Paper-based motivation-to-eat visual analogue scales (VASs) developed for adults are widely used in the pediatric age range. The VAS is comprised of four domains: hunger, fullness, desire to eat, and prospective food consumption. The purpose of the present study was to determine agreement between the traditional paper-based VAS and a novel digital VAS (with and without images), as well as the novel digital VAS’s predictive validity for subsequent food intake (FI) in 9–14-year-old children and adolescents. Following an overnight fast and 3 h after consuming a standardized breakfast at home, children and adolescents (n = 17) completed three different VAS instruments (VASpaper, VASimages, VASno-images) in a randomized order at five time-points: 0 min (baseline), 5 min (immediately after consuming a 147 kcal yogurt treatment), 20 min, 35 min (immediately before an ad libitum lunch), and 65 min (immediately post ad libitum lunch). All three instruments were comparable, as shown by low bias and limits of agreement on Bland–Altman plots, moderate to excellent intraclass correlation coefficients for all domains at all time-points (ICC = 0.72–0.98), and no differences between the incremental area under the curve for any of the domains. All three instruments also showed good predictive validity for subsequent FI, with the strongest relationship observed immediately before the ad libitum lunch (p = 0.56–0.63). There was no significant association between subjective thirst and water intake, except with VASno-images at baseline (r = 0.49, p = 0.046). In conclusion, the present study suggests that a novel image-based digital VAS evaluating motivation-to-eat is interchangeable with the traditional paper-based VAS, and provides good predictive validity for next-meal FI in 9–14-year-old normal weight children and adolescents.