Visuospatial attentive capabilities and saccadic inhibitory control in children with spastic cerebral palsy

Cerebral palsy (CP) is a non-progressive syndrome due to a pre- or perinatal brain injury. Although CP is defined as a movement disorder, it frequently involves attentional and executive impairments, as well as specific learning disabilities. Seven children (5 males and 2 females aged 9-16 years) with spastic CP (CPC) and 13 typically developing children (TDC) (6 males and 7 females aged 9-16 years) participated in the study. Six CPC suffered from right-sided hemiplegia, one was diplegic. Participants had normal IQ and normal or corrected-to-normal visual acuity. CPC did not show visual field defects within ± 10° of visual angle, had normal verbal comprehension and were able to sit independently. Eye movements were recorded by an infrared eye-tracking system (Tobii X120, Sweden), while subjects performed a visually-guided saccade task and a Posner cueing task. A square-shaped grey placeholder was displayed in each quadrant of the visual field at 7° of eccentricity from a central cross, on a black background of a pc screen placed at 80 cm in front of the subject. Participants had to make a saccade as fast as possible to a green target occurring inside one of the placeholders. In the cueing task, a non-informative cue (brief place-holder flash) unpredictably occurred 150 ms before the imperative target, either at the same (valid condition) or at a different (invalid condition) location. No differences in latency and amplitude were found between visually-guided saccades of both groups. By contrast, CPC had a great difficulty to suppress saccades towards task-irrelevant targets both during fixation and in the cuing task. Thus, the expectancy of a relevant target during central fixation elicited in 4 CPC saccades towards a place-holder in 20-60% of the trials (<15% in TDC). Furthermore, while TDC made inappropriate saccades to the cue in only 0-30% of the trials, 5 of 6 CPC did so in a much larger percentage of trials (up to 89%), with a remarkable preponderance for a specific quadrant or hemifield. Interestingly, we found no relationship between the hemifield of prevalent saccadic intrusions and the affected limbs by CP. In addition, in CPC we observed a significant correlation between the mean latency of visually-guided saccades and the percentage of trials in which saccades were erroneously made to the cue. Finally, by taking into account only the correct trials in the Posner task, in both groups saccade latencies were faster in the valid than in the invalid condition. This study shows that in CP, even in presence of a mild symptomatology, prefrontal inhibitory executive control is frequently impaired, in the absence of deficits in the low-level visuospatial attentive capabilities.