Constructing meaning: Task demands and the time course of lexical and semantic processing in visual word recognition
Date
2014-01-14
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
The ultimate goal of word recognition is accessing meaning, indeed, the desire to characterize how and when words contact meaning has driven the development of numerous computational models (e.g., Coltheart et al., 2001; McRae, de Sa & Seidenberg, 1997; Morton, 1969). While progress has been made, there is still little agreement between models when it comes to semantic representation. Many studies suggest that semantic access is not a unitary phenomenon, but is better characterized as a diverse set of processes that construct meaning in order to suit the current context (Yap et al., 2012). Across two studies we investigated the dynamic nature of semantic processing. First, we investigated the relative availability of language- and object-based semantic information by tracking the time course of semantic richness effects in word recognition using a signal-to-respond (STR) paradigm fitted to a lexical decision (LDT) and a semantic categorization (SCT) task. We used linear mixed effects to examine the relative contributions of language-based (number of senses, ARC) and object-based (imageability, number of features, body-object interaction ratings) descriptions of semantic richness at four STR durations (75, 100, 200, and 400 milliseconds). Results showed an early influence of number of senses and ARC in the SCT. In both LDT and SCT, object-based effects were the last to influence participants’ decision latencies. Second, we expanded upon our investigation of task-effects on semantic activation by contrasting the event-related potentials associated with two semantic categorization tasks (SCT) using electroencephalography. The two SCTs involved different decision categories: is it an animal? vs. is it a concrete thing? Participants completed both tasks and, across participants, the same core set of items was presented. Despite null differences in reaction time, under the concreteness decision we observed greater negativities in an N400 window (300-500) than when the same items were presented under the animal decision. This task-driven N400 concreteness effect suggests that top down modulation of semantic processing can be elicited by minor shifts in decision criteria. We interpret our results within a framework in which conceptual representations are not simply accessed from memory, but are dynamically constructed (Kiefer, 2012; Kutas & Federmeier, 2011).
Description
Keywords
Psychology, Psychology--Cognitive, Psychology--Experimental
Citation
Hargreaves, I. (2014). Constructing meaning: Task demands and the time course of lexical and semantic processing in visual word recognition (Doctoral thesis, University of Calgary, Calgary, Canada). Retrieved from https://prism.ucalgary.ca. doi:10.11575/PRISM/25428