Background: Attention deficits are found in children at familial high risk of schizophrenia (FHR-SZ) and bipolar disorder (FHR-BP) using assessment methods relying on motor-based response latency. This study compares visual attention functions in children at FHR-SZ or FHR-BP with controls using an unspeeded task unconfounded by motor components. Methods: Visual attention was assessed in 133 7-year-old children at FHR-SZ (N = 56) or FHR-BP (N = 32), and controls (N = 45) using the unspeeded paradigm, TVA-based whole report. We compared four parameters of visual attention: visual processing speed, visual short-term memory, threshold for visual perception, and error rate. Further, we investigated their potential relationships with severity of psychopathology, adequacy of the home environment, and neurocognitive measures. Results: Children at FHR-SZ displayed significant deficits in perceptual processing speed of visual attention compared with controls (p < .001; d = 0.75) as did children at FHR-BP (p < .05; d = 0.54). Visual processing speed was significantly associated with spatial working memory (beta = -0.23; t(68) = -3.34, p = .01) and psychomotor processing speed (beta = 0.14, t(67) = 2.11, p < .05). Limitations: Larger group sizes would have permitted inclusion of more predictors in the search for neurocognitive and other factors associated with the parameters of TVA-based whole report. Conclusions: Young children at FHR-SZ and FHR-BP display significant deficits in processing speed of visual attention, which may reflect the effect of shared vulnerability risk genes. Early identification of children at FHR-SZ and FHR-BP with perceptual processing speed impairments may represent a low-cost basis for low-risk interventions.