Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Modality-dependent and independent effects of sentence context effects on visual word recognition
Poster Session B, Wednesday, September 30, 4:30 - 6:30 pm, Wangari Maathai
This poster is part of the Sandbox Series.
Sara Guediche1, Harshi Lodha Jain; 1Augusta University
Visual word recognition is facilitated by a predictable sentence context, especially when the word is difficult to read. For example, reduced interletter spacing within a word (crowding) slows visual word recognition. Visual word recognition and reading comprehension, more generally, can also be impaired by many other factors including developmental reading disorders. In such cases, a predictable sentence context may be more beneficial to visual word recognition if presented acoustically. Clark et al. (2021) demonstrated that predictable auditory sentence contexts also facilitate visual word recognition, with larger benefits observed for crowded words. The present study investigates the neural mechanisms underlying sentence-context effects on visual word recognition as a function of sentence modality (auditory vs. visual) using functional magnetic resonance imaging (fMRI). To this end, we developed predictable and unpredictable sentence contexts for a common set of target words. In an online norming study, 20 participants completed a sentence-completion task to assess cloze probability for the predictable sentences. An additional 20 participants completed a sentence-rating task to ensure comparable numbers of highly rated predictable and unpredictable sentences. We are currently running participants in an fMRI study that manipulates sentence modality (auditory, visual), sentence predictability (predictable, unpredictable), and visual crowding (crowded text, standard text) using the same target words, counterbalanced across participants. We predict that activity in the ventral lateral occipital cortex, specifically in the visual word form area (VWFA; Dehaene et al., 2002), will increase under crowded viewing conditions, reflecting greater difficulty in visual word recognition. We further predict that predictable sentence contexts will reduce activity in this region through top-down modulation, regardless of sentence modality. Finally, we predict an interaction between crowding and sentence predictability, such that contextual facilitation will be greater for crowded words in both auditory and visual sentence conditions. In addition, we will examine modality-dependent interaction effects on functional connectivity between regions across temporal, parietal, and frontal cortices. Based on neuroanatomical models of reading that suggest greater dorsal-stream engagement during reading relative to listening (e.g., Carreiras et al., 2014), we predict stronger interaction effects involving the angular gyrus for visually presented sentence contexts. In contrast, based on models of spoken-language comprehension (e.g., Hickok & Poeppel, 2007), we predict greater recruitment of ventral temporal-lobe regions associated with lexical-semantic processing during auditory sentence contexts, including middle and anterior temporal regions. These findings will provide insight into the neural mechanisms underlying contextual facilitation across modalities and may inform neural models that account for both modalities during word recognition and language comprehension.
Topic Areas: Reading, Multisensory or Sensorimotor Integration