Keywords
autism spectrum disorder - executive function - language - intervention
Learning Outcomes: As a result of this activity, the reader will be able to (1) explain the relationship
between executive functions and language in ASD; (2) describe how targeted interventions
may impact executive function or language performance in ASD; and (3) explain how
various executive function task features and inclusion of different comparison groups
may yield distinctive findings.
Autism spectrum disorder (ASD) is a neurodevelopmental disorder that is behaviorally
diagnosed and is characterized by deficits in social communication and the presence
of restricted and repetitive patterns of behavior.[1] Recent estimates from the Centers for Disease Control and Prevention indicate that
this disorder affects approximately 1 in 59 children.[2] Children, adolescents, and adults with ASD often exhibit deficits in various aspects
of executive function, although numerous review articles and meta-analyses indicate
variability in executive function performance in this population, depending on the
tasks and comparison groups included in the study.[3]
[4]
[5]
[6]
[7] Executive functions are defined as cognitive processes that underlie goal-directed
behavior, and include skills such as shifting or cognitive flexibility, inhibition,
and working memory, as well as higher-level functions such as planning.[8]
[9]
[10] Executive functions have a vast impact on daily life, and are closely related to
the ability to communicate with others.[3]
[11]
[12]
[13] Understanding the relationship between communicative skills and executive function
skills is important because both domains have impacts on learning, scholastic achievement,
and social development.[14]
[15]
[16] Additionally, various theories have proposed that language mediates executive functions,
while others have proposed that executive functioning underlies communication skills.[11]
[17]
[18] This article reviews empirical studies that examined executive functions (e.g.,
shifting, inhibition, working memory, and planning) and language (e.g., semantics,
syntax, and pragmatics) in ASD, and discusses clinical implications drawn from existing
evidence.
Language Overview in Autism Spectrum Disorder
Language Overview in Autism Spectrum Disorder
Social communication impairments are a core characteristic and required to obtain
a behavioral diagnosis of ASD. Pragmatic language is one component of social communication
and includes a variety of skills. Children with ASD have challenges with topic maintenance,
greetings and farewells, appropriate turn taking, and using conversational repair
strategies. These deficits are associated with challenges with perspective-taking
and theory of mind.[19] Due to the pervasiveness of pragmatic language deficits, this area has been extensively
studied in ASD.[20]
[21]
[22] Children with ASD are impaired in their pragmatic skills compared with peer groups
with typical development (TD), intellectual disability, or other disorders.[23]
[24]
[25]
While deficits in pragmatics and social communication are a hallmark feature of ASD,
structural language abilities, specifically semantics and syntax, vary and result
in marked individual differences. Several studies indicate a discrepancy in expressive
and receptive vocabulary scores, such that expressive vocabulary scores are higher
than receptive vocabulary scores.[26]
[27]
[28] Other studies have noted no difference, or reported higher receptive vocabulary
scores than expressive vocabulary scores, and the discrepant findings appear to depend
on the measures and ages of participants.[28]
[29]
[30] In addition to deficits in standardized vocabulary scores, children with ASD present
with unique deficits in specific subdomains of vocabulary, including the production
and comprehension of personal pronouns, mental state terms, and prepositions.[23]
[31]
[32] These subdomains of vocabulary may be particularly impacted because they require
perspective-taking, which is often a challenge for individuals with ASD. For example,
use and understanding of mental state terms, which refer to one's emotional or cognitive
state, require theory of mind.[33] Similarly, correct use of personal pronouns requires the speaker to know if the
communication partner is aware of the referent. If not, a communication breakdown
can occur.
Within the domain of syntax, children and adolescents with ASD are variable in their
deficits. Several studies have suggested that children with ASD exhibit syntactic
delays in comparison to peers with intellectual disability or TD.[34]
[35]
[36]
[37] These syntactic abnormalities can include challenges with verb phrases, regular
and irregular past tense, present tense, and regular third person singular verbs.[35]
[36] However, like deficits in semantics, challenges in this domain of language are not
universal among individuals with ASD; several studies have reported intact syntactic
skills in individuals with ASD when compared with TD peers.[38] In fact, several studies have proposed that there exists a subgroup of children
with ASD who have grammatical impairments that mirror those of children with developmental
language disorder (DLD), which is a disorder characterized by deficits in language
without a known etiology.[39]
[40] However, general patterns have indicated that while some children with ASD have
challenges with certain aspects of syntax, this is not true for all children with
ASD.
Executive Function Overview in Autism Spectrum Disorder
Executive Function Overview in Autism Spectrum Disorder
Similar to structural language, many children and adolescents with ASD have difficulties
with executive functions, although unlike pragmatic language, this is not a hallmark
deficit of this population. Many studies have found that executive functions are generally
impaired at the group level in children and adolescents with ASD, particularly in
comparison to peers with TD,[41]
[42]
[43] although this is not always the case.[44] It is important to be mindful of individual differences, though, as some children
with ASD do not exhibit deficits in executive function, indicating that these deficits
are not a core impairment in this population.[45] It is also particularly difficult to draw conclusions about children with ASD because
studies vary in group matching criteria, as well as inclusionary and exclusionary
criteria. However, some studies that have conflicting findings report similar ages
for their participant groups, as well as similar matching criteria, comparison groups,
and covariates.[42]
[43]
[44] Two common comparison groups are children with attention-deficit/hyperactivity disorder
(ADHD) and children with DLD. Specifically, comparisons between ASD and ADHD are frequent
because these two disorders share phenotypic overlap, and understanding their respective
executive function skills can shed light on similarities and differences in these
groups. On the other hand, comparisons between children with ASD and children with
DLD inform the relationship between language skills and executive functions. These
groups are commonly compared when examining language development in ASD.
Executive function domains, as noted earlier, include shifting, inhibition, and working
memory (commonly referred to as updating). These domains, although related, are separable
and unique constructs, as indicated by a factor analysis completed by Miyake and colleagues.[9] We will define these domains and summarize the literature on executive function
skills in children and adolescents with ASD. We delve into more detail for the domain
of planning, which is regarded as a higher-level cognitive skill, given that it is
inherently more complex than shifting, inhibition, and working memory.[9] Notably, some methods used to measure executive functions are verbal or linguistic
in nature, while others are nonverbal or do not have a linguistic component. We will
identify the verbal or linguistic nature of the tasks, as this feature may impact
the interpretation of the findings. After examining executive functions in ASD, we
will discuss studies that have specifically linked shifting, inhibition, working memory,
and planning to language abilities.
Shifting. Shifting is the ability to switch between mental states or tasks and is also commonly
referred to in the literature as cognitive flexibility. There is some evidence that
shifting is impaired in ASD in comparison to peer groups with DLD[46] and ADHD,[42]
[43] as well as TD peers.[47] However, other studies have not found differences in shifting between children with
ASD and children with TD, intellectual disability, or ADHD.[44]
[48] Several studies have used the Wisconsin Card Sorting Task (WCST[49]) to examine shifting. In this task, participants are expected to shift between sorting
cards based on one of three dimensions, including color, number, and shape. Although
the WCST does not require a verbal output, linguistic skills can help the participant
to verbally encode and rehearse the shapes or colors of the targets for sorting. A
perseverative error occurs if an individual does not shift to the new sorting expectation
and responds based on the previous sorting expectation. Children with ASD tend to
make more of these perseverative errors during this task in comparison to peer groups
with TD, ADHD, and DLD.[46]
[50] For example, Liss and colleagues[46] compared shifting abilities between school-age children with high-functioning ASD
and children with DLD matched on IQ, age, and socioeconomic status. Notably, the children
with DLD had an average verbal IQ score that was 10 points higher than the children
with ASD. Shifting was measured using the WCST. They found that the children with
ASD made more perseverative errors on the WCST, but the difference in perseverative
errors was no longer significant when accounting for verbal IQ. The authors ultimately
proposed that there were no unique shifting deficits in ASD, and that verbal IQ played
a key role in shifting performance.
A 2014 meta-analysis[5] of shifting focused on reported data from a large sample of individuals with ASD
and TD. Across a range of performance measures, the authors found that participants
with ASD and TD were not different on shifting performance overall, although they
did find a difference between groups on the Shift subscale of the Behavior Rating Inventory of Executive Function (BRIEF[51]
[52]). The authors of this meta-analysis advise that performance measures of cognitive
flexibility may lack ecological validity, as these measures were not as sensitive
as the BRIEF in distinguishing between clinical and control groups. Overall, they
suggest that their results are not surprising given that cognitive inflexibility,
or difficulty with shifting, is not a core characteristic of ASD. However, the authors
reported that the majority of the studies included in the meta-analysis had compared
individuals with ASD to age-matched controls, but the authors did not consider group
matching status in their inclusionary criteria.[5] Therefore, results from this meta-analysis may not generalize to all individuals
with ASD; comparison groups should be considered in future work.
Inhibition. Response inhibition has been a focus of several studies that compare participants
with ASD to various peer groups, often participants with ADHD and TD. This aspect
of executive function involves suppressing information that may interfere with the
task goal at hand. Like results for shifting, findings related to response inhibition
in children and adolescents with ASD are varied. Some studies have reported no differences
between children with ASD and their peer groups with ADHD and TD,[44]
[48]
[53] while others have reported marked impairments in this domain.[42]
[43]
Several studies have compared verbal response inhibition performance in children with
ASD to that in children with TD. For example, Yang and colleagues[44] examined this skill using the Stroop task[54] in children with ASD (mean age = 8.1 years) in comparison to children with TD (mean
age = 8.0 years) and children with ADHD (mean age = 8.2 years). Children were all
statistically similar on age, and nonverbal IQ scores were controlled in their analyses.
The Stroop task requires participants to provide a verbal output while suppressing
other verbal information. For example, in this task the participant may see the word
BLUE printed in green ink, and they must name the color of the ink instead of the
printed word. The authors found no significant differences between the three groups
in performance on the Stroop task as a measure of inhibition.[44] Corbett and colleagues[42] examined response inhibition in similar groups of children, including children with
high functioning ASD (mean age = 9.44 years), TD (mean age = 9.56 years), and ADHD
(mean age = 9.40 years). They controlled for IQ in their analyses, and measured inhibition
with various tasks, including one verbal task that was similar to the Stroop test.
On the Stroop-like task, they found that children with ASD performed worse on measures
of inhibition compared with the children with TD, but they performed similarly with
the children with ADHD. There were no differences between the children with ADHD and
the children with TD on this task. The authors also measured response inhibition with
a task in which participants pressed a button when they saw or heard a number “1”
but not a number “2.” Thus, this task did not require a verbal output, although the
input was linguistic in nature. Children with ASD performed worse than children with
TD on both tasks, but they only performed worse than children with ADHD on the visual
number presentations.[42]
One study by Johnston and colleagues[55] examined verbal response inhibition in adults with ASD (mean age = 27.8 years) compared
with adults with ADHD (mean age = 27.3 years), and TD adults (mean age = 27.3 years).
The groups were matched on verbal IQ (all participants had verbal IQ scores within
the normal range), age, and gender. The authors used verbal tasks, including the Stroop
task[54] and the Hayling Sentence Completion test,[56] which requires participants to complete a sentence using either a relevant, connected
word, or an irrelevant, unconnected word. Participants with ASD had typical response
inhibition but slow response latencies. The authors suggested that this was likely
due to deficits in response initiation. That is, adults with ASD were slower and more
accurate, while adults with ADHD were more impulsive in the manner in which they completed
the task.[55]
Working memory. Updating of working memory is the ability to revise recalled information as task
goals update or change.[14] Similar to the domains of shifting and inhibition, results from working memory tasks
are variable: some studies have reported that children with ASD are impaired when
compared with peer groups such as children with TD and children with ADHD,[42]
[43] while others have reported no differences in working memory performance between
individuals with ASD and peer groups with TD, ADHD, and intellectual disability.[43]
[44]
[48]
Roelofs and colleagues[48] did not find differences in working memory among adolescents and young adults with
ASD and intellectual disability, compared with age- and IQ-matched individuals without
ASD. Updating of working memory was measured using a verbal fluency task and a verbal
backward digit span test. Kado and colleagues[43] reported similar working memory performance between children with ASD and children
with ADHD, but both groups had impaired working memory compared with chronological
age- and IQ-matched school-age children with TD. In this study, the authors used a
modified Japanese version of the WCST[49] which included fewer cards.[57] The test was purported to measure aspects of inhibition, shifting, and working memory.
They presented several scores for this task, including the number of sorted sets of
six consecutive cards. The authors found that children with ASD and ADHD sorted fewer
sets and had a higher number of total errors than the children with TD, indicative
of inadequate working memory.[43]
To tease out verbal demands of executive function tasks, some studies deliberately
separate working memory into two domains: spatial or visual working memory and verbal
working memory. For example, Hill and colleagues[58] assessed working memory in 5- to 8-year-old children with ASD and children with
DLD. The children with ASD were divided into two groups: children who had co-occurring
language impairment and children who had typical or normal language abilities. They
found that children with ASD and language impairment performed worse on verbal working
memory tasks than the children with ASD and normal language. The children with ASD
and language impairment and children with DLD performed similarly on most verbal working
memory tasks except for a nonword repetition task, in which the children with DLD
produced more errors. However, there were no group differences on the visual working
memory tasks, indicating that the increased demands associated with a verbal working memory task may impact performance and differentiate ASD groups with and
without language impairment.[58]
Planning. Planning is considered to be a higher-level cognitive process, and requires sequences
of planned actions to be monitored and updated when given additional knowledge or
when a preceding action occurs.[6] Generally, children with ASD demonstrate marked deficits in planning compared with
TD peers, as well as peer groups with ADHD and Tourette's syndrome,[47]
[50]
[59] although children with ASD may have similar planning skills to children with a language
disorder.[46]
Kimhi et al noted differences in planning ability between preschool children with
ASD and chronological- and mental age–matched children with TD.[47] Older children with ASD may also exhibit deficits in planning, as a longitudinal
study examining executive function in adolescents with ASD relative to adolescents
with learning disability found that adolescents with ASD have deficits in planning
compared with the children with TD; notably, these deficits did not improve over a
3-year time period.[59] In contrast to these findings, Liss and colleagues did not find differences between
boys with ASD and boys with DLD on the Mazes subtest from the Wechsler Intelligence
Scale for Children—Revised, which is a task that measures aspects of planning, shifting,
and inhibition.[46] Also, Corbett and colleagues[42] did not find significant differences in planning between children with high functioning
ASD, children with ADHD, and children with TD when they used the Cambridge Neuropsychological
Test Automated Battery (CANTAB) Stockings of Cambridge task, which is a spatial planning
task.
Generally, it appears that some individuals with ASD exhibit deficits in areas of
executive functions, but findings seem to be dependent on several factors. First,
outcomes may be dependent on the ages of the participants. For example, studies that
include adolescents or adults generally find that participants with ASD do not perform
differently than TD peers or peers with ADHD or intellectual disability,[44]
[48]
[53] whereas studies that only include school-age or younger participants are more likely
to reveal differences.[42]
[46] Second, verbal and nonverbal tasks can yield differences, such that tasks which
require verbal responses generally differentiate groups, while tasks that are more
visually based do not.[58] This pattern of findings suggests that executive function skills by themselves may
not impact in ASD, but group differences may emerge when executive function tasks
include verbal or linguistic demands. While language deficits in the domains of vocabulary
and syntax are not hallmark features of ASD, some children do experience delays in
these domains.[39] Therefore, it is important to include language-matched groups of children if selecting
executive function tasks that require verbal output. For example, Hill and colleagues
reported that groups of children with ASD and DLD matched on language ability performed
similarly on verbal working memory tasks,[58] highlighting the importance of language matching.
Many studies, however, do not include language-matched groups of children. Rather,
groups are often matched on chronological age, mental age, or IQ.[43]
[46]
[48]
[53] Selection of comparison groups and matching criteria are important to consider.
For instance, comparing children with ASD to children with DLD may yield different
results than comparing children with ASD to children with ADHD or other intellectual
and developmental disabilities. A comparison of age- and IQ-matched children with
ASD and DLD suggests similarities in inhibition, and shifting, and planning.[46] Age-matched children with ASD and children with DLD do not differ in performance
on visual working memory tasks, and children with ASD + language impairment and children with
DLD perform similarly on verbal working memory tasks.[58] In comparison to age- and IQ-matched children with ADHD, children with ASD either
perform worse[42]
[43] or similarly[44] on shifting, inhibition, and working memory measures. Additionally, comparisons
of individuals with ASD + intellectual disability and individuals with intellectual
disability without ASD do not yield differences in shifting, inhibition, or working
memory.[48] Ultimately, these findings suggest that, when controlling for linguistic abilities
and task demands, individuals with ASD are not likely uniquely experiencing executive
dysfunction. Furthermore, language skills, along with comorbid diagnoses, such as
ADHD, must be considered when assessing and targeting executive functions in children
with this heterogeneous disorder.
To elucidate the interplay between executive functions and language, several studies
have explored the relationship between these two domains in children and adolescents
with ASD. We will review studies that have examined executive functions in relation
to vocabulary, syntax, and/or pragmatic abilities.
The Relationship Between Executive Function and Language in ASD
The Relationship Between Executive Function and Language in ASD
Studies examining executive functions and language in children with TD have found
correlations between these two domains. One particularly relevant model for conceptualizing
the relationship between executive functions and language is the Hierarchical Competing
Systems Model (HCSM).[18] This model suggests that there is a habit system which is influenced by previous
experience, and a representational system which is influenced by conscious reflection
on behavior. These two systems are hierarchically arranged, and they compete to guide
goal-directed behavior; thus, the representational system can override the habit system.
While the habit system may lead an individual to be influenced by perceptual information,
the representational system, or language-guided reflection, overrides perceptually
driven behaviors in favor of more mature goal-directed behavior. This model serves
to explain the relationship between executive functions and language throughout the
lifespan.
Vocabulary. Given the HCSM's claims regarding the importance of verbal ability for executive
function performance,[18] it is probable that vocabulary, in particular, is important. Cascia and Barr[60] reported on expressive and receptive vocabulary in relation to executive functions
in children and adolescents with ASD, ranging in age from 9 to 18 years (mean age = 14.65
years). They found that receptive and expressive vocabulary standard scores were related
to parent and teacher ratings of executive functions. Specifically, expressive and
receptive vocabulary scores were related to parent ratings of attention, emotion regulation,
and inhibitory control. Based on the teacher ratings, receptive vocabulary was related
to attention and organization, and both expressive and receptive vocabulary scores
were related to emotion regulation, flexibility, inhibitory control, initiation, and
self-monitoring. The goal of their study was to examine how executive function mediates
the relationship between vocabulary and empathy; they found that both vocabulary and
executive functioning abilities were important for empathy. The authors discussed
how more advanced vocabulary can support social interactions. The same logic/argument
can be extended to executive functioning skills, such that better executive functioning
ability can support social interactions. Therefore, not only are vocabulary and executive
functions related to one another, but they both underlie important social outcomes.[60]
Joseph and colleagues[41] studied executive function and lexical skills in 34 school-age children with ASD
and 31 children without ASD who were matched on age, verbal IQ, and nonverbal IQ.
Children with ASD had poorer expressive and receptive vocabulary raw and standard
scores. Unlike the comparison group, vocabulary scores for the children with ASD were
not correlated to measures of inhibition or planning. Moreover, language and working
memory were not related in either group. The authors posited that children with ASD
did not use their language to internally maintain rules or to verbally encode steps
to complete the tasks. These findings may suggest that children with ASD present with
a deficit in verbal mediation, or the ability to use language skills internally to
facilitate accomplishing goal-directed behavior.
To examine lexical processing skills in relation to executive functions, Haebig et
al[61] used a lexical decision task and executive function tasks that targeted shifting
and updating. Groups included 30 TD children, 27 children with ASD, and 28 children
with DLD. The groups were matched on receptive vocabulary growth scores. The authors
found that both shifting and updating abilities predicted accuracy and reaction time
on the lexical processing task. The authors posited that children, regardless of diagnosis,
process language with the aid of nonlinguistic mechanisms. Therefore, they argue that,
consistent with the HCSM, there exists a relationship between language and executive
functions, such that executive function supports language processing. The directionality
of this relationship cannot be determined from this study, however, given its cross-sectional
design.[61]
Syntax. The HCSM also leads to the hypothesis that better syntactic skills would benefit
performance on executive function tasks. Some studies have used the Clinical Evaluation
of Language Fundamentals (CELF[62]) to examine expressive and receptive language ability, including syntactic ability.
For example, Akbar and colleagues assessed children with ASD between the ages of 8
and 19 years (76% male). Children completed the CELF, Fourth Edition[62] to obtain a receptive, expressive, and core language score. The Children's Communication
Checklist[63] was used as a parent report measure of language. The authors assessed working memory,
organization, shifting, and inhibition using the Delis-Kaplan Executive Function System
(DKEFS[64]), the Developmental Neuropsychological Assessment (NEPSY-II[65]), and the Wechsler Intelligence Scales For Children (WISC[66]). The CELF-4 core language score was a significant predictor of working memory.
Nonverbal IQ scores predicted organization and shifting; ASD severity predicted shifting
as well. There were no predictors for inhibition. The authors suggested that working
memory deficits in ASD may be the result of weak language ability, as working memory
is also a deficit for kids with DLD. The authors also proposed that this connection
between working memory, syntax, and semantics could be due to a limitation in the
ability to develop and utilize verbal mediation strategies during these executive
function tasks.[11]
Ellis Weismer and colleagues[67] also used the CELF-4[62] to specifically look at associations between receptive and expressive language scores
and executive functions (i.e., shifting, inhibition, and updating of working memory).
Their participant groups included children with TD and children with ASD, who were
all between the ages of 8 and 12 years. There were several associations between receptive
and expressive language skills and executive functions in children with ASD. However,
when the children with ASD were split into groups with typical language and language
impairment, receptive language abilities were associated with different executive
function domains. In the group with language impairment, receptive language was correlated
with shifting; in the group with normal language, receptive language was correlated
with inhibition. The authors proposed that the association between language and executive
functioning skills support the HCSM, but they caution that this finding does not answer
the question of directionality.[67]
Durrleman and Delage[32] studied 5- to 16-year-old children and adolescents with ASD (mean age = 9; 07) and
DLD (mean age = 9; 07) compared with age-matched controls and younger controls matched
on expressive grammar abilities. They assessed expressive grammar and specifically
examined pronoun production. They used a backward digit span task to assess working
memory. When they controlled for nonverbal reasoning, they found correlations between
third person pronoun production and backward digit span for both the children with
ASD and the children with DLD; however, they did not look at these relationships in
the children with TD. The authors suggested that the production of complex grammar
recruits working memory ability in children with ASD and DLD.[32]
Pragmatic language. Widespread and pervasive deficits in pragmatic language skills in children and adolescents
with ASD make this area of language particularly interesting and important to study.
Several studies have reported connections between pragmatic language and executive
functions. For instance, in addition to studying the relationship between executive
functions and structural language, Akbar and colleagues used the pragmatic judgment
subtest of the Comprehensive Assessment of Spoken Language (CASL[68]) to examine the relationship between executive functions and pragmatic language
in their group of 8- to 19-year-old participants with ASD. They found that the pragmatic
judgment score of the CASL predicted working memory in children with ASD as measured
by a letter and number sequencing task, but it did not predict other aspects of executive
functions, including shifting and inhibition.[11] The authors proposed that syntactic skills and pragmatic skills both contribute
to the ability to perform well on tasks of working memory.
Other studies have examined the expressive pragmatic skills in children and adolescents
with ASD. For example, McEvoy and colleagues[21] assessed preschool-age children with ASD, developmental delay, and TD. Children
completed a battery of executive function tasks, and the Early Social Communication
Scales[69] was used to measure joint attention and social interaction. The authors found that
the children with ASD had lower scores in social interaction compared with the other
two groups. The children with ASD were also not able to flexibly change sets on the
spatial reversal task, and they had more perseverative errors than the two other groups,
meaning that the children with ASD persisted in using a previously reinforced strategy
even though this strategy no longer resulted in a reward. In addition, a significant
relationship was observed between social interaction behaviors and executive functioning
skills across groups. The social interactions had specific goals (e.g., rolling a
ball back and forth with the examiner), and the ability to maintain and achieve these
goals may be reflected by the relationship between social interaction behaviors and
executive function skills. Finally, there was a relationship between social interaction
behaviors and joint attention, suggesting that this relationship may account for the
link between social interaction and executive functions, as joint attention was also
related to perseverations on the spatial reversal task.[21]
Gilotty and colleagues[13] examined executive function skills in relation to adaptive skills in children and
adolescents with ASD, ranging in age from 6 to 17 years. They found relationships
between executive functioning skills and the communication and socialization domains
of the Vineland Adaptive Behavior Scales. Specifically, poor communication domain
scores and social domain scores were related to less initiation and worse working
memory skills. This supports the hypothesis that working memory and aspects of language
and communication are associated in this population.[13] Pugliese et al[70] reported similar findings, such that parent reports of executive functioning skills
predicted socialization and communication scores on the Vineland Adaptive Behavior
Scales (VABS[71]) in school-age children with high-functioning ASD.[70]
In sum, language abilities (overall language and pragmatic language, specifically)
appear to be particularly important for working memory in this population.[11]
[13]
[32]
[67] It is also possible that working memory underlies performance on language-based
tasks, such as a lexical decision task.[61] Findings regarding the relationship between executive functions and expressive and
receptive vocabulary knowledge appear to be mixed.[41]
[60] Given these findings, and the premise of the HCSM, executive function and language
certainly appear to be related; however, it is difficult to determine the directionality
of the relationship between these two skills. Regardless, deficits in these domains
in children with ASD warrant further research, particularly in how clinical practice
can improve these two skills.
Impact of Intervention
The relationships between executive functions and domains of language ability (e.g.,
vocabulary, syntax, and pragmatics) in ASD raise questions surrounding the impact
of intervention. Specifically, if structural and/or pragmatic language skills are
in fact associated with executive functions, one could argue that an intervention
that targets one domain would lead to generalized improvements in the other domain.
The HCSM would purport that improvements in language ability would lead to improvements
in executive functions, although results reported by Haebig and colleagues[61] would imply the opposite, such that executive function would support language performance.
In line with the HCSM, a few studies have looked at executive functions as an outcome
after administering a social-based pragmatic intervention.
Stichter and colleagues examined the effectiveness of a social competence intervention
for elementary school students with high-functioning ASD.[72] They used the Social Competence Intervention-Elementary (SCI-E[72]) program for participants with ASD between 6 and 10 years of age (mean age = 8.77
years). IQ scores ranged from 72 to 124. This intervention targeted social communication
areas including turn-taking, recognizing feelings and facial expressions in others,
sharing ideas, and problem solving. They obtained several measurements pre- and postintervention,
including the parent-report and teacher-report Social Responsiveness Scale[73] as a measure of social functioning, and tasks that measured emotion recognition,
executive functions, and theory of mind. Executive functions were measured through
parent report and direct assessment. The intervention resulted in improvements in
theory of mind, as well as better problem-solving abilities as measured through the
direct assessment. Parent perceptions of social functioning and executive functions
also increased.[74]
Morgan and colleagues conducted a cluster randomized trial to examine the effects
of classroom Social, Communication, Emotional Regulation, and Transactional Support
(SCERTS[75]) intervention in comparison to a school-based education that included ASD training
modules. Specifically, they were interested in executive functions and social outcomes,
in addition to active engagement in the classroom and adaptive behaviors, for children
with ASD. One hundred eighteen children with ASD were in the classroom SCERTS intervention
group (mean age = 6.82 years), and 79 children were assigned to the ASD training modules
group (mean age = 6.77 years). Participants in these groups were not significantly
different on age, race, ethnicity, gender, ASD severity, or pretest IQ measures. They
found that the group receiving the classroom SCERTS intervention had better outcomes
in social skills and executive functions. One explanation for these findings is that
the SCERTS model is inherently social, in that the main goal is to increase the student's
engagement and social communication. However, it is important to note that the outcome
measures for social skills and executive functions were teacher-report scales, which
may have biased the results as the teachers knew which treatment they were administering.[76]
One study examined an executive function intervention and its impact on social skills.
Kenworthy and colleagues[77] conducted a randomized control trial of Unstuck and On Target (UOT), which is an
executive function intervention. Forty-seven children with ASD were assigned to the
UOT intervention, while 20 children with ASD were assigned to a social skills intervention.
Children were between 7 and 11 years old and all had IQ scores above 70. The children
who received the UOT intervention had greater improvements in domains of executive
function, including flexibility and planning, compared with the children who received
the control social skills intervention, although both groups demonstrated improvements
on pre- and postintervention assessments. Children in both groups also made improvements
in social skills, indicating that treatments targeting executive function may also
help pragmatic language skills.[77]
Clinical Implications
Extending the complex and often contradictory findings of both empirical and clinical
research to practice is difficult given the heterogeneity of ASD. This is also complicated
by the differences between studies in terms of the tasks used for assessment, particularly
the use of verbal- or visual-based tasks; the comparison groups selected, including
children with ADHD, DLD, or TD; and the methods used for group matching. Specifically,
groups matched on language do not yield differences in executive function performance
between children with TD and ASD,[61] but matching on IQ or age inconsistently yields differences between groups.[42]
[43]
[44]
[46]
[48]
[53] Despite the variability in findings, some key patterns do emerge from these studies.
Social communication-based interventions appear to have a positive impact on executive
function. Several studies that targeted social skills in elementary school children found improvements
in executive functions after treatment. One theory for this is that children in social-based
interventions learn to self-reflect, and they also learn to observe the behaviors
of peers. If children strengthen their ability to observe and understand their own
behaviors and the behaviors of their peers, it is likely that they will use these
self-reflection skills during executive function tasks. If verbal mediation does play
a role in executive function tasks, then perhaps this skill is enhanced during a social-based
intervention, and increased practice translates to executive functions. Conversely,
just as social-based interventions can result in improvements in executive functions,
there is evidence that executive function-based intervention can result in improvements
in social skills.[77]
Structural language ability is important for performance on tasks of executive functions. Connections between structural language (lexical and syntactic skills) and executive
functions support the HCSM, such that language guides reflection on behavior and ultimately
impacts performance on these cognitive tasks. For instance, several studies note that
working memory and syntax are related[11]
[32] and lexical abilities appear to be important for shifting and inhibition.[60]
[61] These relationships support the hypothesis that the ability to use linguistic skills
to drive one's behavior is important for tasks of executive functions. Thus, if this
hypothesis is correct, the clinical implications are that focusing on the development
of vocabulary and syntax may ultimately benefit children with ASD both academically
and socially. However, further research is needed to test this hypothesis.
Consideration of ASD symptom severity. ASD symptomatology and severity impacts the presence of both restricted and repetitive
behaviors as well as social deficits, which are the two core deficits in ASD. Several
studies have found connections between executive function and ASD symptomatology and/or
severity. Thus, ASD symptomatology and/or severity is an important factor to consider
in both research and clinical practice. However, it is challenging to disentangle
these features, and it is not clear if increased severity impacts performance on executive
function tasks, or if impairments in executive function result in presentation of
more severe symptoms. For example, children with ASD tend to make more perseverative
errors on tasks measuring shifting in comparison to children with DLD or ADHD,[46]
[50] and these errors are potentially positively correlated with ASD severity and symptomology.
A treatment approach must be carefully selected, knowing that these errors and behaviors
are likely a product of this core deficit of ASD. It is challenging to draw direct
conclusions from present work, however, because the majority of studies examining
executive functions and language include children with high-functioning ASD.
In addition to considering ASD severity and symptomology, it is still important to
comprehensively account for an individual's profile, including strengths, weaknesses,
and motivators, before administering treatment. Consideration of the individual's
profile, use of clinical expertise and judgment, and implementation of research-based
treatment need to be integrated to perform evidence-based practice.
Conclusions
There are large variations in findings and contradictory reports of executive function
performance and its relation to language ability in children and adolescents with
ASD, and future work should aim to carefully characterize the participant samples
to aid in presenting clear conclusions within a heterogeneous disorder. It remains
evident that language and executive function are important for several aspects of
daily life, impacting outcomes such as scholastic achievement, peer victimization,
and reciprocal relationships.[14]
[15]
[16] Given that both executive functions and language ultimately have profound effects
on daily life, and that one can be influenced by targeting the other in intervention,
it is important that future research continues to examine the interplay between these
two domains.
