Evidence towards a continuum of impairment across neurodevelopmental disorders from basic ocular-motor tasks

Abstract

Findings of genetic overlap between Schizophrenia, Attention-Deficit/Hyperactivity Disorder (ADHD) and Autism Spectrum Disorder (ASD) contributed to a renewed conceptualization of these disorders as laying on a continuum based on aetiological, pathophysiological and neurodevelopmental features. Given that cognitive impairments are core to their pathophysiology, we compared patients with schizophrenia, ADHD, ASD, and controls on ocular-motor and manual-motor tasks, challenging crucial cognitive processes. Group comparisons revealed inhibition deficits common to all disorders, increased intra-subject variability in schizophrenia and, to a lesser extent, ADHD as well as slowed processing in schizophrenia. Patterns of deviancies from controls exhibited strong correlations, along with differences that posited schizophrenia as the most impaired group, followed by ASD and ADHD. While vector correlations point towards a common neurodevelopmental continuum of impairment, vector levels suggest differences in the severity of such impairment. These findings argue towards a dimensional approach to Neurodevelopmental Disorders' pathophysiological mechanisms.

Figure 1

The graph represents the percentage of inhibition errors across tasks on which clinical groups most deviated from healthy controls, namely commission errors from the GNG task (“COM_GNG”), anticipatory saccades from the MEM task (“AS_MEM”) and regular direction errors from the ANT task (“DE_ANT”). The error bars represent one standard error of the mean. The effect sizes from the between-group subject ANOVA are also reported here. COM_GNG: η_p² = 0.098, DE_ANT: η_p² = 0.139, AS_MEM: η_p² = 0.148.

Figure 2

(a) Standard deviation and (b) Mean of RT for correct response from GNG, PRO, ANT, MEM. Error bars represent one standard error of the mean. The effect sizes from the between-group subject ANOVA are also reported here. Standard deviation of RT: GNG: η_p² = 0.117, PRO: η_p² = 0.041, ANT: η_p² = 0.103, MEM: η_p² = 0.04. Mean RT: GNG: η_p² = 0.069, PRO: η_p² = 0.026; ANT: η_p² = 0.096; MEM: η_p² = 0.068.

Figure 3

The four plots show the RT distributions in the four groups in GNG. The RT are plotted in the x-axis and the density in the y-axis. The ex-Gaussian model fit on the RT distribution is also plotted (red line), and is calculated based on the average of each ex-Gaussian parameter from all participants, separately for the three parameters and the four groups. Graphs were created using the ggplot2 package for the open-source statistical package RStudio. The effect sizes from the between-group subject ANOVA are also reported here. Mu: η_p² = 0.026, Sigma: η_p² = 0.041, Tau: η_p² = 0.093.

Figure 4

Vectors of deviancy from controls based on multiple comparisons of each clinical group versus TD on all dependent variables. *SD standard deviation, *AS anticipatory saccades, *Saccades No-D FIX block without distractors, *Saccades D FIX block with distractors.