Clinical Validation of Eye Vergence as an Objective Marker for Diagnosis of ADHD in Children

Abstract

Objective: ADHD youth show poor oculomotor control. Recent research shows that attention-related eye vergence is weak in ADHD children. Method: To validate vergence as a marker to classify ADHD, we assessed the modulation in the angle of vergence of children (n = 43) previously diagnosed with ADHD while performing an attention task and compared the results with age-matched clinical controls (n = 19) and healthy peers (n = 30). Results: We observed strong vergence responses in healthy participants and weak vergence in the clinical controls. ADHD children showed no significant vergence responses. Machine-learning models classified ADHD patients (n = 21) from healthy controls (n = 21) with an accuracy of 96.3% (false positive [FP]: 5.12%; false negative [FN]: 0%; area under the curve [AUC]: 0.99) and ADHD children (n = 11) from clinical controls (n = 14) with an accuracy of 85.7% (FP: 4.5%; FN: 19.2%, AUC: 0.90). Conclusion: In combination with an attention task, vergence responses can be used as an objective marker to detect ADHD in children.

Keywords

ADHD eye vergence diagnosis children biomarker binocular

ADHD is one of a common neurodevelopmental disorder affecting 3% to 7% of school-aged children worldwide. It is characterized by a low degree of attention, a high degree of hyperactivity and impulsivity, and the inability to inhibit inappropriate actions.

The altered behavior of ADHD patients is not limited to general conduct but it is also observed at the level of saccadic eye movement behavior. ADHD patients have more difficulty suppressing saccadic eye movements when fixation is required (Karatekin & Asarnow, 1999; Munoz, Armstrong, Hampton, & Moore, 2003) and the rate of microsaccades is reported to be higher in ADHD participants compared with controls, especially in the time intervals around stimulus onset (Fried et al., 2014) and after cue presentation, although suppression has also been reported (Engbert & Kliegl, 2003). The poor eye control of ADHD patients is furthermore expressed as an increased variability in the latencies of saccadic responses to visual stimuli compared with healthy controls (Kuntsi et al., 2006; Leth-Steensen et al., 2000. Poor control of saccadic behavior has been observed in anti-saccades (Leth-Steensen et al., 2001), visually guided saccades (Leth-Steensen et al., 2001; Mostofsky, Lasker, Cutting, Denckla, & Zee, 2001; Munoz et al., 2003), and memory guided saccades (Rommelse, Van der Stigchel, & Sergeant, 2008). Moreover, whereas healthy participants show an asymmetry in eye movement control where eyes move faster when controlled by the right cerebral hemisphere, ADHD children do not show this asymmetry (Rothlind, Posner, & Schaughency, 1991). In contrast to fast saccadic eye movements, slower smooth pursuit eye movements appear to be within normal range in ADHD patients (Ross, Olincy, Harris, Sullivan, & Radant, 2000).

Besides the saccadic eye movements, disconjugate eye movements or vergence, that is, where the eyes move in opposite direction (Figure 1), are affected in ADHD patients (Solé Puig et al., 2015). This finding is of relevance as recent data provide evidence showing that eyes converge during orienting attention (Solé Puig, Pérez Zapata, Aznar-Casanova, & Supèr, 2013; Solé Puig, Puigcerver, Aznar-Casanova, & Supèr, 2013). During gaze fixation, the eyes briefly converge after the presentation of a stimulus, which indicates the location of an upcoming peripheral target but not or weakly after a stimulus that was not informative about the location of the peripheral target (Solé Puig, Pérez Zapata, et al., 2013). The strength and timing of eye vergence correlate with the onset and strength of the visual event–related potentials (vERPs) at parietal locations (Supèr, Marco, Zapata, Crespillo, & Puig, 2014). Also, stimulus contrast is associated with strength of convergence (Sole et al., 2013a) where high stimulus contrast relates to larger modulation of the angle of eye vergence. Moreover, detected targets were accompanied by eye vergence, whereas targets, which were not detected, were not (Solé Puig, Pérez Zapata, Aznar-Casanova, & Supèr, 2013).

Figure 1.

Schematic explanation of the vergence.

The diagnosis of ADHD is clinical, based upon criteria established by classification systems such as the Diagnostic and Statistical Manual of Mental Disorders (5th ed.; DSM-5; American Psychiatric Association [APA], 2013). Even though the clinical diagnosis shows considerable levels of concurrent and predictive validity (Faraone, 2005; Faraone et al., 2000), concerns persist, and additional tools are needed to support ADHD diagnosis. The observation that attention-related eye vergence is poorly present in children with ADHD (Solé Puig et al., 2015) led to the idea that evaluating eye vergence during an attention task can be used as an objective measure to support the clinical diagnosis of ADHD. A recent study reporting abnormalities in the brain structure, which control eye vergence, in ADHD patients supported this notion (Johnton et al., 2014). In fact, in a preliminary report, we were able to classify ADHD in children using vergence responses (Lorena Esposito et al., 2016). The aim of the present study was, therefore, to replicate our earlier findings and to validate eye vergence as a marker tool for classifying ADHD. We assessed the modulation of the angle of vergence in children previously diagnosed with ADHD while performing a child-friendly attention task and compared the results with the responses from age-matched clinical controls and healthy participants.

Method

Participants

Forty-three children aged between 7 and 17 years (M ± SD = 11.95 ± 3.06) diagnosed with ADHD, and 19 clinical controls (M ± SD = 11.57 ± 2.86) participated in the study. Clinical controls were children referred to the hospital for attentional and/or conduct problems but after first clinical assessment diagnosed as not having ADHD but suffering other mental problems. None of the participants was taking medication for ADHD. Patients were recruited through the Child and Adolescent Health Mental Center from the Hospital Mataró of the Consorci Sanitari del Maresme. The participants (n = 30) from the control group were healthy children showing no attention or conduct problems (M ± SD: 8.85 ± 0.49 years) recruited via a public school. In addition, a separate population of 67 children (M ± SD: 10.67 ± 2.64 years; 21 healthy controls; 14 clinical controls, and 32 ADHD children) were used for validation purposes. All the clinical diagnoses of ADHD were made by clinical psychiatrists. All cases were diagnosed using the Diagnostic and Statistical Manual of Mental Disorders (4th ed., text rev.; DSM-IV-TR; APA, 2000) including a psychiatric and psychologist interview to assess the presence of symptoms of inattention, hyperactivity, and impulsivity during the last 6 months. Also, the beginning of the symptoms before 7 years of age and the persistence of clinical dysfunction in at least two settings (school and home) were used as criteria. Furthermore, we analyzed psychopathology and comorbidity using Kiddie Schedule for Affective Disorders and Schizophrenia for School-Age Children–Present and Lifetime version (K-SADS-PL; Kaufman et al., 1997). Part of the medical examination and psychiatric evaluation of all patients for diagnosing ADHD was the inquiry about visual problems. The survey included specific questions on strabismus and accommodation insufficiency. All participants had normal or corrected-to-normal visual acuity.

Inclusion and Exclusion Criteria

The inclusion criteria were as follows: (a) between 7 and 17 years of age, (b) a diagnosis of ADHD without mental retardation, (c) Spanish mother tongue or fluency in Spanish, and (d) informed consent for the study signed by the parent or legal guardian with patient assent. The exclusion criteria were as follows: (a) a history of head injury with loss of consciousness or other neurological illness; (b) mental retardation or other significant disorders like a pervasive developmental disorder; and (c) visual or auditory problems.

Ethics Statement

Before participating in the study, written informed consent from the parents on behalf of the children enrolled in our study was obtained in accordance with the Helsinki Declaration. The study was approved by the Ethics Committee of the University of Barcelona and of Consorci Sanitari del Maresme, and the study was registered at Agencia Española de Medicamentos y Productos Sanitarios (identifier: 548/15/EC).

Apparatus

We used the BGaze system (Braingaze SL, Mataró, Spain) for presenting the visual stimuli synced with a remote eye tracker. The BGaze system includes a 30Hz binocular eye tracker (X2-30, Tobii Technology AB, Sweden). The display resolution was 1024 × 768 pixels.

Procedure

Children sat in a dimly lit room of the hospital or school, in front of the PC monitor at a distance of 55 cm. During the entire task, a chinrest was used to prevent head movements. The eye tracking equipment was calibrated (five points, binocular) for each participant at the beginning of the experiment by BGaze eye tracking software. Before starting the task, all children practiced with cue and no-cue trials (20 trials) to become familiar with the task. The entire procedure took about 12 min to complete. After testing, the saved behavioral and eye data were stored and assigned a random number. Validation was done double blinded (Figure 2).

Figure 2.

Flowchart of the clinical validation.

Visual Cue Experiment

To assess orienting visual attention, we used a paradigm in which the children were required to discriminate cartoon images of a tadpole from a fish (Figure 3). Each trial started with the presentation of a central frog (size of approx. 3 × 4 degrees visual angle) with his eyes closed, together with two small pool on either side (size of 2 × 3degrees visual angle ; eccentricity of 6 degrees visual angle ). After 500 ms fixating, the frog opened the eyes looking toward the left or right side, or straight ahead. In the former case, the frog’s gaze served as a cue to inform the child about the location (left or right) of the upcoming image of a tadpole or fish (informative cue condition). In the latter case, the child was unaware of the stimulus location (no-informative cue condition). In total, there were 128 trials. Fifty percent of the trials contained an informative cue and trials with different cue conditions were randomly interleaved. The fish or tadpole cartoon was presented for 1,500 ms. During the trial, the child was required to maintain fixation at the central frog image, and had to respond by pressing a button when a tadpole was presented and refrain from responding when a fish appeared. When the child correctly identified a tadpole, feedback was given by a small jump of the frog. The next trial started automatically at the end of a trial.

Figure 3.

Task design and images used: (a) Illustration of the different phases of the attention task and stimulus presentation times. (b) Gaze directions of the central frog act as a cue left or right.

Data Analysis

The data analysis was performed in two steps. In the first, we sought statistical proof of significant differences in the vergence signals between the different groups. This was done using the whole available sample. In the second step, we used the outcomes of the first part to build and test a machine-learning model that allowed prediction of ADHD for a given participant. For this, we split the original sample in two subsamples for extended validation purposes: one (S1, 90% total) for parameter adjustment and the other (S2, 10%) for final validation.

First part: In total, there were 3,354 (1,661 no-informative cue trials and 1,693 informative cue trials) trials in the healthy control group, 2,240 (1,133 no-informative cue trials and 1,107 informative cue trials) in the clinical group, and 5,036 (2,537 no-informative cue trials and 2,499 informative cue trials) in the ADHD group. The angle of eye vergence was calculated using the cross product of both gaze vectors. Gaze vectors correspond to the lines between the 3D eye positions and 2D gaze positions in a common coordinate system. Samples that gave a low validity score according to the Tobii eye tracker software were set as missing. In total, 10% to 20% of the samples were invalid. Low validity scores usually happen during saccades and blinks. From the remaining samples, we calculated the pointwise median of all trials for conditions and groups, separately. We chose to use the median instead of the mean to mitigate the effect of occasional outliers and thereby reduce bias. To reduce irregularities, the obtained signal was then smoothed using a moving median and thereafter a moving average with a 200 ms window. For statistical analysis, we used bootstrapping and permutation analysis to simulate pointwise the distributions of the medians.

Second part: The classification model (Figure 2) consists of two layers. In the first layer, a Radial Basis Function SVM model (gamma = 6.5) was trained and tested to separate healthy children from ADHD children using a set of three vergence features extracted from no-cue, left-cue, and right-cue signals. In the second layer, two nearest-neighbors models (1-NN and 3-NN) were used to distinguish between clinical controls and ADHD children. In the second layer, overall vergence level, vergence variations in velocity were used as features. The parameters of all three models were adjusted with a 30-fold stratified cross-validation routine over the S1 subsample, which, at each iteration, was further split into an 80-20 train-test random resampling. Then, the resulting model was tested on the S2 subsample, which so far had been unseen by it. Only participants who were not classified as healthy controls and the ones that reached poor confidence levels in the first layer entered the second layer. The final classification was based on the label that gave the highest confidence level.

The sample size was obtained after a preliminary study of statistical power, where we supposed an effect magnitude of at least d = 0.75 (Cohen’s) with a significance level of .05 and power = 0.75. We used R statistical software to simulate the distribution of the difference in medians and estimated a sample size of at least n = 40 participants. The results showed that we needed about 20 participants per group for each comparison, with at least 20 healthy controls (in this study, we have 30), 20 clinical controls (in this study, we have 19), and 20 ADHD-diagnosed participants (in this study, we have 43).

Results

Behavior

The overall behavioral reaction times to target stimuli (tadpole images) in the informative cue condition were similar for all groups (M ± SD, controls: 678.8 ± 231.1 ms, clinical controls: 644.3 ± 200.8 ms, ADHD: 664.1 ± 248.8 ms; Tukey’s rank test, all possible combinations p > .05). In the no-informative cue condition, children on average responded slower to target stimuli than in the informative cue condition. Healthy controls responded (M ± SD; 712.6 ± 210.6 ms) significantly (p < .01) slower than the children from the other two groups (M ± SD, clinical controls: 661.7 ± 194.6 ms, ADHD: 683.3 ± 240.5 ms). Between clinical controls and ADHD children, no significant differences were observed in reaction times. The variability in reaction times was different between the three groups (p < .01).

For healthy controls, the hit rates for targets were 81.7% and 81.7% and the correct rejection rates for the distractors were 77.3% and 75.9% in the informative and no-informative cue conditions, respectively. In the clinical control group, the hit rates were 60.0% and 64.2%, and the correct rejection rates were 64.3% and 62.2 %, in the informative and no-informative cue conditions, respectively. In the ADHD group, hit rates were 66.1% and 67.8%, and correct rejection rates were 67.6% and 67.6 % in the informative and no-informative cue conditions, respectively. The differences in the hit rates between healthy controls and the clinical populations were statistically significant in both cue conditions (Tukey’s rank test, all p < .05). The differences in the correct rejection rates were not significant (Tukey’s rank test, all possible combinations p > .1).

Eye Vergence Data

During the trials, the angle of eye vergence was not constant even though participants maintained gaze fixation at the central image. Notably, at the end of the trial during the stimulus period (when the target/distractor is presented), the angle of eye vergence decreased, which means that the eyes diverged (Figure 4). Around 600 ms after the onset of the target/distractor, the angle of eye vergence reached a minimum and returned to the initial baseline level. The changes in the angle of eye vergence are observed in trials belonging to the informative cue and no-informative cue conditions but they are more pronounced in the former condition. In contrast to the results obtained from healthy participants, the results from the clinical groups show remarkably less modulation in the angle of eye vergence. Especially, children of the ADHD group showed little to no changes in the angle of eye vergence during the task.

Figure 4.

Average modulation of the angle of vergence.

Cue Versus No-Cue Vergence

In the healthy and the clinical control groups, eye vergence responses are stronger (i.e., the angle of eye vergence is larger, meaning that the eyes converge) during the informative cue period compared with the vergence responses during the no-informative cue period (Figure 5). In the ADHD group, no difference in vergence responses during the cue period is observed (Figure 5).

Figure 5.

Average modulation of the angle of vergence.

Distractor Versus Target

To appreciate the relation of eye vergence to stimulus relevance, we compared the vergence responses to targets (tadpole images) with the responses to distractors (fish images). In the healthy control group, the angle of eye vergence decreased equally strongly during the stimulus period in both cases (Figure 6). In the clinical control group, eye vergence to distractors was as strong as in the healthy control group but was less noticeable in the target condition (Figure 6). In the ADHD group, for both targets and distractors, no changes in the angle of eye vergence were observed.

Figure 6.

Average modulation of the angle of vergence for distractor and target stimuli.

Classification

By applying machine-learning algorithms, we assessed whether, using features of the modulation in the angle of eye vergence, we were able to discriminate ADHD participants from the healthy and clinical controls. We used a three-step classification procedure (Figure 2). In the cross-validation stages, we obtained accuracy of 96% in Layer 1 and 88% in Layer 2. In the validation stage with unseen participants, the results show (Figure 7) that with a 96.3% precision (area under the curve [AUC]: 0.99), ADHD participants can be separated from healthy participants. The false positive rate, that is, healthy patient diagnosed as ADHD, was 5.12% and the false negative rate, that is, ADHD patients classified as healthy was 0%. The separation between clinical groups reached an accuracy of 85.7% with an AUC of 0.90. False positive and false negative rates were 4.5% and 19.23%, respectively. As behavioral performance data showed statistical significance, we added behavioral features (average response time, variability in response time, hit rate, and correct rejection rate) to the models to improve accuracy. Instead of improvement, reduced accuracies were obtained. Neither the addition of pupil size data to the model improved classification outcomes.

Figure 7.

ROC curves based on eye vergence.

Pupil Diameter

Neural circuits for controlling eye vergence partly overlap with those that control pupil size. To see if the size of the pupil changes during the attention task, we calculated pupil diameter. The results show that the diameter of the pupil increases during the task. This increase was stronger in the healthy control group than in the clinical groups. There was no significant difference in pupil diameter between the clinical groups, except in the no-informative cue condition where the clinical control group showed a stronger increase in pupil diameter compared with the pupil changes of the ADHD group (Figure 8). Because healthy participants showed different pupil modulations, we tried to classify ADHD children based on pupil size. However, classification in the validation stage gave poor results with maximal AUCs of 0.6 (Figure 9).

Figure 8.

Average modulation of pupil diameter.

Figure 9.

ROC curves based on pupil size.

Behavioral and Vergence Responses

Previous research has shown an absence of vergence responses when participants fail to detect a stimulus (Solé Puig, Pérez Zapata, et al., 2013). To know whether healthy and clinical groups differ, we compared the vergence responses with the behavioral responses (Figure 10). For correct responses to targets (hits), a strong vergence response in the healthy and clinical control groups was noticed. In contrast, when participants fail to respond to targets (misses), a clear modulation in the vergence angle was absent. For correct rejections and false alarms, healthy participants showed vergence responses, but no or weak modulation in the angle of eye vergence was seen in the clinical groups.

Figure 10.

Average modulation of the angle of vergence separated by behavioral response outcomes.

We next aligned the vergence responses at onset of the behavioral response, that is, when participant pressed the response button (Figure 11). In the healthy control group, there was a clear dip in the vergence responses centered on the behavioral response. This was also true for the clinical control group in the no-informative cue condition and for responses to distractors (false alarms). No clear response dip was observed in the informative cue condition and for hit trials. ADHD participants showed no response modulation around behavioral response onsets.

Figure 11.

Average modulation of the angle of vergence aligned on behavioral responses onset, that is, the moment of button press.

Task Duration

During the 12-min task, participants may become fatigued or bored by stimulus repetition, especially those in the ADHD group, as they have difficulties remaining focused. To assess the possible influence on vergence responses, we compared the modulation in the average vergence angle during the first half of the task and compared that with the responses of the second half of the task. There is a slightly lower modulation in the first half during the stimulus period in the healthy group and during the cue period in the ADHD group (Figure 12). In the clinical control group, no difference was seen.

Figure 12.

Average modulation of the angle of vergence according to task period.

Stimulus Laterality

Many studies report compromised brain lateralization in patients with ADHD. We therefore tested whether vergence responses show such laterality effect. We analyzed vergence to stimuli presented on the left and right sides separately. In none of the groups was there a clear difference in the modulation in the angle of eye vergence between left and right conditions (Figure 13).

Figure 13.

Average modulation of the angle of vergence according to stimulus side.

Discussion

In this study, we tested eye vergence modulation while performing an attention task as a tool to discriminate ADHD from non-ADHD children. We applied a child-friendly attention task for 10 to 12 min and recorded with a remote eye tracker eye position data from which vergence was calculated. The principal finding is that children belonging to the healthy control group showed clear modulation in the angle of eye vergence whereas children in the clinical groups showed weak to no significant modulation in the angle of eye vergence while performing the attention task. Healthy children were younger than the clinical participants. This, however, may not be relevant for the classification outcomes. At the age of 8 to 9 years, children can follow task instructions in a gaze contingency task (Paus, Babenko, & Radi, 1990). In our task, healthy children performed better than ADHD children. Also, at these ages, development of binocular coordination is completed, which can be appreciated in the observed vergence responses in healthy children.

The angle of eye vergence decreased during the stimulus period when the target/distractor stimulus was presented. This signifies that the eyes diverged. In addition, the vergence angle from the healthy control group was larger in the informative cue condition than in the no-informative cue condition during the period of cue presentation. These findings agree with our earlier observations of attention-related eye vergence in children (Solé Puig, Pérez Zapata, et al., 2013). Furthermore, we show modest vergence responses in a clinical control group. Even though children of the clinical control group showed, in general, weaker modulation in the angle of eye vergence, there was a small significant difference in vergence between the cue conditions. Children from the ADHD group, however, showed weak vergence responses and no difference in vergence between cue conditions was observed. Thus, ADHD children appear less sensitive to visual stimulation, but they are also vulnerable when orienting attention is required.

The current results confirms our previous findings showing poor attention-related eye vergence in ADHD children (Solé Puig et al., 2015) and that vergence can be a used to discriminate ADHD in children (Lorena Esposito et al., 2016). In the latter study, we evaluated four classes of supervised machine-learning classifiers (in total 138 different models) with a validation set of 232 children of a school cohort. The average accuracy of these models was 90% (minimum: 86.21%; maximum: 95.26%). In the current study, we applied a two-step classifier. This was necessary because we included a clinical cohort, which compared with healthy controls gave more subtle differences in the angle of eye vergence.

The overall weak modulation in the angle of eye vergence in ADHD children may indicate that the vergence system is less sensitive to visual stimulation. This may explain the absence of clear vergence responses in healthy controls when failing to detect the target (Solé Puig, Pérez Zapata, et al., 2013; current study). The absence of a difference between vergence responses from the informative and no-informative cue condition suggests that attentional control of eye vergence by presumably higher cortical areas is impaired or the vergence system is still immature (Bucci, Nassibi, Gerard, Bui-Quoc, & Seassau, 2012; Kirkby, Blythe, Drieghe, & Liversedge, 2011; Prado, Dubois, & Valdois, 2007).

Markers of ADHD

The clinical diagnosis of ADHD shows considerable levels of concurrent and predictive validity (Faraone, 2005; Faraone et al., 2000). Nevertheless, concerns about diagnostic accuracy persist. The diagnosis has been called “subjective” because it relies on clinician evaluation of responses from patients, parents, and/or informants. Some suggest that the use of subjective diagnostic procedures leads to the overdiagnosis of ADHD (Bruchmuller, Margraf, & Schneider, 2012; Visser et al., 2014), while other studies have raised concerns about the underdiagnosis of ADHD. In response to such concerns, researchers have sought to develop objective measures to diagnose ADHD.

Much research has examined peripheral biochemical markers. Meta-analyses of these studies indicate that five measures differentiated ADHD and control patients (Norepinephrine [NE], 3-Methoxy-4-hydroxyphenylethylene glycol [MHPG], monoamine oxidase [MAO], zinc, and cortisol; Faraone, Bonvicini, & Scassellati, 2014; Scassellati, Bonvicini, Faraone, & Gennarelli, 2012). Moreover, NE, MHPG, MAO, b-phenylethylamine, and cortisol were responsive to ADHD medications. Meta-analysis also shows that peripheral measures of oxidative stress differ between ADHD and control participants (Joseph, Zhang-James, Perl, & Faraone, 2015). Other approaches to biomarker development for ADHD have used neuropsychological (Ritsner, 2009), electroencephalographic (Snyder, Rugino, Hornig, & Stein, 2015), actigraphy (Dane, Schachar, & Tannock, 2000), structural imaging (Silk, Vance, Rinehart, Bradshaw, & Cunnington, 2009), and functional imaging (Bush, Valera, & Seidman, 2005) methods. Continuous performance tests (CPTs; for example, Corkum & Siegel, 1993; Homack & Riccio, 2006; Riccio & Reynolds, 2001) have been evaluated in many studies.

Most methods that are currently being used to support the clinical diagnosis quantify symptoms of ADHD. The AULA Nesplora (AULA) is a CPT that shows clear correlations with the Conners’ CPT (Díaz-Orueta et al., 2014). AULA reports >90% accuracy at their company website, and may be useful in establishing a differential ADHD diagnosis (Areces, Rodríguez, García, Cueli, & González-Castro, 2018). Quotient reports a cross-validated AUC of 0.716 (unpublished data at company website) based on comparing ADHD patients with healthy controls. They report higher receiver operating characteristic curves (ROCs) that have not been cross-validated. The ability of the QbTest to identify ADHD in children is moderate with sensitivity ranging from 47% to 67% and specificity from 72% to 84% (Hult, Kadesjö, Kadesjö, Gillberg, & Billstedt, 2018; Reh et al., 2015). The adult version of the QbTest gives similar results with an overall correct classification of 72.1% (Edebol, Helldin, Holmberg, & Gustafsson, 2011; Edebol, Helldin, & Norlander, 2013; Söderström, Pettersson, & Nilsson, 2014). Measuring body movements with wireless inertial sensors gives a classification of >95% (O’Mahonya, Florentino-Liano, Carballo, Baca-García, & Artés Rodríguez, 2014) and classification based on microsaccades yields an accuracy of 70% (sensitivity: 59%; specificity: 82%; Fried et al., 2014). The Neuropsychiatric EEG-Based Assessment Aid (NEBA) is an EEG brainwave test for ADHD based on theta/beta ratio (TBR). Integration of NEBA outcomes with a clinician’s ADHD evaluations improves diagnostic accuracy from 61% to 88% (Snyder et al., 2015). A study examining the TBR in ADHD and normal children reported a sensitivity of 0.86 and a specificity of 0.57 (Sangal & Sangal, 2015).

So far, no method has shown sufficient sensitivity and specificity when predicting “gold standard” clinical diagnoses of the disorder. According to the task force of the World Federation of ADHD (WF-ADHD), a useful biomarker needs to meet the following criteria (Thome et al., 2012): sensitivity exceeding 80%, specificity exceeding 80%, the putative biomarker is reliable, reproducible, inexpensive, noninvasive, easy to use, and has been confirmed by at least two independent studies. The use of eye vergence recordings to classify ADHD meets all of the WF-ADHD criteria but needs to be confirmed by additional studies. So far, the marker could be a diagnostic support tool to complement clinical ADHD diagnosis. The marker tool could demonstrate attention deficits in children without showing clear symptomology or vice versa. It could be useful for the assessment of borderline cases when ADHD diagnosis is difficult to make. Also, the objectivity may help to convince patients of the outcome of the clinical diagnosis and to adhere to their treatment protocols. Furthermore, as the method assesses attention, it may be helpful for diagnosing children who show little hyperactivity symptoms—for instance, girls. The test uses a child-friendly frog game, and the outcomes alone could be used to support ADHD diagnosis but classification accuracy remains to be tested.

Binocular Vision Difficulties in ADHD

The finding of altered vergence in ADHD is not surprising given that binocular vision in ADHD children is impaired. Convergence insufficiency (CI), which is a common binocular disorder characterized by the inability to obtain a single visual field while working at a near distance (Borsting et al., 2011; Borsting, Rouse, & Chu, 2005; Borsting et al., 2003; Rouse et al., 1999), is prevalent in children with ADHD (Granet, Gomi, Ventura, & Miller-Scholte, 2005), and has been shown to relate to attention problems (Borsting et al., 2003; Poltavski, Biberdorf, & Petros, 2012; Rouse et al., 2009). The primary source of CI symptoms may be accommodative insufficiency (Marran, De Land, & Nguyen, 2006) and thus occurs at close distances. In our study, the target distances fell well outside the range of distances of CI for children. This means that our observed disruption in vergence modulation does not reflect CI but represents a novel role of eye vergence in visual attention (Solé Puig, Pérez Zapata, et al., 2013; Solé Puig, Puigcerver, et al., 2013).

Role for Vergence in Attention

Top down attention originates in the frontal cortex (Arnsten & Rubia, 2012; Bisley, 2011; de Zeeuw, Mandl, Hulshoff Pol, van Engeland, & Durston, 2013; Thompson, Biscoe, & Sato, 2005) and reduced or distorted activation in prefrontal regions of ADHD patients has been observed (Bush, 2011; Di Michele, Prichep, John, & Chabot, 2005; Kraina & Castellanos, 2006; Soliva, 2011; Valera, Faraone, Murray, & Seidman, 2007). The frontal cortex controls eye vergence (Gamlin & Yoon, 2000) and may be the source of our observed attention-related vergence. Thus, reduced functionality of the frontal cortex in ADHD patients may produce a distorted modulation in attention-related vergence. The frontal cortex is connected to the reticular formation in the brainstem, where premotor neurons reside that control eye vergence (Chaturvedi & Van Gisbergen, 2000; Coubard, 2013; Gamlin, 2002; Judge & Cumming, 1986; Mays, 1984; Suzuki, Suzuki, & Ohtsuka, 2004). The reticular formation of the brainstem forms part of a broader pathway, including the frontal and parietal regions of the cerebral cortex (Alvarez, Jaswal, Gohel, & Biswal, 2014; Gamlin & Yoon, 2000; Gnadt & Mays, 1995) and cerebellum (Alvarez et al., 2014; Nitta, Akao, Kurkin, & Fukushima, 2008; Versino, Hurko, & Zee, 1996). These structures, which form part of the attention system of the brain, also are involved in the control of vergence. This suggests that the neural circuits controlling vergence and attention are closely linked.

Study Limitations and Future Research

Together with our earlier study (Solé Puig et al., 2015), this is the second study validating attention-related vergence in children as an objective marker for ADHD diagnosis. In this study, we have 19.2% false negative cases which are ADHD children classified as not-ADHD. The relatively high number of false negatives may be because of shared symptomology. The methods assess attentional processing, which is impaired in ADHD patients but also in the clinical controls, which were patients referred to the hospital for learning and/or conduct problems. ADHD is a multifaceted neurodevelopment disorder and therefore more testing is desirable using different cognitive tasks. The ability of eye vergence to classify ADHD presentation “specifiers” (subtypes) and the effect of medication on vergence responses needs to be investigated in future studies. The findings of disrupted eye vergence in ADHD children agree with the general impression of poor binocular control in children with attention problems. However, further studies are needed to assess role of eye vergence in attention.

Conclusion

Our observations show that attention-related vergence differs between healthy controls, clinical controls, and children with ADHD. Solely based on features of the modulation in the angle of eye vergence, we were able to classify ADHD from healthy and clinical control participants with high precision. The results therefore clearly demonstrate that assessment of vergence during a child-friendly attention task is a useful, observer-independent tool supporting clinical diagnosis of child ADHD.

Footnotes

Acknowledgements

The authors thank Prof. Dr. E. Baca-García for his helpful comments on a previous version of the article.

Declaration of Conflicting Interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: F.L.E. and I.M. are employees of Braingaze, and H.S. is co-founder of Braingaze. J.A.R.Q was on the speakers’ bureau and/or acted as consultant for Eli-Lilly, Janssen-Cilag, Novartis, Shire, Lundbeck, Almirall, Braingaze, Sincrolab, and Rubió.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by Industrial Doctorates grants to F.L.E. and I.M. from Agència de Gestió d’Ajuts Universitaris i de Recerca; financial support from Centro para el Desarrollo Tecnológico Industrial, Spain, to Braingaze; and grants of the Spanish Ministry (PSI2014-57454-P) to H.S. and (PSI2015-63965) to L.E

ORCID iD

Alexandre Perera i Lluna

Author Biographies

Paloma Varela Casal is licenced in Medicine and specialized in Psychiatry. Currently working at Mataró Hospital as child and adolescent psychiatrist, and she is head of the Department of Child and Adolescent Psychiatry at the Hospital de Mataró.

Flavia Lorena Esposito (MSc) studied Psychology and is a PhD student at the Faculty of Psychology of the Univeristy of Barcelona.

Imanol Morata Martínez (MSc) studied Mathematics and is PhD student at the Politecnical University of Catalonia.

Alba Capdevila works as a research assistant at the Department of Child and Adolescent Psychiatry at the Hospital de Mataró.

Maria Solé Puig (PhD) is psychologist and works as a postdoc at the Psychology Faculty of the University of Barcelona.

Núria de la Osa (PhD) is a clinical psychologist and specialist in Language Therapist. PhD in Psychology, associate professor of Diagnostic and Psychological Assessment of Child and Adolescents in the Universitat Autònoma de Barcelona.

Lourdes Ezpeleta (PhD) is a clinical psychologist. Professor of child and adolescent psychopathology at the Universitat Autònoma de Barcelona. She is also the Director of the Unit of Epidemiology and Diagnostic in Developmental Psychopathology.

Alexandre Perera i Lluna (PhD) is an associate professor at the Politecnical University of Catalonia where he leads the Bioinformatics and Biomedical Signals Lab. He is also a researcher at Sant Joan de Déu Research Foundation.

Stephen V. Faraone (PhD) is a distinguished professor in the Departments of Psychiatry and Neuroscience & Physiology at SUNY Upstate Medical University. He is also Senior Scientific Advisor to the Research Program Pediatric Psychopharmacology at the Massachusetts General Hospital and a lecturer at Harvard Medical School.

Josep Antoni Ramos-Quiroga (MD) is head of Department of Psychiatry and President of Innovation Commission at Hospital Universitari Vall d’Hebron in Barcelona and Professor of Psychiatry at Department of Psychiatry and Legal Medicine of the Universitat Autònoma de Barcelona.

Hans Supèr (PhD in Neurobiology) is ICREA researcher at the Psychology Faculty of the Unversity of Barcelona. He is also co-founder and CTO of Braingaze

Jose Cañete (MD) is director of the department of Child and Adolescent Psychiatry at the Hospital Mataro of the Maresme Health Consortium. He is associate professor at the Faculty of Psychology of the Blanquerna-University Ramon Llull.

References

Alvarez

T. L.

Jaswal

Gohel

Biswal

B. B.

(2014). Functional activity within the frontal eye fields, posterior parietal cortex, and cerebellar vermis significantly correlates to symmetrical vergence peak velocity: An ROI-based, fMRI study of vergence training. Frontiers in Integrative Neuroscience, 8, 50. doi:10.3389/fnint.2014.00050

American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (4th ed., text rev.). Washington, DC: Author.

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). Arlington, VA: American Psychiatric Publishing.

Areces

Rodríguez

García

Cueli

González-Castro

(2018). Efficacy of a continuous performance test based on virtual reality in the diagnosis of ADHD and its clinical presentations. Journal of Attention Disorders, 22, 1081-1091.

Arnsten

A. F.

Rubia

(2012). Neurobiological circuits regulating attention, cognitive control, motivation, and emotion: Disruptions in neurodevelopmental psychiatric disorders. Trends in Cognitive Science, 51, 356-367.

Bisley

J. W.

(2011). The neural basis of visual attention. Journal Physiology, 589, 49-57.

Borsting

E. J.

Mitchell

G. L.

Scheiman

Chase

Kulp

Cotter

(2012). CITT-RS Study Group. Improvement in academic behaviors and attention following successful treatment of convergence insufficiency. Optometry and Vision Science,89(1), 12-8.

Borsting

E. J.

Rouse

M. W.

Chu

(2005). Measuring ADHD behaviors in children with symptomatic accommodative dysfunction or convergence insufficiency: A preliminary study. Optometry, 76, 588-592.

Borsting

E. J.

Rouse

M. W.

Mitchell

G. L.

Scheiman

Cotter

S. A.

Cooper

. . . London

(2003). Validity and reliability of the revised convergence insufficiency symptom survey in children aged 9 to 18 years. Optometry and Vision Science, 80, 832-838.

10.

Bruchmuller

Margraf

Schneider

(2012). Is ADHD diagnosed in accord with diagnostic criteria? Overdiagnosis and influence of client gender on diagnosis. Journal of Consulting and Clinical Psychology, 80, 128-138.

11.

Bucci

M. P.

Nassibi

Gerard

C. L.

Bui-Quoc

Seassau

(2012). Immaturity of the oculomotor saccade and vergence interaction in dyslexic children: Evidence from a reading and visual search study. PLoS ONE, 7(3):e33458.

12.

Bush

(2011). Cingulate, frontal, and parietal cortical dysfunction in attention-deficit/hyperactivity disorder. Biological Psychiatry, 69, 1160-1167.

13.

Bush

Valera

E. M.

Seidman

L. J.

(2005). Functional neuroimaging of attention-deficit/hyperactivity disorder: A review and suggested future directions. Biological Psychiatry, 57, 1273-1284.

14.

Chaturvedi

Van Gisbergen

J. A.

(2000). Stimulation in the rostral pole of monkey superior colliculus: Effects on vergence eye movements. Experimental Brain Research., 32, 72-78.

15.

Corkum

P. V.

Siegel

L. S.

(1993). Is the continuous performance task a valuable research tool for use with children with attention-deficit-hyperactivity disorder? Journal of Child Psychology and Psychiatry, 34, 1217-1239.

16.

Coubard

O. A.

(2013). Saccade and vergence eye movements: A review of motor and premotor commands. European Journal of Neuroscience, 38, 3384-3397. doi:10.1111/ejn.12356

17.

Dane

A. V.

Schachar

R. J.

Tannock

(2000). Does actigraphy differentiate ADHD subtypes in a clinical research setting? Journal of the American Academy of Child & Adolescent Psychiatry, 39, 752-760.

18.

de Zeeuw

Mandl

R. C.

Hulshoff Pol

H. E.

van Engeland

Durston

. (2013). Decreased frontostriatal microstructural organization in attention deficit/hyperactivity disorder. Human Brain Mapping, 33, 1941-1951.

19.

Díaz-Orueta

Garcia-López

Crespo-Eguílaz

Sánchez-Carpintero

Climent

Narbona

(2014). AULA virtual reality test as an attention measure: Convergent validity with Conners’ Continuous Performance Test. Child Neuropsychology, 20, 328-342. doi:10.1080/09297049.2013.792332

20.

Di Michele

Prichep

John

E. R.

Chabot

R. J

. (2005). The neurophysiology of attention-deficit/hyperactivity disorder. International Journal of Psychophysiology, 58, 81-93.

21.

Engbert

Kliegl

(2003). Microsaccades uncover the orientation of covert attention. Vision Research, 43, 1035-1045.

22.

Edebol

Helldin

Holmberg

Gustafsson

(2011). In search for objective measures of hyperactivity, impulsivity and inattention in adult attention deficit hyperactivity disorder using the Quantified Behavior Test Plus. Europe’s Journal of Psychology, 7, 443-457.

23.

Edebol

Helldin

Norlander

(2013). Measuring adult attention deficit hyperactivity disorder using the Quantified Behavior Test Plus. PsyCh Journal, 2, 48-62.

24.

Faraone

S. V.

(2005). The scientific foundation for understanding attention-deficit/hyperactivity disorder as a valid psychiatric disorder. European Child & Adolescent Psychiatry, 14, 1-10.

25.

Faraone

S. V.

Biederman

Spencer

Wilens

Seidman

L. J.

Mick

Doyle

A. E.

(2000). Attention-deficit/hyperactivity disorder in adults: An overview. Biological Psychiatry, 48, 9-20.

26.

Faraone

S. V.

Bonvicini

Scassellati

(2014). Biomarkers in the diagnosis of ADHD—Promising directions. Current Psychiatry Reports, 16(1), 497.

27.

Fried

Tsitsiashvili

Bonneh

Y. S.

Sterkin

Wygnanski-Jaffe

Epstein

Polat

(2014). ADHD subjects fail to suppress eye blinks and microsaccades while anticipating visual stimuli but recover with medication. Vision Research, 101, 62-72.

28.

Gamlin

P. D.

(2002). Neural mechanisms for the control of vergence eye movements. Annals of the New York Academy of Sciences, 956, 264-272.

29.

Gamlin

P. D.

Yoon

(2000). An area for vergence eye movement in primate frontal cortex. Nature, 407, 1003-1007.

30.

Gnadt

J. W.

Mays

L. E.

(1995). Neurons in monkey parietal area LIP are tuned for eye-movement parameters in three-dimensional space. Journal of Neurophysiology, 73, 280-297.

31.

Granet

D. B.

Gomi

C. F.

Ventura

Miller-Scholte

(2005). The relationship between convergence insufficiency and ADHD. Strabismus, 13, 163-168.

32.

Homack

Riccio

C. A.

(2006). Conners’ Continuous Performance Test (2nd ed.; CCPT-II). Journal of Attention Disorders, 9, 556-558.

33.

Hult

Kadesjö

Gillberg

Billstedt

(2018). ADHD and the QbTest: Diagnostic validity of QbTest. Journal of Attention Disorders, 22, 1074-1080.

34.

Johnton

B. A.

Mwangi

Matthews

Coghill

Konrda

Steele

J. D.

(2014). Brainstem abnormalities in attention deficit hyperactivity disorder support high accuracy individual diagnostic classification. Human Brain Mapping, 35, 5179-5189.

35.

Joseph

Zhang-James

Perl

Faraone

S. V.

(2015). Oxidative stress and attention deficit hyperactivity disorder: A meta-analysis. Journal of Attention Disorders, 19, 915-924.

36.

Judge

S. J.

Cumming

B. G.

(1986). Neurons in the monkey midbrain with activity related to vergence eye movement and accommodation. Journal of Neurophysiology, 55, 915-930.

37.

Karatekin

Asarnow

R. F.

(1999). Exploratory eye movements to pictures in childhood-onset schizophrenia and attention-deficit/hyperactivity disorder (ADHD). Journal of Abnormal Child Psychology, 27, 35-49.

38.

Kaufman

Birmaher

Brent

Rao

Flynn

Moreci

. . . Ryan

(1997). Schedule for Affective Disorders and Schizophrenia for School-Age Children-Present and Lifetime version (K-SADS-PL): Initial reliability and validity data. Journal of the American Academy of Child & Adolescent Psychiatry, 36, 980-988.

39.

Kirkby

J. A.

Blythe

H. I.

Drieghe

Liversedge

S. P.

(2011). Reading text increases binocular disparity in dyslexic children. PLoS ONE, 6(11), e27105.

40.

Kraina

Castellanos

F. X.

(2006). Brain development and ADHD. Clinical Psychology Review, 26, 433-444.

41.

Kuntsi

Rogers

Swinard

Börger

van der Meere

Rijsdijk

Asherson

(2006). Reaction time, inhibition, working memory and “delay aversion” performance: Genetic influences and their interpretation. Psychological Medicine, 36, 1613-1624.

42.

Leth-Steensen

Elbaz

Z. K.

Douglas

V. I.

(2000). Mean response times, variability, and skew in the responding of ADHD children: A response time distributional approach. Acta Psychologica, 104, 167-190.

43.

Lorena Esposito

Varela

Ozan Alkan

Morata

de la Osa

Ezpeleta

. . . Supèr

. (2016). Validation of BGaze method supporting ADHD diagnosis. European Psychiatry, 33, S135

44.

Marran

L. F.

De Land

P. N.

Nguyen

A. L.

(2006). Accommodative insufficiency is the primary source of symptoms in children diagnosed with convergence insufficiency. Optometry and Vision Science, 83, 281-289.

45.

Mays

L. E.

(1984). Neural control of vergence eye movements: Convergence and divergence neurons in midbrain. Journal of Neurophysiology, 51, 1091-1108.

46.

Mostofsky

S. H.

Lasker

A. G.

Cutting

L. E.

Denckla

M. B.

Zee

D. S.

(2001). Oculomotor abnormalities in attention deficit hyperactivity disorder: A preliminary study. Neurology, 57, 423-430.

47.

Munoz

D. P.

Armstrong

I. T.

Hampton

K. A.

Moore

K. D.

(2003). Altered control of visual fixation and saccadic eye movements in attention-deficit hyperactivity disorder. Journal of Neurophysiology, 90, 503-514. doi:10.1152/jn.00192.2003

48.

Nitta

Akao

Kurkin

Fukushima

(2008). Vergence eye movement signals in the cerebellar dorsal vermis. Progress in Brain Research, 171, 173-176. doi:10.1016/S0079-6123(08)00623-7

49.

O’Mahonya

Florentino-Liano

Carballo

J. J.

Baca-García

Artés Rodríguez

(2014). Objective diagnosis of ADHD using IMUs. Medical Engineering & Physics, 36, 922-926.

50.

Paus

Babenko

Radi

(1990). Development of an ability to maintain verbally instructed central gaze fixation studied in 8- to 10-year-old children. International Journal of Psychophysiology, 10, 53-61.

51.

Poltavski

D. V.

Biberdorf

Petros

T. V.

(2012). Accommodative response and cortical activity during sustained attention. Vision Research, 63, 1-8.

52.

Prado

Dubois

Valdois

(2007). The eye movements of dyslexic children during reading and visual search: Impact of the visual attention span. Vision Research, 47, 2521-2530.

53.

Reh

Schmidt

Lam

Schimmelmann

B. G.

Hebebrand

Rief

Christiansen

(2015). Behavioral assessment of core ADHD symptoms using the QbTest. Journal of Attention Disorders, 19, 1034-1045. doi:10.1177/1087054712472981

54.

Riccio

C. A.

Reynolds

C. R.

(2001). Continuous performance tests are sensitive to ADHD in adults but lack specificity: A review and critique for differential diagnosis. Annals of the New York Academy of Sciences, 931, 113-139.

55.

Ritsner

M. S.

(2009). The handbook of neuropsychiatric biomarkers, endophenotypes and genes, Volume 1: Neuropsychological endophenotypes and biomarkers. Dordrecht, The Netherlands: Springer.

56.

Rommelse

N. N. J.

Van der Stigchel

Sergeant

J. A.

(2008). A review on eye movement studies in childhood and adolescent psychiatry. Brain and Cognition, 68, 391-414.

57.

Ross

R. G.

Olincy

Harris

J. G.

Sullivan

Radant

(2000). Smooth pursuit eye movements in schizophrenia and attentional dysfunction: Adults with schizophrenia, ADHD, and a normal comparison group. Biological Psychiatry, 48, 197-203.

58.

Rothlind

J. C.

Posner

M. I.

Schaughency

E. A.

(1991). Lateralized control of eye movements in attention deficit hyperactivity disorder. Journal of Cognitive Neuroscience, 3, 377-381.

59.

Rouse

M. W.

Borsting

Hyman

Hussein

Cotter

S. A.

Flynn

. . . De Land

P. N.

(1999). Frequency of convergence insufficiency among fifth and sixth graders: The Convergence Insufficiency and Reading Study (CIRS) group. Optometry and Vision Science, 76, 643-649.

60.

Rouse

M. W.

Borsting

Mitchell

G. L.

Kulp

M. T.

Scheiman

Amster

. . . Gallaway

(2009). Academic behaviors in children with convergence insufficiency with and without parent-reported ADHD. Optometry and Vision Science, 86, 1169-1177.

61.

Sangal

R. B.

Sangal

J. M.

(2015). Use of EEG Beta-1 Power and Theta/Beta Ratio over Broca’s Area to confirm diagnosis of attention deficit/hyperactivity disorder in children. Clinical EEG and Neuroscience, 46, 177-182. doi:10.1177/1550059414527284

62.

Scassellati

Bonvicini

Faraone

S. V.

Gennarelli

(2012). Biomarkers and attention-deficit/hyperactivity disorder: A systematic review and meta-analyses. Journal of the American Academy of Child & Adolescent Psychiatry, 51, 1003-1019.e20.

63.

Silk

T. J.

Vance

Rinehart

Bradshaw

J. L.

Cunnington

(2009). White-matter abnormalities in attention deficit hyperactivity disorder: A diffusion tensor imaging study. Human Brain Mapping, 30, 2757-2765.

64.

Snyder

S. M.

Rugino

T. A.

Hornig

Stein

M. A.

(2015). Integration of an EEG biomarker with a clinician’s ADHD evaluation. Brain and Behavior, 5(4):e00330. doi:10.1002/brb3.330

65.

Söderström

Pettersson

Nilsson

K. W.

(2014). Quantitative and subjective behavioural aspects in the assessment of attention-deficit hyperactivity disorder (ADHD) in adults. Nordic Journal of Psychiatry, 68, 30-37. doi:10.3109/08039488.2012.762940

66.

Solé Puig

Pérez Zapata

Aznar-Casanova

J. A.

Supèr

H. A

. (2013). A role of eye vergence in covert attention. PLoS ONE, 8(1), e52955. doi:10.1371/journal.pone.0052955

67.

Solé Puig

Pérez Zapata

Puigcerver

Esperalba Iglesias

Sanchez Garcia

Romeo

. . . Supèr

. (2015). Attention-related eye vergence measured in children with attention deficit hyperactivity disorder. PLoS ONE, 10(12), e0145281. doi:10.1371/journal.pone.0145281

68.

Solé Puig

Puigcerver

Aznar-Casanova

J. A.

Supèr

H. A

. (2013). Difference in visual processing assessed by eye vergence movements. PLoS ONE, 8(9), e72041. doi:10.1371/journal.pone.0072041

69.

Soliva

J. C.

(2011). Neuroimaging in the diagnosis of ADHD: Where we are and where we are going. Expert Opinion on Medical Diagnostics, 5, 307-318.

70.

Supèr

Marco

Zapata

J. P.

Crespillo

J. C.

Puig

M. S.

(2014). Role of vergence during eye fixation in orienting visual attention. Journal of Vision, 14(10), 9.

71.

Suzuki

Ohtsuka

(2004). Convergence eye movements evoked by microstimulation of the rostral superior colliculus in the cat. Neuroscience Research, 49, 39-45.

72.

Thome

Ehlis

A. C.

Fallgatter

A. J.

Krauel

Lange

K. W.

Riederer

. . . Gerlach

(2012). Biomarkers for attention-deficit/hyperactivity disorder (ADHD): A consensus report of the WFSBP task force on biological markers and the World Federation of ADHD. The World Journal of Biological Psychiatry, 13, 379-400.

73.

Thompson

K. G.

Biscoe

K. L.

Sato

T. R.

(2005). Neuronal basis of covert spatial attention in the frontal eye field. Journal Neuroscience, 25, 9479-9487.

74.

Valera

E. M.

Faraone

S. V.

Murray

K. E.

Seidman

L. J.

(2007). Meta-analysis of structural imaging findings in attention-deficit/hyperactivity disorder. Biological Psychiatry, 61, 1361-1369.

75.

Versino

Hurko

Zee

D. S.

(1996). Disorders of binocular control of eye movements in patients with cerebellar dysfunction. Brain, 119, 1933-1950.

76.

Visser

S. N.

Danielson

M. L.

Bitsko

R. H.

Holbrook

J. R.

Kogan

M. D.

Ghandour

R. M.

. . . Blumberg

S. J.

(2014). Trends in the parent-report of health care provider-diagnosed and medicated attention-deficit/hyperactivity disorder: United States, 2003-2011. Journal of the American Academy of Child & Adolescent Psychiatry, 53, 34-46.e2.