Chapter 7 explores an empirical challenge for both representational- and retrieval-based accounts of attraction, focusing on object pronouns and their resistance to attraction effects. While attraction has been observed across various linguistic dependencies, such as subject–verb agreement and reflexives, attempts to induce attraction with object pronouns have consistently failed. This chapter reviews past studies and introduces new high-powered self-paced reading experiments designed to test attraction for object pronouns. The findings show, for the first time, that object pronouns are indeed susceptible to attraction effects, specifically when attractor nouns match the pronoun in gender. The experiments also reveal a grammatical asymmetry, where attraction occurs only in ungrammatical sentences, aligning with the predictions of retrieval-based accounts. These results challenge representational accounts, which predict attraction in both grammatical and ungrammatical configurations. This chapter provides new insights into how gender cues are processed during pronoun resolution and offers crucial evidence favoring the retrieval-based account of attraction.

Chapter 3 focuses on agreement attraction, one of the most well-studied phenomena in psycholinguistics. Linguistic dependencies, particularly subject–verb number agreement, are disrupted by attractors – intervening elements that have the correct information in the wrong position. Attractors lead to the formation of illicit grammatical dependencies, creating the illusion that ungrammatical sentences are acceptable or that well-formed sentences are not. Focusing primarily on subject–verb number agreement, the chapter introduces readers to experimental paradigms used to study attraction effects in sentence production and comprehension. It discusses key factors that modulate attraction, including number morphology, sentence complexity, and the syntactic properties of attractors. A major theme is how attraction-based interference reveals underlying principles of memory encoding and retrieval and real-time language processing. The chapter also introduces methodological tools, such as factorial designs, and experimental techniques like self-paced reading and eye-tracking, which have been critical in uncovering how agreement attraction operates in moment-to-moment language comprehension.

Chapter 2 provides a comprehensive survey of various linguistic illusions, showing how errors in language processing arise. It begins with auditory illusions, such as the Yanny/Laurel effect and McGurk effect, highlighting how multisensory inputs influence speech perception. It covers sentence processing illusions, including lingering misinterpretations, role reversals, and local coherence effects, which reveal systematic misinterpretations of syntactically complex or semantically ambiguous sentences. The comparative illusion, missing VP effect, and illusory NPI licensing are explored, demonstrating how these errors reveal the workings of the fundamental cognitive mechanisms that support language processing, such as memory retrieval and the interaction between domain-specific and domain-general processes. A central theme is the interaction between shallow processing strategies and deeper cognitive mechanisms, which sometimes lead to illusory interpretations. This exploration of linguistic illusions underscores their value as tools for uncovering how the mind processes language in real time, contributing to broader theories about sentence comprehension and cognitive architecture.

Chapter 8 provides a summary of the book’s key findings, emphasizing how the retrieval-based account provides better empirical coverage over the representational-based accounts. This chapter then explores key outstanding questions in the study of linguistic illusions, including the interaction between encoding and retrieval processes, individual differences, the effects of good-enough processing, and the role of different linguistic features across languages. The chapter concludes by outlining future directions for research, suggesting potential interventions to reduce attraction errors through memory training and timing manipulations. As the final chapter, it reflects on how scientific inquiry continues to evolve, encouraging further investigation into the cognitive mechanisms behind real-time language processing.