Lachmann and van Leeuwen’s research over the last two decades has shown that visual context affects letter identification differently than visually similar pseudo-letters and shapes (non-letters). By comparing isolated targets and targets surrounded by congruent and incongruent shapes, they found that shapes and pseudo-letters typically benefit from congruent surrounds relative to the isolated baseline, whereas letters show a negative congruence effect, i.e., accuracy and speed are reduced in the congruent condition while incongruent surrounds produce positive or neutral congruence effects. This finding points to categorical differences in processing: letters rely on highly practiced, automatized, analytic representations that conflict with the grouping tendencies induced by congruent surrounding shapes, whereas non-letters are more susceptible to holistic integration with the surrounding context. In the present study, we test whether these effects can be modulated when the visual context is altered through additional flankers. In a speeded letter-versus-non-letter identification task, we manipulate surround congruency (congruent vs. incongruent) and flanker condition (none, two flankers, four flankers). We hypothesize that flankers may form a broader contextual structure, making the central letter more isolated and thereby reducing or eliminating the influence of the surrounding shape. We report preliminary findings on how immediate and expanded context shape letter perception.