Abstract/Summary

The chapter highlights key insights gained by using the speed-accuracy tradeoff (SAT) technique. SAT evidence has been instrumental in integrating sophisticated memory models into psycholinguistic theory, notably introducing content-addressable retrieval cues stemming from syntactic, semantic, morphological, and pragmatic levels. We describe the SAT procedure, and discuss two key features: retrieval cues support direct-access memory retrieval, yet can also create similarity-based retrieval interference. We explain how SAT modeling disentangles accuracy from time course of processing. Time course measures provide a crucial way to investigate architectural issues, such as number and type of operations involved and organization of component operations (e.g., serial search, cascaded parsing, parallel processing, active maintenance in focal attention). Our review describes how SAT evidence bears on linguistic issues in experimental syntax, with examples involving long-distance dependencies, garden-path sentence ambiguities, and anaphora resolution. Additional psycholinguistic areas where SAT has been informative are briefly noted. Challenges and promising future directions are discussed.