What is this article about?

Neuropsychological abilities are important for psychopathology. For example, they could serve as potential markers of genetic or other disease liability, or as components of pathophysiology. There is a growing movement at the NIMH that wants to improve the nosology (classification) of mental disorders, based on neuropsychological abilities, such as cognitive and emotional functioning.

However, this is hard. There is no one on one association between neuropsychological abilities and mental disorders. Thus, different psychiatric disorders can be characterized by similar neuropsychological deficiencies, such as problems in emotion regulation, problems in executive functioning, or problems in information processing. However, a lot of studies of psychopathology act like there is a one on one relationship between these phenotypes and individual disorders. The current article looks at what is the overall correlational structure of intermediate phenotypes within different mental disorders. Intermediate phenotype is used, because it is not known to what extent executive functioning or processing speed are endophenotypes, thus pathophysiological indicators. It is thus also not fully known how executive functioning and processing speed can serve as a liability indicator for different disorders. In this article, this will become more clear. The authors describe two important candidate intermediate phenotypes which are relevant to improve nosology of mental disorders, namely: executive functions (EF) and processing speed or response speed (Speed).

What is Executive Functioning?

Executive functioning is a very important term in general psychology. It refers to top-down (from the brain), goal-directed cognitive processing. There have been a lot of different, overlapping definitions. For example, some define it as a top-down process that is necessary for goal-directed action, while others define it as complex cognition. When there are novel situations, one is reliant on his or her executive functioning.

In relation to psychopathology, executive functioning determines a lot of capabilities, such as problem solving, impulse control, and emotion regulation. Thus, executive functioning processes have been suggested to be intermediate phenotypes, endophenotypes, or measures for attention-deficit hyperactivity disorder (ADHD), antisocial personality disorder, substance use disorder, depression, anxiety, schizophrenia, autism, and learning disorders.

Specific executive functioning processes differ across different models. In the current paper, the authors chose a few specific key abilities, namely: set-shifting and maintenance, interference control, response inhibition, and working memory. The authors also included response consistence/variability, which may be both a correlate of executive functioning, or ‘Speed’.

The fact that executive functioning includes a lot of different processes, complicates the discussion. Therefore, the authors look at it from a holistic perspective, and focus on a combined measurement of executive functioning. Executive functioning is further important, because there is a growing market of cognitive training which are aimed at improving certain components of executive functioning, such as improving working memory as a way to improve psychopathology. However, before such interventions can be set out, it is important to understand what executive functioning is, and what the relation is with psychopathology.

What is Output Speed?

 The term ‘speed’ is used to refer to different processes that define the speed of response on a task, or output speed. Output speed refers to speed of perception, efficiency of information processing and accumulation, speed of and bias toward response preparation, and speed of response execution. The final outcome of these processes is reflected in the speed of responding, and this will be called ‘speed’. It is important to distinguish this speed from executive functioning processes, and especially from efficiency.

Speed is not always theorized as a marker for psychopathology. However, it did gain some interest as an index of genetic liability and white matter integrity. In the literature, processing speed has been associated with intelligence (IQ), health, and psychopathology. Slow speed is widely used as an outcome measure in intervention studies for different health conditions. However, studies conducted on executive functioning often neglect processing speed. This is weird, because processing speed, which is a lower-order process, informs higher-order processes such as executive functioning.

Which psychopathologies are selected?

In the current article, the authors chose to focus on common, co-occurring disorders such as ADHD, antisocial personality disorder (ASPD), alcoholism, substance use disorder, depression, and anxiety disorder. The authors decided to focus on these disorders, because they often co-occur, and can be grouped into liability dimensions. Thus, they chose a sub-set of psychopathology disorders that are common and often overlapping.

What is the relationship between Speed, EF, and structures of psychopathology?

Weak executive functioning and slow Speed are often important in different disorders. However, EF dysfunction is not a necessary or sufficient cause of mental disorders. But, why is EF implicated in many disorders? There have been reviews that have compared EF across disorders, but the authors of these reviews agree that different studies use different methodologies, and that this makes interpretation difficult.

In the current study, the authors want to clarify what the combined relationship is between EF and Speed for common disorders. By doing this, they aim to understand how intermediate phenotypes can be integrated in nosology of mental disorders. One possibility could be that Speed sometimes accounts for the effects of EF, because EF measures are often confounded with Speed. The authors propose three hypotheses.

1. The Specificity model. This model suggests that different disorders are associated with different types of executive functioning deficits. Thus, according to this model, EF should not be considered as a single construct. Instead, it should be considered as consisting of components. This is called the component structure. This approach may show that ADHD is associated with poor working memory, and that antisocial behavior may be related with poor response inhibition.
2. The Severity model. According to this model, EF and Speed impairments are associated to the overall severity of psychopathology, instead of a specific form of psychopathology. Within disorders, severity varies in terms of impairment (overall functioning level), and other ways. The authors have used two proxies to measure severity. First, they looked at the number of co-occurring disorders. This index can be helpful in clarifying whether a cognitive function that is associated with a disorder is due to that disorder itself, or because the disorder is nested in a cluster of many disorders. Another approach is the clinician rating of Global Assessment of Functioning (GaF).
3. The Dimension Model. According to this model, EF or Speed deficits are related to one or more shared, underlying psychopathology liability dimensions, instead of specific disorders. E.g., Krueger (1999) found that there are two broad, superordinate factors which account for the pattern of correlations among liabilities to common mental disorders. This, he called, an internalizing factor and an externalizing factor. Internalizing factors were related to depression and anxiety disorders, and externalizing factors were related to anti-social and substance use disorders. Internalizing factors refer to that people experience problems in their thoughts, emotions, and feelings, but they do not necessarily show this in their behavior. Externalizing means that someone shows disrupted behaviour. Other studies have suggested that there is a general psychopathology, ‘g’ or ‘p’ factor.

What methods did the authors use?

The authors combined two adult samples from overlapping local communities, and they were chosen because they had extensive clinical evaluations of comorbid disorders, were community-recruited and thus avoided the bias of clinic-referred samples, and contained a broad representation of the disorders of interest.

Sample 1: ADHD Adults and Controls

This sample was initially recruited for a study which evaluated the relationship between executive functioning of adults with ADHD. The participants were recruited from the community through public advertisements. During a face-to-face interview with a clinician, eligible participants were assessed. These clinicians used the Structured Clinical Interview for DSM-IV Axis I Disorders. Disorders that are common in childhood (ADHD, CD, and ODD) were assessed with the Kiddie Schedule for Affective Disorders and Schizophrenia. IQ was assessed using the short form of the WAIS-III. Participants that were in a current period of major depression/mania/hypomania, or participants that were not able to remain sober during testing, participants with IQ < 75, participants that were taking anti-psychotic, anti-depressant, or anti-convulsant medication were excluded. The final sample consisted of 363 adults.

Sample 2: Substance Abuse Study

This sample was initially recruited for the study of the etiology of substance use disorders from preschool years until the time of greatest substance use. The participants consisted of parents from families who had participated in the Michigan Longitudinal Study (Substance Abuse Study), which is an ongoing longitudinal study of the development of alcohol and other substance use disorders. These families were recruited from the community, based on the alcoholism status of the father during 1985-1993, when their target child was in preschool and the parents were aged 20-53. The data was collected at the initial recruitment (Wave 1), and at 3-year intervals after Wave 1. In Wave 5, participants underwent neuropsychological testing. There were 159 court alcoholic families. These families were recruited when the father was ever convicted of drunk driving, and had a high blood alcohol content, but was not undergoing litigation. Also, there were 91 control families, in which neither parent was associated with substance use disorder. These were recruited with door-to-door canvassing in the neighborhoods of the court alcoholic families. Lastly, this canvassing also lead to the discovery of community alcoholic families, which were 61 families. Community alcoholic families refer to families in which the father met the criteria for alcoholism, but was never arrested for drunk driving. This initial sample consisted of 607 men and women. The SMAST, Drinking and Drug History Questionnaire, the Antisocial Behavior Inventory, and the Diagnostic Interview Schedule Version III were completed. DSM-IV alcohol-related and ASPD diagnoses were also established by clinicians. IQ was assessed using the short form of the WAIS-R. The authors counted the total disorders, to achieve the index of severity. Participants were excluded if they met criteria for lifetime diagnosis of psychosis or bipolar disorder, or when their IQ < 75. This led to the exclusion of 2% of the participants. The final sample then consisted of 470 participants.

Neuropsychological Test Battery for Both Samples

The neuropsychological measures were the same for both samples.

Set Shifting

To measure set shifting, the authors used the ‘Trail Making Test’, which is a paper-and-pencil test that consists of two parts. In part A, participants have to connect numbered circles as fast as possible, without making errors. In part B, participants need to switch between numbers and letters in alphabetical-numerical order. Part A reflects measures of output speed, and part B reflects scanning and motor speed, and switching.

Interference Control

To measure interference control, the authors used the Stroop Color-Word Test. In this test, the participant reads words like “red” or “green”, but the ink of the letters varies. So, one can read “green” in red ink. This shows interference control: are they able to say ‘red’, even when the ink is green?

Set Maintenance and Working Memory

To measure higher level cognitive control, the authors administered the Wisconsin Card Sorting Test (WCST). Participants view a computer screen with four ‘key cards’ which they need to match to a stimulus card. This measure reflects rule detection, set maintenance, shifting, interference, and working memory updating, in short working memory.

Response Inhibition

The Go-Stop Task was administered to measure response inhibition. Participants view a computer screen, which displays an X or an O on a black and white screen. The participants need to respond to these stimuli, by pressing the buttons ‘X’ and ‘O’ as quickly as possible. When they hear a tone, they need to withhold their response.

Response Time (RT) variability

To measure RT variability, the authors computed as within-subject standard deviation on the ‘go-trials’ of the stop-go task. This measure reflects EF or arousal state.

What can be concluded?

In this article, the authors looked at the relationship between psychiatric disorders with EF and response speed. They found that psychopathologies can be conceptualized hierarchically. Different manifestations of psychopathology such as alcoholism or depression are nested within higher order factors. The authors found that processing speed is associated with higher order liabilities for psychopathology, and for its  severity. This was especially true for the externalizing domain and the domain of ADHD. In contrast, EF did not operate as a general factor, but on a componential level, in which specific cognitive weaknesses were related to specific manifestations of psychopathology. This suggests that there is a two-stage model of neuropsychology and psychopathology that may guide how we approach intermediate phenotypes. From this idea, it can be concluded that processing or response Speed is an intermediate phenotype for both externalizing psychopathology and overall severity of psychopathology. EF components are intermediate phenotypes for subsets of that liability. It may also be the case that the neurobiological basis of processing speed reflects delayed or immature white matter development. For instance, in individuals with ADHD, there have been notations of delayed cortical maturation, and diffuse white matter alterations. This is also the case for individuals with antisocial personality disorder, and substance abuse.