Autism and the Complex Visual System
Abstract
Researchers examined if visual processing in children with autism spectrum disorder differs from visual processing in their typically developing counterparts. Participants viewed a series of feature search tasks that were either single feature or conjunction feature tasks. Statistical analysis revealed that children with ASD have enhanced search accuracy and when compared to their typically developing counterparts. Statistical analysis also revealed that children with ASD have decreased amplitude and increased latency in event-related potentials as well as visual related potentials. These results suggest that children with ASD are more accurate in search tasks than typical children while simultaneously exhibiting visual processing deficits.
————————————-
Autism Spectrum Disorder (ASD)
Autism spectrum disorder (ASD) is a term that describes a group of pervasive developmental disorders characterized by qualitative abnormalities in reciprocal social interactions and patterns of communication, and by a restricted, stereotyped, repetitive repertoire of interests and activities, generally appearing in the first two years of life (Dover & Le Couteur, 2007). Early signs and symptoms may include atypical attention, poor eye contact, failure to follow gaze, limited imitation of others, or sensory hypo/hypersensitivity. This study focuses on the early sign of atypical attention and investigates the neural mechanisms involved with attention in children with ASD.
ASD and Attention
Two well-studied types of attention are spatial and feature-based attention. Spatial attention refers to attention focused on certain positions in space while feature-based attention refers to the attention focused on specific features, such as color (Galashan & Siemann, 2017). Previous research has found that children ages 7-18 with ASD have difficulty in overriding incorrect attentional cues, particularly with non-social cues, such as a vehicle or a chair. The children in this study were successful in processing predictive value of directional cues but were (1) less engaged by visual information than their typically developing peers and (2) less able to reallocate attention after it was directed (Jaworski & Eigsti, 2017),
A study done on toddlers with ASD utilized a visual search task to explore looking behavior and visual search skills in 17 children around the age of 29 months. The researchers developed a classic visual search paradigm that contrasts feature search and conjunction search without the use of verbal instructions. Their results showed that toddlers with ASD were more successful at finding a target among distractors (feature conjunction search) when compared to their typically developing peers. The reason for these results is still debated among researchers but it is likely that enhanced discrimination, an increased ability to discriminate targets from distractors, would allow toddlers with ASD to search for the target more efficiently. This enhanced search is believed to be due to increased salience for the target, which improves their search performance (Kaldy et al., 2011; O’Riordan & Plaisted, 2001). Currently, there is no research done on the mechanisms to explain the process of enhanced discrimination in children with ASD.
ASD and Visual Processing
Individuals with autism have atypical cognitive profiles, such as impaired social cognition and social perception, executive dysfunction, and atypical perceptual and information processing. These profiles are underpinned by atypical neural development at the systems level (Lai et al., 2014). Yamasaki et al. found: (1) enhanced and impaired processing co-exists within the lower visual area (V1), (2) local information integration from area V1 is impaired in higher-level visual areas after V4 and V5/MT and (3) the dorsal attention network is impaired while the ventral attention network is intact in ASD. These researchers believe that the findings indicate complex functional alterations in visual and attention networks that support the possibility of altered connectivity within and between distributed brain regions in ASD (Yamasaki et al., 2017).
The attention network involves various regions of the brain, including the superior temporal cortices, the superior frontal lobe, and the inferior parietal lobe (Hopfinger, 2000). Studies have found that the early stages of attentional modulation take place in the superior temporal gyrus (Hill & Miller, 2010). Further, research has shown increased activity in the superior frontal cortex during attentional shifts to peripheral locations when compared to maintenance of attention at fixation (Corbetta et al., 1990). Large regions within the frontoparietal network, including the superior frontal lobe, have been found to play a role in the top-down bias signals sent to various regions of the visual system (Kanwisher & Wojciulik, 2000). The inferior parietal lobe has been shown to aid in top-down attentional orienting (Shomstein, 2012). Although there is much research investigating the role of brain activity in attention tasks as well as research investigating atypical neural development in autism, there is limited research that combines these two aspects to better understand brain processing in children with autism.
Recent research has begun to explore this topic through the use of visual evoked potentials (VEPs) and event-related potentials (ERPs). These are objective and non-invasive methods to delineate subtle functional abnormalities in the human visual system which directly measure neuronal activities with high temporal resolution. The visual evoked potential (VEP) is measured by the electrodes overlying the scalp region and breaches anywhere in the visual system can generate abnormal VEPs. VEP averages occipital lobe activity evoked from contrast stimulation of the visual system and can identify visual malfunction in patients lacking visual symptoms (Kothari et al., 2016). A study done on pre-school-aged children with ASD found a correlation between longer VEP latency and abnormal behaviors, which suggests a delayed neural communication within other neural circuits, apart from the visual pathway (Sayorwan et al, 2018).
Event-related potentials (ERPs) are changes in an EEG in response to specific events or stimuli that are time-locked to sensory, motor or cognitive events and are thought to reflect the summed activity of postsynaptic potentials produced when a large number of similarly oriented neurons synchronously fire during information processing (Sur & Sintha, 2009). Previous research done on adults with Asperger’s syndrome found reduced amplitude of ERP wavelengths, suggesting an alteration of visual processing (Kornmeier, 2014).
Overview of the Current Study
Current studies are often limited to investigating singular aspects of the visual system in children with autism, such as enhanced visual search speed or visual processing dysfunction. Previous research studies have found that children with ASD are more successful at finding target stimuli among distractors while others have found that children with ASD have impaired visual processing systems. This study aims to integrate both of these topics of interest into a novel design that examines both target search capabilities and visual processing in children with ASD. In order to accomplish this aim, the study utilizes feature search trials (used to investigate search accuracy) as well as conjunction search trials (used to investigate visual processing). Researchers utilized a visual task search in combination with eye-tracking, visual evoked potentials, and event-related potentials (ERPs) to investigate whether children with ASD have a unique visual system when compared to their typically developing peers.
First, we hypothesized that children with ASD will be more successful at finding target stimuli among distractors in feature search trials when compared to the control group due to enhanced discrimination (O’Riordan & Plaisted, 2001). Further, we hypothesized that children with ASD will exhibit visual processing dysfunction in conjunction with search trials due to delayed neural communication and altered connectivity. We expect that children with ASD will have longer VEP latency as well as reduced ERP amplitude, which would suggest delayed neural communication as well as altered connectivity within and/or between distributed brain regions.
Method
Participants
Ten children with ASD under the age of five (6 male, 4 female) and ten children without ASD under the age of five (5 male, 5 female) completed the study. All participants with ASD were evaluated using the Autism Diagnostic Observation Schedule and results were confirmed by a clinical psychologist. The clinical severity of autism was evaluated using the Autism Treatment Evaluation Checklist (ATEC). For detailed participant characteristics, please see Table 1. None of the participants had first-degree relatives with colorblindness.
Apparatus
A screen-based Tobii Pro Spectrum eye-tracker was used to measure eye movement patterns. This eye-tracker was chosen due to its screen attribute which allows for more accurate data recording by eliminating participant discomfort that may occur with other devices such as glasses. An electroencephalogram (EEG) was used to obtain VEP and ERP recordings. The EEG procedure was replicated from previously successful studies with children on the autism spectrum. The participants head was measured and marked with a washable wax pencil to ensure accurate placement of the net, which was then placed over the scalp. Prior to fitting the sensor net over the scalp, the sponges were soaked in an electrolyte solution. For most participants, at least two minutes of baseline activity were recorded but the recording may be limited to no more than two minutes depending on the child’s tolerance (Bosl et al., 2018).
Procedure
Participants sat on their caregivers’ lap, approximately 60 centimeters away from the monitor. Caregivers were given an eye mask to wear over their eyes in order to ensure that they were not able to view the screen. Following a 30-second calibration phase, participants saw two blocks of trials, consisting of four familiarization trials and eight test trials. Each trial lasted for approximately 30 minutes. In familiarization trials, the three items that were used in the search displays (the red apple target, green apple color distractor, and a red, elongated popsicle shape distractor) appeared on the screen for three seconds. In each trial, the three items were presented in a different spatial configuration. The test trials consisted of four single feature trials (set size 3 or 5) and four feature conjunction trials (set size 4, 8, or 10). Single feature trials can be defined as feature searches in which participants look for one aspect of stimuli, such as color. Conjunction search trials can be defined as feature search in which participants look for more than one aspect of stimuli, such as color and shape.
Figure 1 provides an example of a trial with a set size of 10. Test trials were presented in random order, with the exception that the first three test trials were single feature trials to emphasize the special status of the target through pop-out. Before each trial began the target (a red apple) swirled in from the upper portion of the screen, stopped at the center of the screen for approximately two seconds, and then disappeared. Participants were then presented with a search display, presented in a randomized order for each participant, for six seconds. After the end of the trial, the target item spun clockwise 180 degrees, then counterclockwise 180 degrees (for a total of two seconds).
Results
The age distribution in both groups was not statistically different by means of the chi-squared test. The clinical characteristics of the participants including age, gender, and ATEC screening scores are displayed in Table 1.
Search Performance
A trial was considered successful if the participant fixated on the target at least once for at least three seconds. Figure 2 shows the success rates as a function of set size for each group. In the ASD group, the mean success rate was 88.20 for single feature trials (SD = 4.31) and 70.80 for conjunction feature trials (SD = 5.82). In the control group, the mean success rate was 82.20 for single feature trials (SD = 4.88) and 52.80 for conjunction feature trials (SD = 6.17). A repeated-measures ANOVA was run to determine whether there is an effect of ASD diagnosis on success rate. The between-subjects factor was a group (ASD or control) and the within-subjects factor was a trial type (single or conjunction). There was a significant main effect of trial type, F (1, 9) = 23.72, p < .05. There was also a significant main effect of group F (1, 9) = 21.87, p < .05. The interaction between trial type and group was also highly significant F (1, 9) = 22.47, p < .01. Both the control group and the ASD group were more successful in the single feature trials and this effect was stronger in children with ASD when compared to the typically developing control group.
VEP Analysis
For the VEP analysis, the amplitude and the latency of the VEP waveforms were used as parameters for comparing the outcomes between the groups. Amplitude, the difference between the height of the current peak and the highest point of the previous peak, was measured in microvolts (µv). Latency, the duration from the visual stimulus to the peak of each VEP waveform, was measured in milliseconds (ms). The amplitude and latency of the VEP waveforms were used as a basis for comparison between groups. Wavelength amplitude for single feature trials was found to be 4.45 µV ± 3.10 for the ASD group and 6.82 µV ± 4.60 for TD children. Wavelength amplitude for conjunction feature trials was found to be 5.27 µV ± 3.10 for the ASD group and 7.24 µV ± 4.60 for TD children (see Figure 3). Wavelength latency for single feature trials was found to be 82.56 ms ± 5.20 for the ASD group and 81.24 ms ± 3.33 for TD children. Wavelength latency for conjunction feature trials was found to be 127.44 ms ± 8.64 for the ASD group and 112.75 ms ± 4.58 for TD children (see Figure 4). These findings suggest that there is altered connectivity within and between distributed brain regions in children with ASD.
ERP Analysis
A mixed-model ANOVA was run to determine whether there is an effect of ASD diagnosis on altered connectivity using ERP amplitude. The difference traces (dERPs) between single feature trials and conjunction feature trials show a sizable effect of set size at occipital electrodes and thus in visual areas (see Figure 5). The analysis indicates a strong effect of the factor group, F(1,9) = 19.99, p < .05, as well as a strong effect of the factor trial type (F(1, 9) = 25.17, p < .05, reflecting larger amplitudes for the conjunction feature trials compared to single feature trials. Further, the analysis indicates an interaction between the group factor and trial type factor (F(1,9) = 1.95, p < .01, reflecting a larger difference of the ERP trial-type effect between observer groups at central electrodes (Kornmeier et al., 2014).
Discussion
The current study conducted a novel design that investigated both target search capabilities as well as visual processing mechanisms in children on the autism spectrum. This design has never been utilized before and presents key findings for the field of autism research. In support of our hypothesis that children with ASD will be more successful at finding target stimuli among distractors due to enhanced discrimination, we found higher mean success rates for the ASD group in both single feature trials and conjunction feature trials. Our results replicate and extend the existing literature on the visual system in individuals with an autism spectrum disorder. In support of previous studies, we found that children with ASD are better at finding a target stimuli among distractors when compared to their typically developing peers (Kaldy et al., 2011). Further analysis found a significant main effect of trial type, group, and an interaction between the two. This statistical evidence suggests that the mean success rates were influenced by both ASD diagnosis as well as the type of feature search. Children with ASD outperformed the TD group in both trial types. Further, both groups were more successful with the single feature search and this finding was strongest in the ASD group.
Although our study replicated findings that support the use of enhanced discrimination in children with ASD, a limitation of the current study is that researchers did not investigate the mechanisms to explain this process. Future studies should make this a priority by further analyzing EEG recordings in combination with eye-tracking methods. Understanding the mechanism of enhanced discrimination has the potential to aid typically developing children who have trouble with focusing on tasks, such as children diagnosed with ADHD.
Additionally, we found support for our hypothesis that children with ASD will exhibit visual processing dysfunction in conjunction search trials due to delayed neural communication and altered connectivity. We found a correlation between longer VEP latency and “abnormal behaviors” which can be understood as autism for the purpose of this research (Ossenkopp, 2017). The ASD group displayed smaller VEP amplitude and longer VEP latency for both trial types. Our study presents a more detailed analysis of VEP wavelengths in relation to autism by examining both the latency as well as the amplitude for two types of feature search tasks. Further, the ASD group displayed smaller ERP amplitude and the statistical analysis suggests that amplitude is affected by ASD diagnosis, trial type, and set size. The existing literature on ERP wavelengths is limited to amplitude recordings for individuals diagnosed with Asperger’s syndrome. (Kornmeier, 2014). Our study broadens this research by including all individuals on the autism spectrum as well as targeting a younger age demographic, which had not been investigated prior to our publication. Both the VEP data and the ERP data indicate that the autism group does have delayed neural communication. Our analysis presents novel findings that the delay in neural communication is dependent upon task complexity as well as task size.
This study is limited in that we only utilized common shapes (apples and popsicles) in the feature search tasks. Although this was an important first step in breaking ground for the complexity of creating a research design that utilizes three different analyses, it does not answer many questions that remain about visual processing in autism. Future studies should extend our research by adding a layer of depth to the visual search tasks through the use of more complex images such as faces or animals. This would allow for better understanding as to why individuals with autism process visual stimuli but often fail to respond. Additionally, this may provide insight into how therapy can utilize various techniques to teach social cue responses based on what children with autism are innately inclined to pay attention to.
The current study conducted a novel design that investigated both target search capabilities as well as visual processing mechanisms in children on the autism spectrum. Our study has important implications for further understanding the visual process and dysfunctions in children with ASD. Our findings suggest that children with ASD are better at search accuracy when compared to their typically developing peers but they do not utilize the same neural network or processes when examining a visual field.
References
Bisley, J. W. (2011). The neural basis of visual attention. The Journal of Physiology, 589(1), 49–57.
Bosl, W.J., Tager-Flusberg, H. & Nelson, C.A. EEG Analytics for Early Detection of Autism Spectrum Disorder: A data-driven approach. Sci Rep 8, 6828 (2018).
Corbetta, M., Miezin, F. M., Dobmeyer, S., Shulman, G. L., & Petersen, S. E. (1990). Attentional modulation of neural processing of shape, color, and velocity in humans. Science, 248(4962), 1556–1559.
Davis, E. T., & Palmer, J. (2004). Visual search and attention: An overview. Spatial Vision, 17(4– 5), 249–255.
Dover, C. J., & Le Couteur, A. (2007). How to diagnose autism. Archives of Disease in Childhood, 92(6), 540–545.
Galashan, D., & Siemann, J. (2017). Differences and Similarities for Spatial and Feature-Based Selective Attentional Orienting. Frontiers in Neuroscience, 11.
Hopfinger, J. B., Buonocore, M. H., & Mangun, G. R. (2000). The neural mechanisms of top- down attentional control. Nature Neuroscience, 3(3), 284–291.
Jaworski, J. L. B., & Eigsti, I.-M. (2017). Low-level visual attention and its relation to joint attention in autism spectrum disorder. Child Neuropsychology : A Journal on Normal and Abnormal Development in Childhood and Adolescence.
Kaldy, Z., Kraper, C., Carter, A. S., & Blaser, E. (2011). Toddlers with Autism Spectrum Disorder are more successful at visual search than typically developing toddlers. Developmental Science, 14(5), 980–988.
Kanwisher, N., & Wojciulik, E. (2000). Visual attention: Insights from brain imaging. Nature Reviews. Neuroscience, 1(2), 91–100.
Kornmeier, J., Wörner, R., Riedel, A., Bach, M., & Tebartz van Elst, L. (2014). A Different View on the Checkerboard? Alterations in Early and Late Visually Evoked EEG Potentials in Asperger Observers. PLOS ONE.
Kothari, R., Bokariya, P., Singh, S., & Singh, R. (2016). A Comprehensive Review on Methodologies Employed for Visual Evoked Potentials. Scientifica.
Lai, M.-C., Lombardo, M. V., & Baron-Cohen, S. (2014). Autism. Lancet (London, England), 383(9920), 896–910.
National Institute of Mental Health (2018). Autism spectrum disorder.
O’Riordan, M. A., Plaisted, K. C., Driver, J., & Baron-Cohen, S. (2001). Superior visual search in autism. Journal of Experimental Psychology, Human Perception and Performance, 27(3), 719–730.
Sayorwan, W., Phianchana, N., Permpoonputtana, K., & Siripornpanich, V. (2018). A Study of the Correlation between VEP and Clinical Severity in Children with Autism Spectrum Disorder. Autism Research and Treatment, 2018, 5093016.
Shomstein S. (2012). Cognitive functions of the posterior parietal cortex: top-down and bottom-up attentional control. Frontiers in Integrative Neuroscience, 6, 38.
Sur, S., & Sinha, V. K. (2009). Event-related potential: An overview. Industrial Psychiatry Journal, 18(1), 70–73.
Wolfe., J & Horowitz, T., (2008). Visual search. Scholarpedia, 3(7):3225
Yamasaki, T., Maekawa, T., Fujita, T., & Tobimatsu, S. (2017). Connectopathy in Autism Spectrum Disorders: A Review of Evidence from Visual Evoked Potentials and Diffusion Magnetic Resonance Imaging. Frontiers in Neuroscience, 11.
Figures