You are viewing the SNL 2017 Archive Website. For the latest information, see the Current Website.

 
Poster B57, Wednesday, November 8, 3:00 – 4:15 pm, Harborview and Loch Raven Ballrooms

Processing Demands of Word Frequency on Verbal working Memory as measured by functional near-infrared spectroscopy (fNIRS)

Amy Berglund1, Julia L. Evans1, Andrea W. Fung1, Chen Song1, Fenghua Tian2, Holly Watkins1;1University of Texas at Dallas, 2University of Texas at Arlington

A learner’s ability to hold words in working memory impacts their ability to learn language. This working memory dependency is due to its characteristic of constantly updating information with temporally incoming input (Baddeley, 2003; Szmalec, Brysbaert, & Duyck, 2012). Behavioral data suggests that low-frequency words (those words that occur rarely in language) place greater processing demands on working memory than high-frequency words (those words that occur frequently in language) (Polich & Donchin, 1988). The question of how neural activity varies with changes in cognitive processing demands has been the focus of considerable research partly because understanding these dynamics may provide insights into how mental activity is organized in the cortex. Functional magnetic resonance imaging (fMRI) studies show greater cortical activity, as measured by increased task-related cerebral blood flow, in left-prefrontal blood-oxygen-level-dependent (BOLD) signal changes when processing low-frequency words as compared to high-frequency words, suggesting low-frequency words have greater processing demands than high-frequency words (Chee, Hon, Caplan, Lee, Goh, 2002; Diana & Rder, 2007). Functional near-infrared spectroscopy (fNIRS) is a non-invasive, low-cost neuroimaging modality that measures the relative changes of oxygenated hemoglobin (HbO2) and deoxygenated hemoglobin (Hb) based on near-infrared light absorption (650-900nm). This technology allows researchers to extract functional brain activity information with high temporal resolution and low sensitivity to movement artifacts (Ferrari & Quaresima, 2012). fNIRS has been shown to effectively measure cerebral hemodynamic changes resulting from cognitive processing demands during n-back working memory tasks. Thus proving an effective means of quantifying mental work load and resulting cognitive fatigue with changes in lexical processing demands (Herff et all, 2014). As reviewed by Owen et al. (2005), the n-back tasks primarily activate the frontal poles, dorsolateral, and ventrolateral prefrontal cortex (Owen et. al., 2005). In this study we used fNIRS to record the prefrontal cortex activity in eleven neurologically typical adults engaging in an auditory 2-back task. Word frequency was manipulated by alternating 7 blocks containing high-frequency words with 7 blocks containing low-frequency words. By using the auditory 2-back task to control task demands while simultaneously manipulating the lexical processing demands of word frequency, we were able to examine both the cognitive processing demands and resulting fatigue effects of word frequency on working memory in typical adults. Raw fNIRS data was collected using a CW-6 system (TechEn, Inc.) and processed using HOMER software (Huppert et al., 2009b). Behavioral measures were collected using E-prime. The preliminary results indicated large cortical responses from both the left and right dorsolateral prefrontal cortices. Additionally, low-frequency words evoked higher cortical activation as compared to high-frequency words. Participants showed patterns of fatigue both within individual blocks and over the course of the entire session. Despite the general fatigue trend, analysis revealed no influence of fatigue on individuals’ overall accuracy. In summary, the results from this study support the theory that cognitive load requirements differ to process low-frequency words versus high-frequency words. Interestingly, despite the differing requirements of cognitive load and increasing fatigue, accuracy appears to not be influenced by cognitive difficulty or fatigue in typical adults.

Topic Area: Phonology and Phonological Working Memory

Back to Poster Schedule