The use of eye tracking to assess reading fluency has been proposed as a novel and efficient screening method for identifying school children with atypical reading development or risk of dyslexia. Currently, however, little is known about the relationship between the predictive outcomes produced by eye tracking screening systems and specialist cognitive assessments. Here we investigate this relationship in the context of a neuropsychological case study involving eight subjects (9–10 years) who were identified as being at risk of dyslexia by an eye tracking screening system. We analyze to what extent these subjects displayed cognitive limitations when assessed with a neuropsychological test battery, and to what extent cognitive difficulties were limited to reading and decoding, or whether they extended to other domains as well. As a group, the subjects performed on average significantly worse in reading and decoding than in other cognitive domains. Five subjects performed more than 1.25 standard deviations below the mean for age. In line with empirical evidence, co-occurring difficulties in the attention domain were also observed in a few struggling readers. Overall, the results support the view that eye tracking can be a useful tool to assess and monitor children’s reading development during the early school years.