This thesis is concerned with the mental representation and online processing of conceptual structure. In particular, it investigates two issues: First, it explores some of the processing mechanisms underlying the comprehension of conceptual combinations. Second, it explores some of the processing mechanisms underlying the production of coerced expressions. These two issues are investigated experimentally using a priming paradigm. Noun-noun combinations like dog scarf are common in everyday discourse but often have more than one interpretation. How do language users arrive at an interpretation of the relationship between the two nouns? The first half of the thesis reports four expression-picture matching experiments that used priming to investigate the influence of modifier and head constituents on the comprehension of novel ambiguous noun-noun combinations. Experiment 1 examined the effects of lexical repetition and semantic relation. Results showed reliable relation priming, regardless of whether the modifier or head was repeated between prime and target: Participants tended to choose target pictures involving the same relation as a preceding prime picture. Experiment 2 demonstrated significant relation priming when neither constituent was repeated. Experiment 3 showed significant relation priming when each picture contained both possible semantic relations, arguing against a possible visual-priming account of the effect. Experiment 4 showed that relation priming did not have an effect on the time taken to comprehend a combination. The findings are interpreted in light of competing models of conceptual combination. The second half of the thesis reports four experiments designed to investigate the effects of priming on the production of complement coercions like The author began the book. Recent work in lexical semantics has demonstrated that verbs such as begin and enjoy semantically select for event complements. Where such verbs occur with entity-denoting nouns (e.g., begin the book, enjoy the wine), the NP complement undergoes semantic type coercion, inducing a reference shift to the event associated with that NP. Using a combined picture-description/sentence completion task, participants were presented with pictures followed by sentence fragments which they were instructed to complete. Experiment 5 showed a reliable effect of Prime: Participants tended to produce a target description involving the same level of semantic specification as the preceding prime. Experiment 6 did not show fully significant priming in the absence of (coercing) verb overlap between prime and target. Experiment 7 revealed evidence of semantic and syntactic components to the priming effect. Experiment 8 showed no evidence for differing global and local contextual influences on priming. Taken together, the results of Experiments 5–8 are interpreted in terms of a model of language production based on Levelt, Roelofs, and Meyer (1999). Overall, this study offers insight into the representation and processing of conceptual structure in comprehension and production from a psycholinguistic perspective.