Selective Attentional Networks Deficits in Parents of Children With Autism Spectrum Disorder: Evidence From Attention Network Test
Article information
Abstract
Objective
We primarily aimed to investigate the attention network function among parents of children with autism spectrum disorder (ASD) using the Attention Network Test (ANT). The secondary objective was to observe whether the three attention networks of all participants were related to each other.
Methods
We included 28 parents of children with ASD and 28 well-matched parents of typically developing children. All participants underwent the neuropsychological assessment and ANT test. The three distinct attention networks, including alerting, orienting, and executive control, were also measured.
Results
Compared with controls, parents of children with ASD showed less-efficient alerting and executive control network (all p<0.05), but not orienting network (p=0.74). No significant correlation was found between the alerting, orienting, and executive control network for either group.
Conclusion
Our findings showed that parents of children with ASD had deficits in alerting and executive control attention functions. The deficits are indications of a broad autism phenotype.
INTRODUCTION
Autism spectrum disorder (ASD) is a lifelong neurodevelopmental condition that affects social, communicative, behavioral, and cognitive functioning [1,2]. The prevalence rates of autism have increased dramatically over the past decades, and the new estimate is a global prevalence of 1% [3].
Individuals with ASD typically exhibit a common characteristic, that is, some form of attention deficit, although they differ in age and range of abilities [4]. Studies have documented that individuals with ASD, including children with ASD [5], show deficits when performing tasks involving attention [6-8]. Attentional deficits in ASD are commonly observed in social contexts, however, it occasionally occurs in elementary forms of attentional processing such as simple target detection tasks as well as tests of vigilance or sustained attention [7]. Neuroimaging research has revealed different patterns of brain activity in children with ASD, even when ASD children and neurotypical children show the same performance on attention tasks [9]. Additionally, a meta-analysis by van der Meer et al. [10] showed that children with ASD had deficits in both visual and auditory attention, indicating that the deficiency was not modality specific. Specifically, individuals with ASD show increased activation in the frontal lobe, which is thought to reflect the recruitment of alternative brain regions as a means of compensating for achieving the same level of attentional performance [9]. Although individuals with ASD often exhibit attentional difficulties, the exact nature of their attentional deficits remains unclear.
As a vital cognitive function, human attention work is a complex physiological process. It integrates external sensory inputs with internal information and decision-making and requires the interaction of many subcomponents to complete even the simplest attention task [11,12]. Now that attentional dysfunction have been demonstrated to be a consistent observation in ASD. It would be worthwhile to isolate the subcomponents of attention to determine whether these impairments represent a global deficits or localized dysfunction [13]. Understanding the underlying locus of the attentional deficits as exhibited by individuals with ASD would contribute to managing and treating their attentional problems.
The attention network theory proposed by Posner and Petersen [11] provides a useful theoretical framework for identifying the subcomponents of attention. This theory suggests that the attention system can be divided into three distinct subcomponents, that is, alerting, orienting, and executive control. These three neural networks have been shown to be functionally and structurally independent [14] and are related to specific neural substrates and specific functions [11,12,15,16]. Specifically, alerting specializes in the achievement and maintenance of appropriate sensitivity to incoming stimuli and is associated with activation in the locus coeruleus, right frontal and parietal brain areas. Orienting is responsible for selecting information from numerous sensory inputs, and is associated with the temporoparietal junction (TPJ), superior parietal lobe, and frontal eye fields (FEFs). Executive control involves the resolution of response conflicts among competing sources of input and is related to the anterior cingulate cortex and prefrontal cortex. But they may interact to influence performance [17,18].
Based on attention network theory [11], Fan et al. [14] developed the Attention Network Test (ANT) to simultaneously evaluate the three aforementioned attention components within one experiment. Participants were instructed to determine the direction of the “target” arrow by pressing a key as quickly as possible [14]. The efficiency value of each network is determined by calculating the average response time under several different prompts and flanker conditions [19]. To date, the ANT has been widely utilized to assess the attention function in various psychiatric disorders, including schizophrenia,20 depression [21], attention deficit hyperactivity disorder [22], as well as ASD [8].
Previous studies used the ANT task to investigate the attention in ASD, showing different results with deficits being observed in three attention components. A study done by Fan et al. [23] found weaker alerting attention and executive attention in 20 adults with ASD compared to healthy controls, while no differences were found for orienting attention. Another two study in children with ASD showed different result [8,24]. One was found that, compared with the control group, orienting and executive control attention was weaker, but the other study found that the attention deficit in children with ASD was only existed in the orienting attention.
Genetic factors have been considered to play a significant role in the pathophysiological mechanisms of ASD. Twin studies and epidemiological studies indicated that the heritability of ASD is estimated to be approximately 80%–90% [25,26]. This means that ASD is a highly genetically neuropsychiatric disorder [27]. A considerable number of family members of children with ASD, especially in first-degree relatives (including parents, siblings, and offspring), were found to exhibit a set of characteristics that are resemble some of the core symptoms of ASD [28,29]. These characteristics are inadequate to meet the diagnostic criteria for ASD, and they are termed as a broad autism phenotype (BAP) [30,31]. However, thus far, few studies have focused on the attentional capacities of first-degree relatives of individuals with ASD. It remains unclear as to whether first-degree relatives with ASD share similar attentional dysfunction as those observed in ASD.
This study aimed to investigate and compare the differences in attention function between parents of children with ASD and parents of typical developmental children (TDC). We specifically designed to employ the ANT task to simultaneously investigate the three subcomponents of attention system in each group. According to the literature, we predicted that parents of children with ASD would exhibit less efficient alerting, orienting, and executive attention, compared to controls, or show selective impairment of attention subcomponents.
METHODS
We recruited 28 parents of children with ASD (9 men, 19 women) from autism rehabilitation centers. According to the diagnostic criteria for ASD, the Diagnostic and Statistical Manual of Mental Disorders, 5th edition [32], a structured clinical interview was used to confirm that these recruited parents were not diagnosed with ASD by experienced clinical psychiatrists, and excluded with other neurodevelopmental disorders and other psychiatric disorders. Furthermore, their children were diagnosed with ASD by more than two child psychiatrists. Twenty-eight healthy controls (8 men, 20 women), enrolled from a primary school, were matched with parents of children with ASD for age, gender, and educational level. Two experienced psychiatrists assessed the healthy controls and their children, and confirmed that they all did not meet the diagnostic criteria for ASD.
None of the participants had a history of neuropsychiatric disorders, head injury, or substance abuse. And they were all right-handed. All participants gave their written informed consent prior to the experiment according to the Declaration of Helsinki. The protocol of this study was reviewed and approved by the Institutional Review Board (IRB) of Bengbu Medical College (IRB No. 2019-061).
Measures
Neuropsychological evaluation
All participants underwent neuropsychological tests to evaluate their basic cognitive conditions. The Montreal Cognitive Assessment (MoCA) test was used to measure general cognitive function [33]. The Digit Span (DS) (forward, backward) of the Wechsler Adult Intelligence Scale was used to assess attention [34]. Trail Making Test (TMT) (A, B) was used to evaluate common execution functions [35]. We used the Chinese version Autism Spectrum Quotient (AQ) to evaluate the autistic characteristics and broader autism phenotypes of all participants [36]. On average, it took 25–30 minutes for participants to complete the assessment. Skilled psychologists and psychiatrists administered all assessments.
ANT
The ANT task required the participants to watch the stimuli presented on the computer screen, and we collected the reaction time (RT) by two corresponding keys. The stimuli consisted of five horizontal arrows that pointed to the left or right on a gray background. The target item was presented in the center of the five arrows. As an arrow faced left or right, the arrows on both sides of the target created three possible conditions: neutral condition, congruent condition, and incongruent condition. When appeared neutral condition, the arrows on both sides do not appear. In the congruent condition, the arrows on both sides pointed in the same direction as the target item; when appeared incongruent condition, the directions of the arrows on both sides were inconsistent with the target item. Throughout the task, participants were required to focus on the centrally located fixation cross, and they needed to press the keyboard direction keys as fast and accurately as possible by pressing the corresponding button with their finger of their left or right hand. Each stimulus of the test was divided into four steps, and then repeated cycles until the end of the task. In the first step, a fixation point “+” appeared in the screen randomly, and this point would last for 400–1,600 ms. In the second step, the warning cue would or would not appear for a fixed time of 100 ms. In the third step, a fixation point with a duration of 400 ms appeared on the screen. In the last step, the target stimulus appeared until the participants responded by pressing the button. In this step, the stimuli could last up to 2,700 ms. If the participants completed the response shorter than 2,700 ms, the target stimulus would disappear. The computer would adjust the duration based on the duration and RT of the first step after the participants pressed the button of the target stimulus to ensure a fixed experimental interval.
There were four different cue conditions: no cue, central cue, double cue, and spatial cue. In the no cue condition, the fixation cross appeared directly in the center of the screen for only 100 ms. In the central cue condition, the cue was located at the center of the fixation point. In the double cue condition, cue position appeared above and below the fixation point, this kind of cue does not provide any spatial information and is just a hint of the stimulus to come for participants. In the spatial cue condition, the cue position appeared directly above or below the fixation point, this kind of cue reminded the participants of the target location.
Each ANT task consisted of four blocks. The first block consisted of 24 full feedback exercises. The subsequent three blocks consisted of 96 trials which included 48 different conditions: 4 types of cues×3 flanker conditions×2 target locations×2 target stimuli directions, with two repetitions, and the order of the trails presented to the participants was random. The experiment was carried out in a quiet environment. Figure 1 illustrates the ANT task.
Calculation of the three subtypes of attention networks
We calculated the efficiencies of the alerting, orienting, and executive control networks using the RTs under different test conditions, including 12 conditions as described above. Firstly, the mean RT of the no cue condition minus the mean RT of the double cue condition was used to represent the efficiency of the alerting network. The larger the score of the alerting network efficiency, the higher the alerting attention function of the participants. Secondly, the mean RT in the central cue condition minus the mean RT in the spatial cue condition is used to represent the efficiency of the orienting network. The larger the value of orienting network efficiency, the higher the orienting attention function of the participants. Lastly, we calculated the efficiency of the executive control network by using the mean RT of the congruent target conditions minus the RT of the incongruent target conditions [14]. The greater the value of the executive control attention network efficiency, the lower the executive control function of the participants.
Statistical analyses
We used SPSS, version 22.0 (IBM Corp., Armonk, NY, USA) for Windows to examine all statistical analyses. According to this study, a p<0.05 (two-tail) was considered to be statistically significant. We used independent sample t-test to analyze the neuropsychological data (including age, education, DS, TMT, MoCA, and AQ), background information data, and three efficiency values of attention networks of the two groups. And we used the chi-square test to compare the gender differences between the two groups. Apart from these, we applied Pearson’s correlation test to assess the relationships between three networks’ efficiency and the relationships between the AQ and the three networks’ efficiencies.
RESULTS
Neuropsychological and background tests
Table 1 shows the detailed demographic statistics and neuropsychological data of the participants. Independent sample t-tests, chi-square test showed no difference between parents of children with ASD and parents of TDC in age, education, gender, DS, TMT, MoCA, and AQ (p>0.05).
Efficiencies of the three subtypes of attention networks
The results of participants’ ANT are shown in Table 2. The data shows that the mean RT of alerting network of parents of children with ASD was significantly shorter than that of the parents of TDC (p=0.007). While the mean RT of the executive control network of parents of children with ASD was significantly longer than that of the parents of TDC (p=0.011). But there were no significant differences between the two groups in the orienting network (p=0.740). In addition, there were no significant statistical differences between the two groups in the overall mean RT and global accuracy.
Figure 2 shows the scores of three networks ratios between parents of children with ASD and parents of TDC. There were significant differences in the alerting and executive control networks (p=0.009, p=0.016, respectively), but there was no significant difference in the orienting network (p=0.77). It indicated that there were selective differences in the alerting and executive control networks, but the orienting network was not affected.
Correlations between attention networks and AQ
In order to examine the relative independence of three subtypes of attention networks in the literature, we used Pearson correlation analysis to analyze the data of 56 participants. The results shows that there was no significant correlation between alerting, orienting, and executive in any combination of the two groups. Table 3 illustrates the results of the calculations.
As for the relationship between the three subtypes of attention networks and various dimensions of AQ. The results shows that there was no significant correlation between the scores of the three attention networks and the various dimensions of AQ, whether it is the parents of children with ASD or the parents of children with TDC. Table 4 illustrates the results of the calculations.
DISCUSSION
This study aimed to use ANT, a nonverbal, nonsocial task, to measure whether there are differences in attention function between parents of children with ASD and parents of TDC. The results showed that there were significant differences in alerting and executive control networks between parents of children with ASD and parents of TDC, which showed that parents of children with ASD had decreased alerting and executive control functions. Moreover, we found no significant functional correlation between the alerting, orienting, and executive control functions of the attention network, which showed that these three attention subtypes were independent of each other. As far as we know, this study was one of the few studies that used the ANT task to analyze the attention function of parents of children with ASD. Our findings suggested that selective impairment of attentional function in parents of children with ASD was an indication of a BAP. Our study showed that, with comparable correctness rates, the alerting network efficiency of the parents of children with ASD was significantly lower than that of TDC parents. This finding indicated that parents of children with ASD had a weaker alerting function than parents of TDC, and was unable to react faster to what happened in their environment. According to the alerting network system, the alerting network system is a state in which the body is able to react quickly to its surroundings in order to receive incoming information more quickly, and it is mainly responsible for maintaining the sensitive state of receiving incoming information. Evidence from previous studies showed that children with autism were less sensitive to new things [37,38]. And it has been reported in the literature that autistic traits negatively regulate normal people’s perception of social stimulation, such as emotional faces [39]. These results indicated that autism or autistic traits would affect the alerting function of normal individuals to some extent. Therefore, in order to figure out the mechanism of impaired alerting network function and genetic characteristics of parents of children with ASD, the specific neural mechanism of parents of children with ASD needs to be further researched.
Our study results demonstrated that the efficiency of the executive control network of the parents of children with ASD was significantly longer than those of the parents of TDC. This indicated that when dealing with incongruent flanker information, it was difficult for the parents of children with ASD to ignore the conflicting directional information represented by the incongruent flanker information. In order to obtain higher accuracy, it took more time to deal with conflicting directional information. According to the result of McDowell et al. [40] and Petersen and Posner [41] on the neural aspects of the executive control network, the executive control network was regulated by two different neural circuitries: 1) Cinguo-opericular control system, which performed the functions of attention behavior monitoring and task-set maintenance, and the function of stable maintenance throughout the whole control process. 2) The frontoparietal control system, which performed the functions of task-set initiation and real-time adjustment. Previous studies have shown that individuals with ASD have diminished activity levels of anterior cingulate compared with normal adults [23]. With regard to the mechanism of this defect, we hypothesized that this may account for the poorer executive attentional function of the parents of children with ASD compared with the parents of TDC in the present experiment.
Notably, we found that there is no significant difference in orienting attention efficiency between parents of children with ASD and parents of TDC. However, there are many factors that affect the orienting efficiency. We still could not tell whether the parents of children with ASD had abnormalities in the orienting attention network. First of all, similar to executive control attention, Petersen and Posner [41] proposed the existence of two types of neural circuits governing the orienting attention: 1) Dorsal network system: this system works primarily on fast strategic control of attention, and it is composed of FEFs and the intraparietal sulcus. 2) Ventral network system: this system is more active in tracking targets, and it is considered to be a part of the response network for sensing events. It consists of the TPJ and the ventral frontal cortex [41]. These two attention systems jointly determine the orienting attention network. Based on the existing research results on ASD, there are inconsistent researches between two systems of orienting attention network [42]. Renner et al. [43] reported that the deficits in the orienting attention network of ASD only existed in the ventral system, while an intact dorsal system is preserved [43]. And a subset of studies found that the dorsal system in ASD was enhanced [44]. In addition, to some extent, the age of the participants also affects the orienting network efficiency values. Based on the results of the existing literature, we found that the dorsal system visuospatial function is lower and plastic in children compared to adults. But ANT has shortcomings: it cannot distinguish whether the abnormally orienting attention network present in the participants was a dorsal system or ventral system. Therefore, we have the following three speculations: 1) Parents of children with ASD retain intact orienting networks. 2) Dorsal system and ventral system in parents of children with ASD are not synchronized in efficiency, and there may be an enhancement of the dorsal system [44] along with an impairment of the ventral system brain areas [43], but there is some extent of neutralization in terms of overall effect. And ANT can only make judgments in general and cannot be specific to the two attention systems of orienting [3]. 3) The dorsal system has a certain plasticity. The participants in this experiment were all parents of children with ASD, with an average age of 34 years. The dorsal system of parents of children with ASD was enhanced with age, and their original differences in the orienting network gradually disappeared.
In addition, our results showed that parents of children with ASD had selective impairment of attention function, similar to the results of Hughes et al.’s [45] research on the executive control network of parents of children with ASD. Hughes et al.’s [45] findings showed that parents of autistic children had impairment of executive control function. This impairment was only observed in a quarter of parents of children with ASD, but its incidence was 5–10 times higher than that of learning disability and control group. This indicated that poor attentional function was a characteristic of parents of children with ASD, rather than a universal phenomenon. In our study, it is worth noting that the impairment not only lies in the executive control function, but also in the alerting function. We suggest that it was due to a broader autistic phenotype, or BAP, resulting from a higher-than-normal proportion of autism-related impairments in relatives of autistic individuals due to genetic factors. But it is a challenging task to define the boundaries of BAP. In previous work, family history interviews have been used to determine that the proband of autism has social and communication impairments, but there is no effective method to specifically diagnose BAP. Some people believed that executive function deficiency and other cognitive defects would jointly produce corresponding diseases, and the generation of ASD is related to the interaction of executive function defects and facial processing defects. Hill [46] found that the impairment of executive function might be one of the promising candidate factors. Hughes et al.[47] believed that to determine the genetic tendency of ASD, it is necessary to investigate which component of executive function is the dominant factor. In addition, a study by Wong et al. [48] showed that parents of children with ASD were worse than normal people in the two executive functions of ideational fluency and generative ability, and fathers of children with ASD also had defects in setting shifting to a previously unrelated dimension. Our experimental results are consistent with those of Wong et al. [48]. The genetic tendency of ASD is more complex, and the study of its inner phenotype is more helpful for the specific diagnosis of BAP. The selective impairment of executive function (such as conceptual fluency and generative ability) can be used as an auxiliary diagnostic criterion in the later confirmation of the BAP, and the core defects of executive function in ASD can be further comprehensively understood in future research. This will be more conducive to the diagnosis and treatment of ASD.
In addition, we noticed that no significant relationship was discovered between the three attention networks in the two groups. This finding implies that three subtypes of networks of participants were functionally independent and did not interfere with each other among ASD parents and TDC parents. Similarly, a study conducted by Fan et al. [14] reported that the three subtypes of attention networks were independent of each other in ANT tests of normal adults. However, Mutreja et al. [8] found that there was a significant relationship between the alerting and orienting network in both children with ASD and neurotypical children, but no significant correlation was found between orienting and alerting with executive control network in either group. In the future, maybe we should add some larger-scale in-depth experiments and magnetic resonance imaging (MRI) images to verify the correlations between the three subtypes of attention networks in the first-degree relatives of ASD.
Because of practical constraints, four limitations of our study should be presented. First, the sample size included in our experiments was small and participants were selected mainly included mothers, not both parents. ASD shows different psychological and phenotypic characteristics in girls, and the genetic transmission path may be different from boys. Second, although ANT can be well observed by dividing the human attentional function into three parts, but it can be further subdivided according to the neural circuitry that regulates it. ANT cannot make further observations. Thirdly, we only divided the participants into the ASD group and the control group, but did not divide the ASD group into two or more groups according to the symptom severity of ASD and analysis of variance between groups. Finally, while the ANT task can give intuitive values of the efficiency of the relevant attention system, however, attentional function involves the activation or inhibition of relevant brain regions, which is a complex and unpredictable process. No imageological examinations were performed in this study. In future research works, larger samples may be used to verify our findings. If conditions permit, participants should have a functional imageological examination (e.g., functional MRI). This can help us to study the formation mechanism of attention deficit of parents of children with ASD more accurately.
In conclusion, the results of our study showed that the attentional function of parents of children with ASD was selectively impaired in the alerting and executive control network. The efficiency of the executive control network was higher than that of parents of TDC, indicated that the executive control function was weakened. In the meanwhile, the efficiency of the alerting network was obviously lower, which indicated that the alerting function of parents of children with ASD was decreased too. There was no significant correlation between the three networks of parents of children with ASD and parents of TDC, and they were three functionally independent attention systems. Moreover, the correlation analysis between AQ and the three attention networks of the two groups showed that there was no significant relationship, suggesting that we cannot predict the efficiency of attention networks through AQ scores. Our results suggested that autism, a genetic factor disorder of neurocognitive dysplasia, and abnormalities in the alerting and executive control network of parents of children with ASD are part of BAP. Our results represented a further increase in our understanding of attention deficits in parents of children with ASD and expand the understanding of selected attentional deficits in parents of children with ASD.
Notes
Availability of Data and Material
The datasets generated or analyzed during the study are available from the corresponding author on reasonable request.
Conflicts of Interest
The authors have no potential conflicts of interest to disclose.
Author Contributions
Conceptualization: Shunhua Zhang. Data curation: Tian Qin, Shunhua Zhang. Formal analysis: all authors. Funding acquisition: Shunhua Zhang. Investigation: all authors. Methodology: Tian Qin, Shunhua Zhang. Project administration: Shunhua Zhang. Resources: Shunhua Zhang. Software: all authors. Supervision: Shunhua Zhang. Validation: Tian Qin, Shunhua Zhang. Visualization: Tian Qin, Shunhua Zhang. Writing—original draff: Tian Qin. Writing—review & editing: Shunhua Zhang.
Funding Statement
This work was funded by the National Innovation and Entrepreneurship Training Program for college students of China under Grant 201910367009, 202010367036, 202110367049, and 202210367022, as well as Anhui Provincial Colleges and Universities Ideological and Political Work Ability Improvement Project (No. sztsjh2019-7-14), and the Key Projects of Teaching Research for Colleges and Universities in Anhui Province, China (No. 2020jyxm1174).
Acknowledgements
We would like to appreciate all the participants for their unique contribution. We are also grateful for the valuable support from Anhui Province Key Laboratory of Cognition and Neuropsychiatric Disorders, and Gongjun Ji, Dandan Li, and Tingting Yang.