The processing of ambiguous pronominal reference is sensitive to depth of processing

Previous studies on the processing of ambiguous pronominal reference have led to contradictory results: some suggested that ambiguity may hinder (Stewart, Holler, & Kidd, 2007), while others showed an advantage (Grant, Sloggett, Dillon, 2020) similar what has been reported for structural ambiguities. This study provides a conceptual replication Stewart et al. (2007, Experiment 1), examine whether discrepancy in earlier results is caused by depth participants engage (cf. Swets, Desmet, Clifton, Ferreira, 2008). We present from word-by-word self-paced reading experiment with Dutch sentences contained personal pronoun embedded clause was either or disambiguated through gender features. Depth manipulated offline comprehension questions. The difference times versus unambiguous depends depth: significant penalty found under deep but not shallow processing. No found, regardless depth. replicates (2007) using different methodology and larger sample size appropriate statistical power. These findings provide further evidence resolution flexible process, such way which are processed relevant information. Theoretical methodological implications these discussed.