Compositional parsing in adjective-noun phrases: The role of adjective semantics

A central question in language comprehension is how the mind combines the meanings of individual words. Minimal phrases such as adjective-noun pairs (red cup, big boat) provide a tractable model for studying how linguistic and perceptual features are integrated during comprehension. Previous research has often assumed a uniform compositional process, independent of adjective meaning. Across three experiments, we examined how adjective type (intersective vs. subsective) and word order influence visual verification performance. Experiment 1 used intersective colour adjectives and replicated the established facilitation for double-feature cues, showing faster responses to adjective-noun phrases than to single-word cues. Experiment 2, using subsective size adjectives, produced a similar but weaker pattern, with the advantage no longer significant once individual variability was modelled. Experiment 3 directly contrasted both adjective types within participants and revealed slower overall responses for subsective adjectives but no reliable interaction between adjective type and feature number. Taken together, these results indicate that the apparent "compositional advantage" in this paradigm depends on the semantic transparency of the features involved rather than on a categorical compositional mechanism. Intersective adjectives provide directly alignable perceptual information, yielding faster visual-linguistic integration, whereas subsective adjectives rely on context-dependent interpretation and show less consistent facilitation. Word order did not systematically affect comprehension in any experiment.