between “normality” and a disorder. Instead, this approach recognizes that many symptoms of psychopathology are experienced on a continuum that can change over time. The DSM-5 began to introduce an integration of the dimensional approach to diagnosis within the current categorical approach (APA, 2013). For example, some diagnostic categories like substance use disorders have introduced dimensional severity ratings. Future revisions of the diagnostic classification systems may include more dimensional approaches to mental disorders although this is widely debated (e.g., Wakefield, 2016).

There is also debate surrounding the necessity and usefulness of determining a mental disorder diagnosis. For example, a group of researchers conducted a study investigating the heterogeneity in psychiatric diagnostic classification and concluded that diagnostic labels are scientifically meaningless given the high degree of overlap between disorders (Allsopp et al., 2019). They also argued that diagnostic labels do not tell us much about the individual patient or the treatment that would be most meaningful to them (Allsopp et al., 2019). Mental disorder diagnosis remains controversial despite advances that have been made in the field. For example, there unfortunately continues to be stigma associated with receiving a mental disorder diagnosis (Rose and Thornicroft, 2010). Also, the diagnostic label does not necessarily explain the etiology of the disorder. Despite these shortcomings with the current way of understanding and communicating mental disorders, many advances have been made within the current categorical model of mental disorders that have increased our understanding of mental disorders and their treatment.

4.6 Clinical implications

Our findings point to the issue that the distinction between physical and mental conditions is not as clear as suggested by diagnostic classification systems, since many conditions share both physical and psychological aspects. Generally, three different relationships between chronic physical diseases and mental conditions are possible (Turner and Kelly, 2000): (1) Chronic physical diseases can lead to the manifestation of mental conditions, often depression or anxiety. (2) In individuals with pre-existing mental conditions, the development of a chronic physical disease can aggravate their symptoms. (3) If physical symptoms in individuals with chronic diseases worsen or new ones develop, this can constitute an expression of emotional distress (Turner and Kelly, 2000). In the clinical care of chronic tinnitus patients, these possible connections between tinnitus and mental health need to be addressed.

Furthermore, recent literature has begun to address the limitations of traditional diagnostic classification systems for mental disorders which classify psychopathology in distinct categories that are not based on evidence (Hofmann, 2014; Kotov et al., 2017). New approaches include empirically-based frameworks such as structural approaches using dimensional classification (Kotov et al., 2017), theory-based cognitive behavior classifications (Hofmann, 2014), or network approaches (Fried et al., 2017).

Dimensional classification approaches are based on the assumption that psychopathology lies on a continuum and can be described by different dimensions in a systematic hierarchy (Kotov et al., 2017; Lahey et al., 2017). It has been proposed that a hierarchical taxonomy consisting of a general psychopathology factor encompassing several dimensions/spectra (internalizing, thought disorder, disinhibited externalizing, antagonistic externalizing, detachment, and somatoform) comprised of different syndromes is suitable to characterize the majority of psychopathology (Kotov et al., 2017). In line with this approach, Ivansic et al. (2019) found that mental health in tinnitus patients can best be described by a general psychopathology factor and a somatization factor. They found that the expression of the general psychopathology factor was as high in severe tinnitus as in depressed patients, but more pronounced in mild tinnitus than in healthy controls. The somatization factor, on the other hand, was higher in both mild and severe tinnitus than in depressed patients or healthy controls (Ivansic et al., 2019).

The cognitive-behavioral approach, on which CBT is built, looks at psychopathology as complex causal networks (Hofmann, 2014). In this framework, certain triggers (moderated by attentional processes and trait cognitions) can activate maladaptive cognitive processes, which in turn lead to psychological distress manifesting as a specific interplay of subjective experiences, physiological symptoms, and behavioral responses (Hofmann, 2014). The focus of this approach lies on cognitive processes and their consequences for emotion regulation, which have proven to be important—and modifiable by CBT—for many different mental conditions (Hofmann et al., 2012; Hofmann, 2014). CBT also is known to have a positive effect on tinnitus management (Martinez-Devesa et al., 2010).

In a similar approach, the network perspective conceptualizes psychopathology as complex dynamic networks of mutually interacting symptoms (Fried et al., 2017). In this conceptualization, comorbidity between different mental conditions is thought to be explained by interactions between symptoms, in that the presence of a specific disorder can lead to the manifestation of another disorder via bridge symptoms (Fried et al., 2017). With this approach, the high comorbidity among severe tinnitus and mental disorders could potentially be explained by shared bridge symptoms (e.g., insomnia, concentration problems). Moreover, network approaches have the potential to predict transitions from a healthy network state to a disease state (Fried et al., 2017; van de Leemput et al., 2014), e.g., from mild to severe tinnitus-related distress, which has high clinical relevance.

In sum, all of these approaches appear suitable to better conceptualize tinnitus-related distress (emotions, cognitions, reactions), comorbid mental and physical symptoms, and their interrelationships than current diagnostic classification systems. In line with Stobik et al. (2005), we argue that bothersome tinnitus should be understood as a complex psychosomatic phenomenon including somatic, auditory, and psychosocial aspects, which can mutually reinforce each other. Consistent with this view, our results implicate the need for multimodal psychosomatic treatment for bothersome tinnitus in an interdisciplinary setting. Treatment-induced reductions of affective or anxiety symptoms by CBT can directly improve tinnitus-related distress as well as reduce negative effects of comorbid physical symptoms and hearing-related effects, whereas measures to restore hearing impairment have the potential to decrease aggravated negative effects of mental symptoms. Thus, multimodal treatment approaches combining psychological interventions, hearing aid provision, and medical treatment of comorbid physical symptoms appear to have the highest clinical potential to alleviate tinnitus-related distress.

Is NLD a Distinct Diagnostic Category Than Other Neurodevelopmental Disorders?

Most of the currently recognized neurodevelopmental disorders have seen important changes in their definition and the criteria used to identify them in the diagnostic classification systems (see, for example, how the definitions of specific learning disorders or autism spectrum disorders have changed from the first edition of the DSM or of the ICD till now). However, in the case of NLD the problem is more serious because even the existence of this disorder, as a specific profile, has been debated (Pennington, 2009; Spreen, 2011). Nevertheless, a growing number of cross-disorder comparisons are collecting evidences that clearly show neuropsychologic and neuroanatomical distinctions between children with NLD and those with other neurodevelopmental disorders (Semrud-Clikeman et al., 2010, 2013; Fine et al., 2014; Mammarella et al., 2019). In the next sections a brief overview of the studies in which groups with NLD have been compared with groups having other disorders will be given, also considering neuroanatomical and neuroimaging studies.

The problem with a diagnosis of NLD is that children with this disability are characterized by a core problem in visuospatial processing, but they also present a series of other symptoms that could be shared with other neurodevelopmental disorders. However, in NLD, these symptoms may be less intense and qualitatively different than in the case of the other disorders. For instance, similar social problems may be seen to some extent in individuals with NLD and in those with ASD (without intellectual disability); comparable academic difficulties may be observed in both NLD and specific learning disorders (such as dyscalculia); fine motor problems are present in NLD and in the developmental coordination disorder; difficulties in understanding the pragmatics of language, frequently present in NLD, are a characteristic of the social (pragmatic) communication disorder; finally, impairments of executive functions may be found both in children with NLD and in those with ADHD. Despite the importance of differentiating among these disorders and the related clinical and educational implications, not many studies have investigated the differences among them, and most of the research has focused on the comparison between NLD, ASD, and ADHD.

Semrud-Clikeman et al. (2010) compared an NLD group with children who had ASD or ADHD and typically developing children on several neuropsychologic tasks. They found the children with NLD more impaired than those with ASD or ADHD on measures of visuoconstructive skills, such as the visual-motor integration test (VMI; Beery and Buktenica, 2006) and the Rey Complex Figure Test (ROCFT; Rey, 1941), and also on measures of visual perception, such as the Judgment of Line Orientation Test (Benton et al., 1994). As for fine motor skills, both NLD and ASD groups performed lower than typical controls; however, their performances did not differ from those of the ADHD group. In addition, the NLD and ASD groups’ performances were similar in the fluid reasoning subtest, and in a test of spatial reasoning from the Woodcock–Johnson Cognitive III test set (WJ-Cog III; Woodcock et al., 2001). However, when Semrud-Clikeman et al. (2014) administered the same tasks to another sample of children, they found that the NLD group's performance was worse than that of the ASD group only in the spatial relations task. This latter finding supports the hypothesis that NLD weaknesses in reasoning tasks may be related to the manipulation of visuospatial information. In a further study, Semrud-Clikeman et al. (2014) collected three measures of executive functioning using the Delis–Kaplan Tests of Executive Functioning (D-KEFS; Delis et al., 2001), finding differences between children with NLD, children with ASD, and typically developing controls, especially in the trail-making task, which involved visuospatial working memory and sequencing. Overall, these findings support the hypothesis that clinical differences exist between NLD, ASD, and ADHD. However, to strengthen the hypothesis that children with NLD, differently from those with ASD without intellectual disability, are characterized by a core deficit in visuospatial processing, Mammarella et al. (2019) compared groups with NLD or ASD with typical controls in visuoconstructive and visuospatial working memory tasks, manipulating global versus local processing of the stimuli (Navon, 1977; Schooler, 2002; Caron et al., 2006). Although previous findings with ASD revealed mixed findings (e.g., Mammarella et al., 2014; Van der Hallen et al., 2015), there is evidence of the presence of a local bias, characterized by a focus on details, in the ASD group (Caron et al., 2006; see also Kuschner et al., 2009). Mammarella et al. (2019) used the paradigm proposed by Caron et al. (2006), and previously employed to study global versus local processing in ASD, to shed light on whether the ASD group shared any characteristics with the NLD group in terms of visuospatial processing. The results revealed that the NLD group's performance was worse in both the visuospatial domains examined (i.e., in visuoconstructive and visuospatial working memory tasks), whereas the group with ASD had less general difficulty and was affected by the global versus local (Navon, 1977) differentiation of the stimuli. Importantly, the NLD and ASD groups had similar full-scale IQ and visuoperceptual reasoning index of the Wechsler's scales; therefore differences in the experimental tasks could not be due to a low overall or visuospatial intelligence of the NLD group. The authors concluded that the manipulation of global–local processing styles in visuospatial tasks helped to better distinguish the profiles of children with NLD or ASD.

Neuroimaging studies have also been conducted to provide evidences in favor of a distinction between the NLD and other profiles. Semrud-Clikeman and Fine (2011), studying 28 children with NLD, found that 25% presented unsuspected brain abnormalities, generally including cysts or lesions in the occipital region, while this happened in only 4% of children with ASD, or controls. Another research (Fine et al., 2014) found that the area of the splenium was significantly smaller in children with NLD than in those who had ASD or ADHD, or in typical controls. Within the NLD group, those with a smaller splenium fared worse on spatial intelligence measures, whereas this association was not found in the group with ASD. In further studies, Semrud-Clikeman et al. (2013) identified significantly larger volumes of the amygdalae and hippocampi bilaterally in a group of children with ASD compared with controls or children with NLD. Moreover, both children with ASD and those with NLD had smaller left and right anterior cingulate cortex volumes than controls. This was the first evidence of children with NLD differing in some respects from children with ASD, but possibly sharing the same abnormal connectivity.