Keywords Cerebral visual impairment; Visuoperceptual profile; Serious game-based therapy; Functional vision; Individualized; Adaptive Highlights * Individualized adaptive serious games were developed for cerebral visual impairment. * A multi-informant approach is crucial in serious game-based therapy development. * Visuoperceptual profiles serve as a basis for individualized entry-level difficulty. * Automatic in-game adaptivity can be preliminarily defined using handcrafted rules. * Iterative approach in different user testing groups supported fine-tuning of games. Abstract Children with cerebral visual impairment (CVI) exhibit a very heterogeneous clinical visuoperceptual picture, which implies that a targeted individualized and adaptive therapy is necessary. Serious games for CVI are limited, especially those that train multiple visuoperceptual skills in an individualized and adaptive manner without the frequent manual input from clinicians. In this paper, we describe the design, development, and evaluation process of novel individualized and adaptive serious mini-games for visual perceptual skills. First, an informant-led design, including focus groups with relevant stakeholders implementing a participation-via-proxy and brainstorming sessions, was performed. This resulted in rapid prototypes of four mini-games (MatchMaker, Hurricane Chaos, Maze Explorer 2D, and Maze Explorer 3D), targeting six visuoperceptual skills commonly impaired in children with CVI: (1) visual discrimination and matching, (2) object or picture recognition, (3) visual spatial perception, (4) figure-ground perception, (5) motion perception, and (6) visual short-term memory. Therapeutic game content (selecting, manipulating, and rating images; defining entry-level difficulty and in-game adaptivity rules) was evaluated by clinical and research experts. Finally, formative testing and expert feedback, including usability and user experience by clinicians, researchers, and typically developing children, led to important modifications in the mini-games. Author Affiliation: (a) Department of Development and Regeneration, University of Leuven (KU Leuven), O&N IV Herestraat 49, Box 805, 3000 Leuven, Belgium (b) Department of Electronics and Informatics (ETRO), Vrije Universiteit Brussel (VUB), Brussels, Belgium (c) imec, Leuven, Belgium (d) Department of Brain & Cognition, University of Leuven (KU Leuven), Leuven, Belgium (e) Leuven Brain Institute (LBI), Leuven, Belgium (f) Child Youth Institute (L-C&Y), Leuven, Belgium * Corresponding author. Article History: Received 3 August 2021; Revised 16 November 2021; Accepted 26 November 2021 (footnote)[white star] The views expressed in the submitted article are our own and not an official position of the institution or funder. This work was supported by the Fund Scientific Research Flanders (FWO-project) (grant number T003817N). Author JW was supported by the Flemish Government (grant number METH/14/02). Byline: N. Ben Itzhak [nofar.benitzhak@kuleuven.be] (a,*), I. Franki (a), B. Jansen (b,c), K. Kostkova (b,c), J. Wagemans (d,e), E. Ortibus (a,f)