Recent evidence suggests attentional templates can be flexibly adapted based on a range of context expectations to reflect the most relevant object features in the current environment, e.g., features that are unique to the target, the target's expected retinal size or features of associated objects