Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Feb 1.
Published in final edited form as: Child Neuropsychol. 2021 Sep 20;28(2):266–286. doi: 10.1080/09297049.2021.1972957

What is a processing speed weakness? Importance of cognitive ability when defining processing speed in a child psychiatric population

GA Forchelli 1,2, PJ Vuijk 3, MK Colvin 1,2, A K Ward 1, M R Koven 3, A Dews 3, AE Doyle 1,2,3, E B Braaten 1,2
PMCID: PMC9284538  NIHMSID: NIHMS1736408  PMID: 34544318

Abstract

There is growing evidence that processing speed (PS) deficits in youth with neuropsychiatric conditions are associated with functional difficulties. However, there is no consistent definition of slower PS; specifically, whether slower PS should be defined as a discrepancy from same-aged peers (normative weakness) or as an intrapersonal deficit relative to overall cognitive ability (relative weakness). In a sample of clinically-referred youth, we calculated slower PS both ways and examined the impact on adaptive, academic, and psychopathology outcomes in relation to different levels of cognitive ability. Significant PS x cognitive ability interactions were found on adaptive and academic outcomes. A norm-based weakness in PS (PSI Standard Score <85) was associated with lower adaptive skills and lower academic skills regardless of cognitive ability. In the above average cognitive ability group, relatively lower PS (PSI<15 point difference from VCI) was associated with significantly lower academic performance. No significant associations were found for general psychopathology. Results suggest a normative weakness in PS impacts functional outcomes interactively and differently with level of general cognitive ability. Data suggest that higher cognitive ability may be somewhat protective from the impact of normatively weak PS on adaptive outcomes; however, youth across all abilities with normatively weak PS showed weaker academic performance. Second, children with high cognitive abilities and relatively weak PS showed discrepant performance compared to comparison group. Implications and areas for future research are discussed.

Keywords: Processing Speed, Cognitive Ability, Academic Performance, Adaptive Skills, Learning Disability

Introduction

Processing speed (PS) refers to the time it takes for an individual to perceive, process, and respond to a stimulus. In other words, it is a measure of how quickly and efficiently an individual can get things done in a given period of time (Braaten & Willoughby, 2014). Research has shown increasing evidence of PS’s clinical relevance in children with neurological and medical conditions, such as epilepsy (Bobholz, Dabbs, et al., 2019), hypoglycemia in type 1 diabetes (Bortolotti, Zarantonello, et al., 2018), and pediatric onset multiple sclerosis (Wallach et al., 2020).

PS weakness has also been identified as a clinically relevant finding in neurodevelopmental and neuropsychiatric conditions. PS weaknesses are found in children with Attention Deficit/Hyperactivity Disorder (AD/HD; Goth-Owens et al., 2010; Shanahan et al., 2006), Autism Spectrum Disorder (ASD; Doobay et la., 2018), and psychosis (Kelleher et al., 2013). In a recent study of 751 clinically-referred group of participants 6 to 21-years-old, there were significantly more PS weaknesses in children with any neurodevelopmental or psychiatric diagnosis when compared to those without a diagnosis (Braaten, Ward, Forchelli et al., 2020). In this same study, slow PS also was related to inattention across diagnostic groups, indicating its importance as a cross diagnostic construct. Recent studies have not consistently found significant PS weaknesses in children with anxiety or mood diagnoses or dimensions (Braaten, Ward, Forchelli, et al., 2020; Calhoun & Mayes, 2005; Mayes & Calhoun, 2007). A recent meta-analysis, however, did find some implications of PS weakness when adolescents with Major Depressive Disorder showed remission of depressive symptoms (Baune et al., 2014). Children with psychotic disorders have also have higher rates of PS weaknesses across studies (Braaten, Ward, Forchelli, et al., 2020; Niarchou et al., 2013), consistent with the adult literature.

In child and adolescent populations, PS weaknesses are related to functional impairments. In one sample of children with high-functioning ASD, children with slower PS demonstrated significantly increased social communication and adaptive problems, as rated on behavioral rating scales (Oliveras et al., 2012). Furthermore, slower PS partially mediated increased social difficulties in children with inattention symptoms and AD/HD (Thorsen et al., 2018). Research has also linked PS weakness to academic outcomes. In the previously mentioned study by Braaten and colleagues (2020), PS weaknesses had a negative effect on both math and reading achievement in children. Similarly, in a study exploring the relationship between PS and symptoms of “sluggish cognitive tempo” in children with AD/HD, a PS weakness predicted lower academic achievement outcomes (Cook et al., 2019). These studies suggest adaptive, social, and academic implications for PS weakness.

In clinical and school-based settings, the Processing Speed Index (PSI) from the Wechsler Intelligence Scales is arguably one of the most commonly used measures of PS. However, the definition of PS weakness using the Wechsler PSI varies across studies. Some use a clinical cutoff, such as Braaten and colleagues’ (2020) “PS impairment,” which was defined as a normative weakness on the PSI (Standard Score less than 85). This cutoff may, however, be problematic and underrepresent children who show more of a relative pattern of PS weakness and stronger overall IQ. To this point, in another large study that looked at PS impairment in relation to overall intelligence/FSIQ, the extent of the IQ-PS difference differentiated groups of children with ADHD, ASD, and learning disabilities (Calhoun & Mayes, 2005). Overall, it is unclear in the literature whether the way we define a PS weakness matters across groups.

Results from a study by Thaler and colleagues (2012) went further and suggested that the relationship between PSI and functional outcomes may depend on cognitive ability. In a sample of 189 children with ADHD, they found distinctive clusters of individuals based on their performance on the WISC-IV and functional outcome measures. Children who showed “somewhat low average” PS compared to their IQ scores (i.e., WISC-IV VCIM = 104, PSI M = 90) or below average PS (i.e., Standard Score <85) showed differences on ADHD symptoms, academic performance, and adaptive functioning compared to counterparts with Average to High Average cognitive skills and at least Average PS. More specifically, children with the relative weakness in PS showed significant differences in their presentation of inattentive symptoms on behavior rating scales compared to children with Above Average IQ and at least Average PS. The below average PS group demonstrated weaker adaptive skills than the Average to Above Average cognitive groups. Both the relative and below average PS groups also showed weaknesses in academic fluency tasks compared to the other participants. This suggests that there may be a distinction in how PS weaknesses are defined in relation to cognitive ability and may differentially impact outcome measures. However, this study only explored this in a small sample of children with ADHD.

In a more recent study, PS weakness was a prominent characteristic found in children with lower cognitive ability who also had a co-occurring psychiatric diagnosis (Santegoeds et al., 2021). When PS weakness was controlled, there was no substantial difference in attention or other executive function measures in a group of children with mild to borderline intellectual disabilities when compared to the age- and sex-based normative data. The authors postulated that children with intellectual impairments do not show uniform weakness across all neurocognitive domains but show particular vulnerability to tasks that require efficient processing (i.e., PS).

Overall, research has shown a connection between PS weakness, overall cognitive ability, and functional, behavioral, and academic outcomes, independently. We and others have previously addressed how PS impacts performance across various psychiatric populations (e.g., Braaten, Ward, Forchelli, et al., 2020; Calhoun & Mayes, 2005; Mayes & Calhoun, 2007). Smaller studies have also suggested that PS weakness may be related to cognitive ability (e.g., Santegoeds et al., 2021). Studies, however, have not explicitly considered how high or low cognitive ability may differentially impact the influence of PS weakness on outcomes in a cross-diagnostic group. Further, the definition of a PS weakness is variably defined across research studies, making its clinical meaningfulness somewhat unclear. That is, in relation to cognitive ability, we need to consider whether a clinical cutoff or a relative weakness in PS is differentially important for functional and clinical outcomes.

In our current study, we examined the contribution of PS weakness to clinical and functional outcomes in a large, generalizable outpatient child psychiatry sample in relation to different levels of cognitive ability. Due to the clinical utility and popularity of the Wechsler scale PSI as a complex measure of PS, this was used in the current study. It is hypothesized that participants with normatively weak PS, defined as a weakness on the Wechsler Intelligence PSI (Standard Score < 85), will show significantly more functional impairment on outcome measures, including adaptive skills, academic performance, and psychopathology as compared to participants without a PS weakness. Participants with a relative impairment in PS (<15 points lower than overall cognitive ability) will also show greater functional weakness than those without such relative discrepancy.

Methods and Materials

Participants

Participants were from the Longitudinal Study of Genetic Influences on Cognition (LOGIC). LOGIC recruits youth consecutively referred for evaluation at a pediatric assessment clinic within the Psychiatry Department at Massachusetts General Hospital (MGH). Patients with neuropsychiatric symptomatology are referred to this clinic for a comprehensive evaluation to assist with differential diagnosis and/or treatment or educational planning. Patients are approached in the clinic waiting room and receive a gift card if they agree to participate. To enroll, youth must contribute their clinical data and are asked to contribute supplemental assessments to create a uniform battery of measures across participants. They are also asked to provide DNA; however genomic information was not part of the current analyses. Study procedures were in compliance with the Partners Institutional Review Board. Parents provide written informed consent after a description of risks and benefits; youth 7–17 provide written assent. Participants in the current analysis were consecutively enrolled, unrelated patients ages 6 to 17 with a mean age of 11.1 ± 3.0 years and 468 (37.1%) were female. LOGIC is an ongoing project. Participants with a Full Scale IQ > 55 were included in this study. We did not rule out youth with an IQ below 70 so as to explore PS weakness across the entire cognitive spectrum and because IQ estimates may be impacted by a low PSI. At the time of these analyses, there were 1261 unrelated youth who met these criteria (Table 1).

Table 1.

Characteristics of a youth clinical outpatient sample ages 6–17 (n=1261)

Variables Total sampleM (SD)
Age; mean (SD) 11.1 (3.0)
Sex; Nmale (%) 793 (62.9%)
Psychotropic medication; Nyes (%) 516 (40.9%)
ABAS-2/3
Conceptual Composite 86.5 (14.7)
Practical Composite 84.2 (17.1)
Social Composite 89.4 (17.1)
WIAT-2/3
Word Reading 98.5 (15.5)
Numerical Operations 96.3 (15.6)
CBCL
Internalizing Problems 60.3 (11.5)
Externalizing Problems 56.3 (11.8)
WISC-4/5 or WAIS-4
Full Scale IQ 97.7 (15.7)
Processing Speed Index (PSI) 92.5 (15.3)
PSI groups:
PSI<85 74.4 (8.8); n=365 (28.9%)
PSI≥85 99.9 (10.3); n=896 (71.1%)
Verbal Comprehension Index (VCI) 101.6 (16.3)
VCI groups:
VCI<90 78.3 (10.9); n= 249 (19.7%)
90≤VCI≤110 100.5 (5.4); n = 669 (53.1%)
VCI>110 120.7 (8.7); n = 343 (27.2%)
Open in a new tab

Note. ABAS-2/3 = Adaptive Behavior Assessment System, 2nd or 3rd Edition; WIAT-2/3 = Weschler Individual Achievement tests, 2nd or 3rd Edition; CBCL = Child Behavior Checklist; WISC-4/5 or WAIS-4 (Weschler Intelligence Scales)

Measures

Tests were administered using published instructions by licensed psychologists or by advanced trainees or psychometricians under their supervision. We used measures with robust psychometric properties that are commonly used in child clinical practice and research. Measures included:

General cognitive ability/verbal comprehension.

We assessed cognitive ability using the verbal comprehension index (VCI) from the Wechsler Intelligence Scale for Children – Fourth and Fifth Edition (Wechsler, 2004; Wechsler, 2014)for youth 7 to 16 years and the Wechsler Adult Intelligence Scale – Fourth Edition (Wechsler, 2008) for youth 16 and 17 years. The VCI is highly correlated across the WISC-IV and WISC-V (r = 0.85) (Wechsler, 2014). This index score has been used in prior studies as an estimate of intelligence (Cook et al., 2019) and was selected for the current study to represent cognitive ability. We did not use any another global intelligence index (i.e., FSIQ, GCA, FRI, VSI) from the Wechsler tests because they all include subtests with timed subtests, such as Block Design or PS subtests, and we wanted to ensure that measures tapping into PS were not calculated into our intelligence estimate. Furthermore, the correlation between VCI and FSIQ in our sample of 1,261 participants is r = .84 and between VCI and GAI it is r = .89 which is indictive of common variance between VCI and GAI, and VCI and FSIQ is very high. Lastly, verbal IQ shows stronger relations with achievement relative to performance IQ measures (Wechsler 2003, 2014).

Processing speed (PS).

We assessed processing speed using the processing speed index (PSI) from the Wechsler Intelligence Scale for Children – Fourth and Fifth Edition (Wechsler, 2004; Wechsler, 2014)for youth 6 to 16 years old and the Wechsler Adult Intelligence Scale – Fourth Edition (Wechsler, 2008) for youth 16 and 17 years old. The PSI is highly correlated across the WISC-IV and WISC-V (r = 0.71) (Wechsler, 2014). The PSI is composed of 2 subtests: Coding and Symbol Search. Coding requires participants to quickly draw abstract visual symbols associated with a specific number in a corresponding visual key, as quickly and accurately as possible. In Symbol Search, participants quickly scan and attempt to match one visual symbol in an array of other visual symbols as quickly and accurately as possible. Raw scores for each represent the total number of items completed in a 2-minute time period. These are, then, converted to scaled scores (M = 10, SD = 3) which are then combined to form the PSI and produce a standard score (M = 100, SD = 15).

PS weakness definition.

The same clinical sample was analyzed in two different ways to define PS weakness. For Aim 1 of our study, Normative PS weakness was defined as youth with a normatively-based weakness in PS were defined by a Wechsler PSI score < 85 (i.e., one standard deviation below the mean according to normative data). This strategy identifies youth with a score that is one standard deviation below the population mean and was used in a recent prior study by Braaten and colleagues (2020). For Aim 2 of our study, Relative PS weakness was defined as youth with a relatively lower PS score compared to VCI (discrepancy ≥ 15 points relative to their own Wechsler VCI).

Dimensional rating of psychopathology.

Ratings of psychopathology were taken from symptom rating scales completed by participants’ primary caregivers. We used the Internalizing and Externalizing indices from the Child Behavior Checklist/6–18 (CBCL; Achenbach & Rescorla, 2001) for a global indication of psychological symptoms (Cronbach’s alpha = .94 and .72, respectively).

Adaptive skills.

Ratings of adaptive skills were taken from symptom rating scales completed by participant’s primary caregivers. We used the three composite scores from the Adaptive Behavior Assessment System, 2nd or 3rd Edition (ABAS-2, ABAS-3; Harrison & Oakland 2003, 2015).

Academic skills.

The Word Reading and Numerical Operations subtests from the Wechsler Individual Achievement Tests, 2nd or 3rd Edition (WIAT-2/WIAT-3; Wechsler (2005, 2009) were used as untimed measures of reading and math achievement. The WIAT-2 and WIAT-3 are well normed and widely used individually-administered instruments developed to measure academic achievement of students ages 4 to 50 years old using a normative sample that is generally comparable to that of the WISC. The Word Reading subtest asks a participant to read sight words in isolation. The Numerical Operations subtest asks participants to complete straightforward paper-and-pencil math computation and calculation problems. The age-referenced standard score for these two subtests were used.

Clinical diagnoses.

Participants in the current sample received either clinical DSM-IV-TR or DSM-V diagnoses, reflecting the diagnostic system at the time of their enrollment. Diagnoses were made by or under the supervision of doctoral-level licensed psychologists who were hospital faculty. Our source clinic is an accredited training site for clinical psychology interns and post-doctoral fellows, with an emphasis on thorough, accurate diagnosis. Information is gathered from interviews with parents/guardians and patients, review of medical records, and rating scales. Diagnoses were coded by the clinician in the clinic record if full DSM-IV-TR or DSM-V criteria were met. Diagnoses remained largely the same from the transition from DSM-IV to DSM-V, with the following exceptions. First, an attention deficit/hyperactivity disorder (ADHD) diagnosis made at the time of the DSM-IV was allowed to co-occur with an autism spectrum disorder (ASD), in anticipation of DSM-V. Second, at the start of the study, four conditions (conduct disorder (CD), oppositional defiant disorder (ODD), social phobia/ social anxiety disorder and separation anxiety disorder were only coded by clinicians if they were considered primary diagnoses. For research purposes, these conditions were supplemented with an algorithm that also coded them as positive if they were positive based on a consensus of symptom-level parent reports from questionnaires in the extended clinic record. For those patients who had one positive questionnaire or limited questionnaire data available, two clinicians independently reviewed the diagnoses and a final consensus diagnosis was made based on a discussion of their ratings.

Youth manifested a range of conditions and comorbidity. Under the disruptive behavioral diagnoses heading from DSM-IV, the rate of ADHD was 60.2% (n=759), the rate of Conduct Disorder was 15.9% (n=201) and the rate of Oppositional Defiant Disorder was 12.4% (n=156). The Autism Spectrum Disorders (ASD) were diagnosed in 17.2% (n=217) of the youth. Regarding mood disorders, diagnoses included: 2.8% (n=35) Bipolar Disorder (including Other Specified BPD or Related Disorders), 7.1% (n=90) Major Depressive Disorder, 0.9% (n=11) Dysthymic/ Persistent Depressive Disorder, and 9.1% (n=115) had an “other” mood disorders (i.e., DSM-4 Mood Disorder NOS, DSM-5: Other Specified Depressive Disorder, Disruptive Mood Dysregulation Disorder). Within the anxiety domain, the sample included: 0.3% (n=4) Panic Disorder, 10.4% (n=131) Generalized Anxiety Disorder (GAD), Social Phobia 6.2% (n=78), Separation Anxiety Disorder 4.0% (n=50), and Other Specified Anxiety Disorders 16.3% (n=206). On the psychosis spectrum, diagnoses were: 0.4% (n=5) Schizophrenia/ Schizoaffective Disorder and 2.5% (n=31) Prodromal/Other Specified Schizophrenia Spectrum and Psychotic Disorders. Additionally, 16.3% (n=205) of referred youth did not meet criteria for a full Axis I diagnosis. Rates of diagnoses surpassed 100% due to comorbidity (Table 2).

Table 2.

Diagnosis information of a youth clinical outpatient sample ages 6–17 (n=1261)

Diagnosis* Total Samplen (%)
ADHD 759 (60.2)
ASD 217 (17.2)
Mood Disorders 251 (19.9)
Anxiety Disorders 469 (37.2)
Psychosis (broad) 36 (2.9)
ODD 242 (19.2)
Conduct Disorder 145 (11.5)
Other** 196 (15.5)
Number of diagnoses
0 196 (15.5))
1 446 (35.4)
2 319 (25.3)
3 196 (15.5)
≥4 104 (8.2)
Open in a new tab

Note. ADHD = Attention-Deficit/Hyperactivity Disorder; ASD = Autism Spectrum Disorder; ODD = Oppositional Defiant Disorder

*

Due to comorbidity, numbers do not add up to 100%

**

Other = subclinical diagnoses and/or learning disabilities

Medication.

Detailed data regarding current use of psychotropic medication (dose, type, onset, offset) was obtained as part of the clinical evaluation. In this sample 22.7% (n=286) children were taking stimulants, 10.4% (n=131) were on non-stimulant medication to treat ADHD (e.g., atomoxetine), 8.7% (n=110) were taking an atypical antipsychotic, 13.6% (n=171) were taking a Selective Serotonin Reuptake Inhibitor (SSRI), 3.6% (n=46) were taking a non-SSRI antidepressant, 1.8% (n=23) were taking a benzodiazepine, and 5.4% (n=68) were taking another type of psychotropic medication. These medications were used to create a binary variable to indicate current use of one or more types of psychotropic medications versus non-use, resulting on a total of 40.9% (n=516) youth using any psychotropic medication.

Data Analyses

For our first aim, the group with a normative PS weakness (Wechsler PSI score < 85) were compared to the group without this weakness (Wechsler PSI score ≥ 85). Three groups were created based on the VCI score as a measure of general cognitive ability:1) below average Wechsler VCI (<90), 2) average Wechsler VCI (90≤VCI≤110), 3) above average VCI (>110). Two-way ANCOVA models (PS with 2 levels and general cognitive ability with 3 levels) were used to examine the effect of normative PS weakness on types of adaptive functioning (i.e., practical, social and conceptual), academic functioning (i.e., reading and math), and psychopathology (i.e., internalizing and externalizing problems) at different levels of general cognitive ability while controlling for age, sex, and psychotropic medication. A significant interaction effect between the PS group and the cognitive ability group variable indicates that the effect of PS weakness on the outcome is dependent on the level of general cognitive ability and was followed by post-hoc ANCOVA analyses at each level of the three levels of general cognitive ability.

For the second aim, in the subset of youth with above average general cognitive ability (Wechsler VCI > 110), ANCOVA models were used to examine whether a group with a relative PS weakness (Wechsler VCI - PSI ≥ 15) compared to a group without such difference (i.e., VCI- PSI <15) is associated with worse outcomes on the aforementioned areas while controlling for age, sex, psychotropic medication, and cognitive ability. For all ANCOVA analyses the η2 effect size is calculated by dividing the sum of squares of the effect by the total sum of squares. An η2 of .01 is small, .06 is moderate, and .14 is considered a large effect size (Cohen, 1988). STATA 14 was used for all analyses. A False Discovery Rate (FDR) to correct for multiple testing (Benjamini & Hochberg, 1995) using a critical value of .05 before correction separate for aim 1 and 2. Please refer to Table 1 for all descriptive statistics of this sample.

Results

Aim 1. The effect of a normative PS weakness on adaptive functioning, academic functioning, and psychopathology in the context of different levels of general cognitive ability

Significant differences between the normative PS weakness and comparison group were found on age where the normative weakness group was 0.5 years older on average (t(1259) = 2.59, p =.005), had a lower proportion of girls (χ2(1) =29.79, p<.001), a higher proportion of usage of psychotropic medication (χ2(1) =13.09, p<.001), and a lower mean score on verbal comprehension (t(1259) = 10.96, p <.0001). We therefore control for all these variables in aim 1. There were also higher rates of ADHD, ASD, Mood disorders and psychosis, and we note we have addressed cross-diagnostic effects in our previous work using a multivariate method (see Supplementary Table S1.)

Adaptive functioning:

A significant interaction between PS and cognitive ability level was found for each of the 3 outcomes of adaptive functioning. For practical adaptive skills, there was a significant interaction (F (2, 1068) = 4.46, p = .01, η2=0.01). Post hoc analyses showed that practical adaptive skills were lower for those with below average (F (1, 216) = 8.02, p = .01, η2=0.03) and average (F (1, 562) = 4.48, p = .01, η2=0.01) cognitive ability and normative PS weakness. There was no significant impact of PS for those with above average cognitive ability (Figure 1a). For social adaptive skills, the normative PS weakness and general cognitive ability level (F(2,1095)=4.83, p = .01, η2=0.01) was driven by normative PS weakness having a significant negative impact on social adaptive skills at a below average level of general cognitive ability (F (1, 223) = 6.48, p = .01, η2=0.02) (Figure 1b). It was not at any of the other two levels of general cognitive ability. For conceptual adaptive skills, similar to practical adaptive skills, the significant interaction between normative PS weakness and general cognitive ability (F(2,1072)=3.81, p = .02, η2=0.01) was driven by a negative impact of normative PS weakness only at below (F (1, 217) = 13.38, p < .001, η2=0.05) and average (F (1, 564) = 9.48, p = .002, η2=0.02) levels of general cognitive ability (Figure 1c).

Figure 1a.

Figure 1a

Open in a new tab

Main effect analyses between cognitive ability (VCI) and PS groups on practical adaptive skills scores on ABAS-2/3.

Figure 1b Main effect analyses between cognitive ability (VCI) and PS groups in social adaptive skill scores on ABAS-2/3.

Figure 1c Main effect analyses between cognitive ability (VCI) and PS groups in conceptual adaptive skill scores on ABAS-2/3.

Academic functioning:

For word reading, there was a significant interaction between normative PS weakness and general cognitive ability level (F(2,1235)=3.57, p = .03, η2=0.01). Word reading was lower for all cognitive groups with normative PS weakness, but the effect was strongest for the below average (F (1, 243) = 16.52, p < .001, η2=0.06) and above average (F (1, 327) = 11.56, p < .001, η2=0.05) cognitive groups (Figure 2a). For math, a significant interaction between general cognitive ability level and PS (F(2,1252)=9.64, p < .001, η2=0.01) was found. Math performance was significantly lower for all cognitive groups, but seemed to be strongest at below average (F (1, 244) = 75.32, p < .001, η2=0.22), followed by average (F (1, 664) = 38.55, p < .001, η2=0.03) and above average (F (1, 338) = 15.33, p < .001, η2=0.04) levels of general cognitive ability (Figure 2b).

Figure 2a.

Figure 2a

Open in a new tab

Main effect analyses between cognitive ability (VCI) and PS groups in Word reading subtest performance on the WIAT-2/3.

Figure 2b Main effect analyses between cognitive ability (VCI) and PS groups in Numerical Operation subtest performance on the WIAT-2/3.

Psychopathology:

No significant interaction effects between normative PS weakness and general cognitive ability level were found for internalizing problems (F(2,1245)=0.55, p = .58, η2=0.00) and externalizing problems (F(2,1245)=0.31, p = .73, η2=0.00). Dropping the interaction terms from the models did not result in a significant main effect of normative PS weakness for internalizing (F(1,1252)=0.18, p = .67, η2=0.00) or externalizing problems (F(1,1247)=0.83, p = .36, η2=0.00). (Figure 3).

Figure 3a.

Figure 3a

Open in a new tab

Main effect analyses between cognitive ability (VCI) and PS groups on Internalizing Symptom Index on the CBCL.

Figure 3b Main effect analyses between cognitive ability (VCI) and PS groups on Externalizing Symptom Index on the CBCL.

Aim 2. The effect of relative PS weakness in youth with above average general cognitive ability

No significant differences in this subsample of youth with above average general cognitive ability were found between a group with and without a relative PS weakness on age, sex, use of psychotropic medication and diagnostic status (Table 3). The ANCOVA analyses yielded small to moderate effects for academic functioning, where the relative PS weakness group performed significantly lower than the comparison group on reading (~5 points lower; F(1,298)=13.33, p <.001, η2=0.03) and math (~6 points lower; F(1,309)=16.01, p < .001, η2=0.04). For adaptive functioning there were significant lower scores for the relative PS weakness group on practical adaptive skills (~4.5 point lower; F(1,256)=4.29, p = .039, η2=0.01)) and conceptual adaptive skills (~4 point lower; F(1,257)=5.07, p = .025, η2=0.02) however these results did not survive correction for multiple testing. No significant differences were found on internalizing and externalizing problems between the 2 groups in youth with above average general cognitive ability (Table 4.).

Table 3.

Comparison of the relative PS weakness group (VCI – PSI ≥ 15) and the comparison group (VCI-PSI <15) on demographics and diagnoses (n=343)

Variable Comparison group (n=114) Relative PS weakness group (n=229)
Mean (SD) Mean (SD) t-test p-value
Age 11.0 (3.3) 11.6 (3.2) 1.66 .10
Verbal Comprehension 116.9 (5.5) 122.5 (9.4) 5.88 <.0001
Processing Speed 110.6 (9.1) 91.9 (11.1) 15.62 <.0001
N(%) N(%) χ 2 (1) p-value
Sex (girls) 46 (40.4) 75 (32.8) 1.93 .17
Psychotropic Medication 45 (39.5) 94 (41.0) 0.08 .78
ADHD 52 (45.6) 128 (55.9) 3.23 .07
ASD 16 (14.0) 44 (19.2) 1.41 .23
Mood Disorders 25 (21.9) 48 (21.0) 0.04 .84
Anxiety Disorders 47 (41.2) 95 (41.5) 0.01 .96
Psychosis (broad) 2 (1.8) 7 (3.1) 0.51 .48
ODD 21 (18.4) 39 (17.0) 0.10 .75
Conduct Disorder 9 (7.9) 17 (7.4) 0.02 .88
Open in a new tab

Note. Non-discrepant group (VCI-PSI<15), discrepant group (VCI-PSI≥15); ADHD = Attention-Deficit/Hyperactivity Disorder; ASD = Autism Spectrum Disorder; ODD = Oppositional Defiant Disorder.

Table 4.

ANCOVA models examining the differences between the relative PS weakness group (VCI – PSI ≥ 15) and the comparison group (VCI-PSI <15) in a subsample with Above Average cognitive ability (VCI >110)

Non-discrepant group Discrepant group
Domain F-value df p-value η2 Estimated marginal mean (SE) Estimated marginal mean (SE)
Adaptive Functioning
Practical Adaptive Skills 4.29 1,256 .039 0.01 90.45 (1.74) 85.93 (1.18)
Social Adaptive Skills 0.28 1,261 .596 0.00 91.32 (1.72) 92.46 (1.15)
Conceptual Adaptive Skills 5.07 1,257 .025 0.02 94.48 (1.45) 90.38 (0.97)
Academic Functioning
Word Reading 13.33 1,298 <.001 0.03 112.19 (1.18) 106.85 (0.79)
Numerical Operations 16.01 1,309 <.001 0.04 110.69 (1.31) 104.11 (0.90)
Psychopathology
Internalizing Problems 0.00 1,310 .992 0.00 60.90 (1.07) 60.89 (0.71)
Externalizing Problems 0.07 1,310 .798 0.00 55.03 (1.01) 55.35 (0.67)
Open in a new tab

Note. In all models the covariates were age, sex, use of psychotropic medication and verbal comprehension index score.

Discussion

Prior research has shown that slower PS in ADHD populations (Cook et al., 2017, 2019) and in a larger cross-diagnostic pediatric population (Braaten et al., 2020; Calhoun & Mayes, 2005; Calhoun & Mayes, 2007; Mayes & Calhoun, 2004) is associated with negative functional outcomes. Our results echo this literature base, showing that in a large sample of children with varying psychiatric conditions, those with PS weakness showed significantly lower academic and adaptive skills. Initial univariate analyses demonstrated that there were differences according to age, sex, and medication, as well as cognitive ability levels. We determined there was a need to control for these and look at them differentially across cognitive levels.

The definition of PS weakness or slow PS has varied across studies in both relation to the normative population and in relation to cognitive ability. Our research has expanded the literature base in showing that, in a large, clinically-referred group of children with various neurodevelopmental and psychiatric presentations, it is beneficial to conceptualize PS weakness in the context of cognitive ability. In our sample, the impact of PS weakness was not uniform across all cognitive ability groups. While prior research has uniformly identified that, generally, children with slow PS have adaptive weaknesses (e.g., Oliver as et al., 2012, Thorsen et al., 2018), our study further clarifies those adaptive outcomes may be dependent on the relationship between cognitive ability and slow PS. For adaptive skills, children with Average and Below Average cognitive ability and normative PS weakness had significantly poorer practical/daily living skills and conceptual skills. Children with Below Average cognition ability and normative PS weakness also demonstrated more social weaknesses. For children with Above Average cognitive ability, normative PS weakness did not significantly impact adaptive outcomes.

The variability in the types of adaptive difficulties in relation to cognitive ability and slow PS is clinically relevant. While it may be expected that children with Below Average cognitive ability show significantly weaker adaptive skills, our work suggests that slower PS impairs skills even further. Activities measured by these scales, such as more struggles with completing chores and management of peer relationships, may be more difficult to remediate in children with both lower cognitive aptitude and slower PS across various presentations of psychopathology. Clinicians may need to be mindful of the intensity and frequency of interventions to support growth in these areas when compared to those without slow PS.

The ratings of adaptive skills in those with Average or Above Average cognitive ability and PS weaknesses (or lack thereof) are also noteworthy. First, children with Average intelligence and normatively slow PS show more difficulties on parent ratings of daily living skills (i.e., ABAS-3 Practical). This is less expected for children with appropriate intellectual ability and indicates that these children may struggle more than peers with daily expectations, such as chores and day-to-day tasks. They may require more adult support and guidance than their peers to build these skills. Children, however, with Above Average cognitive ability and normative PS weakness did not show parent-reported adaptive struggles. This may suggest that stronger cognitive ability may help compensate for negative adaptive outcomes related to normatively weak PS skills.

Consistent with prior research (e.g., Braaten et al., 2020, Cook et al., 2019), academic weaknesses were more uniform across cognitive groups in children with normatively weak PS. At all cognitive levels, children with normative PS weakness had significantly lower scores on measures of reading and math. For reading, this relationship is strongest in children with Below Average and Above Average cognitive abilities. There was a 9.5 standard score point difference for children with Below Average cognition and slow PS and 5.6 standard score point difference for children with Above Average cognition. In math, the relationship was strongest for children with Below Average cognition. There was a difference of over 16.8 points for math between those with and without a normative PS weakness. For the Above Average group, there was an 8.5 standard score point difference for math. The differences in the Average group, while less robust, were also significant. There was a differences 3.8 standard score points for reading and 7.6 standard score points for math. These data suggest that children with normative PS weaknesses across cognitive levels are at risk for poor academic outcomes, particularly those with lower and higher cognitive abilities.

In our second aim, we hoped to expand on research by Thaler and colleagues (2012) and Calhoun and Mayes (2005, 2007), by exploring how a relative PS weakness in relation to cognitive ability (i.e., 15 pts or more) may impact functional outcomes. We previously highlighted that children with a normative PS weakness with Above Average cognitive ability showed more academic difficulty but they demonstrated relatively fewer adaptive weaknesses when compared to those without a normative PS weakness. In this further analysis in children with Above Average cognitive ability, a large relative PS weakness was also associated with statistically significant weakness in academic functioning. These children showed significantly more difficulties with both math calculation and word reading than children with Above Average cognitive ability without a relative PS weakness.

These data suggest that children with higher cognitive abilities and relatively weaker PS are at a greater disadvantage in meeting academic demands compared to peers without PS weakness. Results suggest that it may be important to consider this relative PS difference as a risk factor for poorer educational outcomes. In the post hoc analysis, however, the relative PS weakness group still had academic skills that are considered age-appropriate when compared to the larger normative population (Table 4). This relationship needs further exploration. Our measures of academic skills included basic reading and math calculation, both of which are untimed and do not necessarily require significant speed or efficient processing. This could be a reason why scores are generally still “age appropriate” but are impacted to a degree. Future studies should assess the relationship with measures that require more speed and efficiency, such as measures of math and reading fluency, timed written composition, and timed reading comprehension. Academic skills requiring more PS may show a more of a normative and clinically significant difference and, thus, more impairing to functioning.

There are clinical implications for these findings. Children with relative PS weakness with Above Average cognitive ability may require more accommodations, such as extended time or flexible scheduling, despite showing “intact” cognitive abilities. Our data hint at the possibility that the difference in academic outcomes could be considered a precursor to a mild learning disability, particularly as demands for efficient and fluent processing increase in later grades. When concretely considering criteria in the DSM-5, they would not be identified with a learning disability when compared to the normative population; however, there is a discrepancy in their performance compared to their potential. This idea of these “bright kids who can’t keep up” has been postulated by Braaten & Willoughby (2014) and require further consideration. This has implications for long-term outcomes, as they show academic underperformance but may have less access to treatment without a classification or diagnosis of a learning disability, particularly in the school setting. Conversely, research has suggested that children with psychiatric diagnoses (i.e., ADHD) and stronger PS show some evidence of better treatment outcomes than those with weak PS (Adalio et al., 2018). It would be important to explore the implications of PS strength in the context of cognitive ability as a potential protective factor for treatment in pediatric psychiatric populations.

Regarding psychological outcomes, no significant relationships were found between PS and broad measures of internalizing and externalizing psychological symptoms on the CBCL at any level of cognitive ability. When considering psychological symptoms dimensionally, this is consistent with our prior research (Braaten et al., 2020). In this prior study, we found that only inattention symptoms was a significant predictor of slow PS when using a norm-based definition of PS (index score <85) and simultaneously examining 8 symptoms of psychopathology as predictors. Dimensional measures of depression, anxiety, and hyperactive/oppositional symptoms were not associated with PS in that prior study. The difference between our results and prior research is likely explained by the different samples and methods used across studies. For example, Thaler and colleagues (2012) used a hierarchical agglomerative cluster analysis, with Wechsler Index scores as attributes in a sample of children with ADHD to discover a subsample of children with reduced PS. This group showed more inattention problems than a subsample characterized as having high average scores on the Wechsler. In the current paper, our objective was different. We aimed, in addition to exploring different operationalizations of PS weakness, to find correlates of PS weakness in domains beyond psychopathology (i.e., adaptive and academic functioning). Because of the abundance of analyses, we opted to use general measures of psychopathology (i.e., internalizing and externalizing problems). Thus, it could very well be that the “signal” for elevated inattention symptoms found in our previous work and in other research (Thaler et al., 2012 and Dickerson Mayes, 2000) was masked in the broad internalizing problem score. Future research should further explore the relationship between PS across cognitive abilities and symptoms of inattention.

Results should be considered with limitations in our data. First, we acknowledge that we used a singular index score as the measure of complex PS. Certainly, there are many advantages to using the PSI from the Wechsler IQ tests. It is based on a large, nationally representative normative sample and it is used regularly in clinical and school-based settings. However, it is important to examine the utility of normative and relative PS performance based on other measures of PS, as well as consider operationalizing PS across multiple tasks. Research has varied in their definition of PS, including measures of simple or complex reaction time (e.g., Motes et al., 2018), scanning (e.g., Tsourtos, Thompson, & Stough, 2002), sequencing/switching (e.g., Goth-Owens et al., 2010), verbal fluency/naming speed (e.g., Moll et al., 2017), and visual-motor speed (e.g., Brydges et al., 2018). The choice of PS measure impacts its relationship to other outcomes or measures across development, where some tasks are less sensitive or discriminant than others (Cepeda, Blackwell, & Munakat, 2013). These differences are important to explore further. If a measure of PS is less sensitive than others, it may be less helpful in clinical decision-making around diagnosis or treatment. Such issues have been raised by Rommelse, Luman, & Kievit (2020). In order to make meaningful conclusions on what and how PS impacts clinical populations, a more uniform definition must be developed, or specifically defined substrates of PS (including, but not limited to motor speed, verbal fluency, set-shifting, and reaction time) need to be identified. While our results do not directly address this issue, it is possible that there may be different sources of slow PS, particularly those with higher cognitive ability. For example, for some, it may be a matter of effort while for others slow PS may reflect a cautious approach. Additional work is needed to address potential heterogeneity within children with slow PS, as this would have implications for interventions.

Further, our study used a measure of verbal intelligence as a measure of general cognitive ability. As mentioned previously, this index score has been used in prior studies as an estimate of intelligence (Cook et al., 2019) and was selected for the current study to represent cognitive ability, as it does not include any timed measures. Also, verbal IQ has been found to be more predictive of academic outcomes than perceptual reasoning ability (Wechsler 2003, 2014). We did not use other global intelligence indexes (i.e., FSIQ, GCA, FRI, VSI) from the Wechsler tests because they all include a timed element. We wanted to ensure that measures tapping into PS were not calculated into intelligence estimates. We hoped that by using the VCI, we could look at the “purest” relationship between higher-order reasoning and PS on outcomes. The use of VCI, alone, however, does have its limitations. Other studies have included other tasks measuring nonverbal reasoning and intelligence (e.g, Thaler et al., 2012) and future studies should explore how this may impact the relationships with both normative and relative PS weaknesses.

Some other limitations should be considered. In our analyses, some of the participants in this sample were taking medication. We prioritized the generalizability of these results over limitations to a medication-naive sample and medication was generally controlled for in analyses. We note that our use of clinical diagnoses facilitated our ability to collect a large sample of consecutive referrals. We also note that the collection of our sample over time required the combination of DSMIV and 5 diagnoses, as well as select measures with different editions. In the current paper, such combinations were important to maximize statistical power and facilitate subgroup analyses, such as youth with different strata of VCI. Moreover, there is good evidence to support the overlap of the constructs. For example, Kappa agreements between DSM-IV and DSM-5 versions of ADHD and ASD in the literature are high (.75; Ghanizadeh, 2013 and .8; Wiggins et al., 2018, respectively). Correlations of key Wechsler components (VCI and PSI) of the WISC 5 and WAIS-IV are also high (~.8 Weschler, 2014b), as are the WIAT-II and WIAT III numerical operations subtests and the word reading subtests (~.8; Weschler, 2009). In addition to the relationship across versions, we note that individuals with different versions of diagnoses or measures are distributed within groups stratified for analyses. Nonetheless, we cannot rule out the possibility that some noise was introduced as a result of the combinations across versions, which in turn would have underestimated the strength of analyses that were found.

Overall, our study suggests that the potential impact of PS should be considered in the context of more global cognitive ability. In a large, clinically-referred sample of children and adolescents, both a relative and normative PS weakness, in relation to general cognitive ability, was significantly related to academic and adaptive outcomes.

Supplementary Material

Supplementary Table S1

Acknowledgements:

This research was supported by the Iacocca Family Foundation (EBB) and R01MH116037 (AED).

Footnotes

Deceleration of Statement of Interest: The authors report no conflicts of interest. Drs. Forchelli, Vuijk, Ward, Doyle, and Braaten, and Ms. Koven and Dews have no additional disclosures. Dr. Braaten serves on the board of Magination Press. She receives royalties from books published by Guilford Press Bright Kids Who Can’t Keep Up and The Child Clinician’s Report Writing Handbook and by Sage Publishing The Sage Encyclopedia of Intellectual and Developmental Disorders. Dr. Colvin receives travel support from Parent Project Muscular Dystrophy (PPMD) and also is a speaker honoraria from Tourette Syndrome Association.

References

  1. Achenbach TM, & Rescorla LA (2001). Manual for the ASEBA school-age forms &profiles. Burlington, VT: University of Vermont, Research Center for Children, Youth, & Families. [Google Scholar]
  2. Adalio C, Owens J, McBurnett E, Hinshaw B, & Pfiffner K. (2018). Processing speed predicts behavioral treatment outcomes in children with attention-deficit/hyperactivity disorder predominantly inattentive type. Journal of Abnormal Child Psychology, 46(4), 701–711. 10.1007/s10802-017-0336-z [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Ball K, Edwards JD, &Ross LA (2007). The impact of speed of processing training on cognitive and everyday functions. The Journal of Gerontology, Series B: Psychology Sciences and Social Sciences, 62(1), 19–31. 10.1093/geronb/62.special_issue_1.19 [DOI] [PubMed] [Google Scholar]
  4. Baune BT & Fuhr M. (2014). Neuropsychological functioning in adolescents and young adults with major depressive disorder—A review. Psychiatry Research, 218(3), 261–271. 10.1016/j.psychres.2014.04.052 [DOI] [PubMed] [Google Scholar]
  5. Benjamini Y, & Hochberg Y. (1995). Controlling the false discovery rate: A practical and powerful approach to multiple testing. Journal of the Royal Statistical Society: Series B (Methodological), 57(1), 289–300. 10.1111/j.2517-6161.1995.tb02031.x [DOI] [Google Scholar]
  6. Bobholz SA, Dabbs K, Almane D, Jones JE, Hsu DE, Strafstrom CE, Seidenberg M. & Hermann BP (2018). Neurobiological substrates of processing speed in childhood epilepsy. Brain Imaging and Behavior, 13, 1719–1725. 10.1007/s11682-018-0005-z [DOI] [PubMed] [Google Scholar]
  7. Bortolotti S, Zarantonello L, Uliana A, Vitturi N, Schiff S, Bisiacchi P, Avogaro A, Amodio P, & Maran A. (2018). Impaired cognitive processing speed in type 1 diabetic patients who had severe/recurrent hypoglycemia. Journal of Diabetes and Its Complications, 32, 1040–1045. [DOI] [PubMed] [Google Scholar]
  8. Braaten E, Ward A, Forchelli G, Vuijk P, Cook N, McGuinness P, & Doyle A. (2020). Characteristics of child psychiatric outpatients with slow processing speed and potential mechanisms of academic impact. European Child & Adolescent Psychiatry, 29(10), 1452–1464. 10.1007/s00787-019-01455-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Braaten E. & Willoughby B. (2014). Bright kids who can’t keep up. New York, NY: Guilford Press. [Google Scholar]
  10. Brydges C, Reid C, Campbell C, French N, & Anderson M. (2018). Executive functioning(fully) and processing speed (mostly) mediate intelligence deficits in children born very preterm. Intelligence, 68, 101–108. 10.1016/j.intell.2018.03.013 [DOI] [Google Scholar]
  11. Calhoun SL & Mayes SD (2005). Processing speed in children with clinical disorders. Psychology in the Schools, 42(4), 333–343. 10.1002/pits.20067 [DOI] [Google Scholar]
  12. Calhoun SL & Mayes SD (2007). Learning, attention, writing, and processing speed in typical children and children with ADHD, autism, anxiety, depression, and oppositional defiant disorder. Child Neuropsychology, 13(6), 469–493. 10.1080/09297040601112773 [DOI] [PubMed] [Google Scholar]
  13. Cepeda N, Blackwell K, & Munakata Y. (2013). Speed isn’t everything: Complex processing speed measures mask individual differences and developmental changes in executive control. Developmental Science, 16(2),269–286. 10.1111/desc.12024 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Cohen J. (1988) Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Erlbaum. [Google Scholar]
  15. Cook NE, Braaten EB & Surman CBH (2018) Clinical and functional correlates of processing speed in pediatric attention-deficit/hyperactivity disorder: a systematic review and meta-analysis. Child Neuropsychology 24(5), 598–616. 10.1080/09297049.2017.1307952 [DOI] [PubMed] [Google Scholar]
  16. Coyle T, Pillow D, Snyder A, & Kochunov P. (2011). Processing Speed Mediates the Development of General Intelligence (g) in Adolescence. Psychological Science, 22(10), 1265–1269. 10.1177/0956797611418243 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Doobay AF, Foley-Nicpon M, Ali SR, & Assouline SG (2014) Cognitive, adaptive, and psychosocial differences between high ability youth with and without autism spectrum disorder. Journal of Autism and Developmental Disorders, 44(8), 2026–2040. 10.1007/s10803-014-2082-1 [DOI] [PubMed] [Google Scholar]
  18. Gadow KD & Sprafkin JN (2002) Child symptom inventory- 4th edition: Screening and norms manual. Stony Brook, NY:Checkmate Plus. [Google Scholar]
  19. Gadow KD & Sprafkin J. (2013). Child &adolescent symptom inventory (5th ed.). Stony Brook, NY: Checkmate Plus. [Google Scholar]
  20. Ghanizadeh A. (2013). Agreement between diagnostic and statistical manual of mental disorders, fourth edition, and the proposed DSMV attention deficit hyperactivity disorder diagnostic criteria: an exploratory study. Comprehensive Psychiatry, 54(1), 7–10. 10.1016/j.comppsych.2012.06.001 [DOI] [PubMed] [Google Scholar]
  21. Goth-Owens T, Martinez-Torteya C, Martel M, & Nigg J. (2010). Processing speed weakness in children and adolescents with non-hyperactive but inattentive ADHD (ADD). Child Neuropsychology, 16(6), 577–591. 10.1080/09297049.2010.485126 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Haigh S, Walsh M, Mazefsky J, Minshew A, & Eack C. (2018). Processing speed is impaired in adults with autism spectrum disorder, and relates to social communication abilities. Journal of Autism and Developmental Disorders, 48(8), 2653–2662. 10.1007/s10803-018-3515-z [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Harrison P, & Oakland T. (2003). Adaptive behavior assessment system (2nd ed.). San Antonio, TX: Psychological Corporation. [Google Scholar]
  24. Harrison PL& Oakland T. (2015). Adaptive behavior assessment system (3rd ed.). Torrance, CA: Western Psychological Services. 10.1007/978-3-319-57111-9_1506 [DOI] [Google Scholar]
  25. Kelleher I, Murtaugh A, Clarke MC, Murphy J, Rawdon C, & Cannon M. (2013). Neurocognitive performance of a community-based sample of young people at putative ultra high risk for psychosis: support for the processing speed hypothesis. Cognitive Neuropsychiatry 18(1–2), 9–25. 10.1080/13546805.2012.682363 [DOI] [PubMed] [Google Scholar]
  26. Lee T, Wright M, Ames D, & Sachdev P. (2013). Genetic and environmental influences on processing speed, executive functions, and general cognitive ability: Findings from the older Australian twins study. Behavior Genetics, 43(6), 528–529. [Google Scholar]
  27. McGrath L, Braaten E, Doty N, Willoughby B, Wilson KH, O’Donnell E, Colvin M, Ditmars H, Blais J, Hill E, Metzger A, Perlis R, Willcutt E, Smoller J, Waldman I, Faraone S, Seidman L, & Doyle A. (2015). Extending the ‘cross-disorder’ relevance of executive functions to dimensional neuropsychiatric traits in youth. Journal of Child Psychology and Psychiatry, 57, 462–471. 10.1111/jcpp.12463 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Moll K, Göbel S, Gooch D, Landerl K, & Snowling M. (2016). Cognitive risk factors for specific learning disorder: Processing speed, temporal processing, and working memory. Journal of Learning Disabilities, 49(3), 272–281. 10.1177/0022219414547221 [DOI] [PubMed] [Google Scholar]
  29. Motes MA, Spence JS, Brier MR, Chiang HS, DeLaRosa BL, Eroh J, Maguire MJ, Mudar RA, Tillman GD, Kraut MA, & Hart J. (2018). Conjoint differences in inhibitory control and processing speed in childhood to older adult cohorts: Discriminant functions from a go/no-go task. Psychology and Aging, 33(7), 1070–1078. 10.1037/pag0000299 [DOI] [PubMed] [Google Scholar]
  30. Nettelbeck T, & Burns N. (2010). Processing speed, working memory and reasoning ability from childhood to old age. Personality and Individual Differences, 48(4), 379–384. 10.1016/j.paid.2009.10.032 [DOI] [Google Scholar]
  31. Niarchou M, Zammit S, Walters J, Lewis G, Owen M, & Van Den Bree M. (2013). Defective processing speed and nonclinical psychotic experiences in children: Longitudinal analyses in a large birth cohort. American Journal of Psychiatry, 170(5), 550–557. 10.1176/appi.ajp.2012.12060792 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Rommelse N, Luman M, & Kievit R. (2020). Slow processing speed: a cross-disorder phenomenon with significant clinical value, and in need for further methodological scrutiny. European Child & Adolescent Psychiatry, 29, 1325–1327. 10.1007/s00787-020-01639-9 [DOI] [PubMed] [Google Scholar]
  33. Santegoeds E, van der Schoot E, Roording-Ragetlie S, Klip H, & Rommelse N. (2021). Neurocognitive functioning of children with mild to borderline intellectual disabilities: Profile characteristics and predictors of behavioural problems. Journal of Intellectual Disability Research, 10.1111/jir.12874 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Shanahan MA, Pennington BF, Yerys BE, Scott A, Boada R, Willcutt EG, Olson RK, & DeFries JC (2006) Processing speed deficits in attention deficit/hyperactivity disorder and reading disability. Journal of Abnormal Child Psychology, 34(5),585–602. 10.1007/s10802-006-9037-8 [DOI] [PubMed] [Google Scholar]
  35. Thaler NS, Bello DT, & Etcoff LM (2012). WISC-IV profiles are associated with differences in symptomatology and outcome in children with ADHD. Journal of Attention Disorders, 17, 291–301. 10.1177/1087054711428806 [DOI] [PubMed] [Google Scholar]
  36. Thorsen AL, Meza J, Hinshaw S, & Lundervold AJ (2018) Processing speed mediates the longitudinal association between ADHD symptoms and preadolescent peer problems. Frontiers in Psychology, 8, 2154. 10.3389/fpsyg.2017.02154 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Tsourtos G, Thompson JC, & Stough C. (2002) Evidence of an early information processing speed deficit in unipolar major depression. Psychological Medicine, 32, 259–65. 10.1017/s0033291701005001 [DOI] [PubMed] [Google Scholar]
  38. Wallach A, Walts M, Casper TC, Aaen G, Belman A, Benson L, Chitnis T, Gorman M, Graves J, Harris Y, Lotze TE, Mar S, Moodley M, Ness JN, Rensel M, Rodrigues M, Rose JW, Schreiner T, Tillema J-M. Waubant E, Weinstock-Guttman B, Charvet LE & Krupp LB (2020). Cognitive processing speed in pediatric-onset multiple sclerosis: Baseline characteristics of impairment and prediction of decline. Multiple Sclerosis Journal, 26(14) 1938–1947. 10.1177/1352458519891984 [DOI] [PubMed] [Google Scholar]
  39. Wechsler D. (2004). Wechsler intelligence scale for children (4th ed). London, English: Pearson. [Google Scholar]
  40. Wechsler D. (2005). Wechsler individual achievement test (2nd ed). London, English: Pearson. [Google Scholar]
  41. Wechsler D. (2009). Wechsler individual achievement test (3rd ed). London, English: Pearson. [Google Scholar]
  42. Wechsler D. (2008). Wechsler adult intelligence scale (4th ed). San Antonio, TX: Pearson. [Google Scholar]
  43. Wechsler D. (2014a) Wechsler intelligence scale for children (5th ed). San Antonio, TX: Pearson. [Google Scholar]
  44. Wechsler D. (2014b). WISC-V technical and interpretive manual. Bloomington, MN: Pearson. [Google Scholar]
  45. Wiggins L, Christensen D, Van Naarden Braun K, Martin L, & Baio J. (2018). Comparison of autism spectrum disorder surveillance status based on two different diagnostic schemes: Findings from metropolitan Atlanta developmental disabilities surveillance program, 2012. PLoS One, 13(11). 10.1371/journal.pone.0208079 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Table S1
NIHMS1736408-supplement-Supplementary_Table_S1.docx (12.6KB, docx)

RESOURCES