This thesis aimed to describe the coevolution of orthographic information encoding mechanisms during learning to read. Building on the work of Grainger et al. (2016), the first chapter performs a theoretical characterization of three orthographic mechanisms, namely: Position-Specific Encoding (PSE), Position-Invariant Encoding (PIE), and Spatial Integration (SI). This characterization was used to specify two research questions. The first was to determine how the implementation of one specific mechanism influenced or was influenced by the implementation of one or more of the other mechanisms during reading instruction. The second question focused on the efficacy relationships, i.e., the functional interdependence, between these mechanisms. These questions were examined empirically in the second chapter through two original studies that measured the impact of each of these mechanisms in children from the 2nd to the 6th year of learning to read. The third chapter proposes a theoretical reflection on how these mechanisms encode reading units, such as letters, words, and sentences. This reflection led to the application of previous research findings to the processes involved in encoding these units, particularly regarding the effect of visual word identification abilities on the ability to process multiple words in parallel to construct a sentence representation. These issues were examined empirically in the fourth chapter through two studies. The first study, conducted with adults, aimed to describe the efficiency relationships between the encoding of letter, word, and sentence units. The second study, conducted with children in 2nd, 4th, and 6th grades, aimed to specify the developmental trajectory of the ability to construct a syntactic representation of the sentence through parallel word processing and to determine the extent to which this ability was related to the ability to identify printed words. The results of these four studies were critically reviewed in the discussion section. These results confirmed that functional interdependence relationships linked the EPS, EPI and IS mechanisms and were likely to constrain the implementation of the EPS, EPI and IS mechanisms during reading acquisition. In particular, a degree of efficiency of the PPE mechanism, close to that expected in the 5th year of learning, is necessary to implement the IS mechanism under natural reading conditions. Finally, the critical analysis of these results highlighted three points of concern for future studies exploring the coevolution hypothesis of orthographic mechanisms. First, the nature of the processes involved in the tasks and their overlap must be precisely identified. Second, rigorous methodological work must be initiated to develop experimental procedures adapted to beginning readers. Third, the first year of learning, and the quantification of exposure to writing during this period, is an essential time window for fully understanding the coevolution hypothesis of orthographic mechanisms. In this respect, this thesis will have provided the first elements of theoretical and empirical characterization of this original hypothesis and will have outlined its future exploration.