Presentation
Search Abstracts | Symposia | Slide Sessions | Poster Sessions | Poster Slams
Meet my associates: Neural correlates of free word association measures
Poster D53 in Poster Session D with Social Hour, Friday, October 7, 5:30 - 7:15 pm EDT, Millennium Hall
Nicholas Riccardi1, Rutvik Desai1; 1University of South Carolina
Introduction: Free word association provides a measure of the likelihood that one word can prompt another word to come to mind with minimal constraints. Therefore, large-scale associative networks of words can capture shared lexical-semantic and socio-cultural experiences of many people. How these associative structures influence the neural representation of words remains understudied. It stands to reason that associative properties (e.g., how many associates a word has or how strongly a word is associated to other words) could be represented in putative lexical-semantic ‘hub’ areas such as the anterior temporal lobe (ATL), angular gyrus (AG), posterior cingulate cortex (PCC), and medial temporal lobe (MTL). Here, in an fMRI study, we investigated the relationship between brain activity and free word association measures provided by University of South Florida Free Association Norms (USF-FAN). Methods: Twenty-one healthy volunteers participated (10 female, mean age 25). In the scanner, participants completed a lexical decision task while being presented with concrete and abstract words (matched in frequency and length) and pronounceable pseudowords. Participants indicated via button press whether the word was real or not. For each word, three associative measures were gathered from USF-FAN and used as parametric regressors of interest: que set size (QSS), que probability resonance (QPR), and que resonance strength (QRSG). QSS is a measure of how many strong associates a word has (e.g., farm may have many strong associates such as cow, chicken, barn, etc.). QPR is the probability that each associate in a cue word’s association set will produce the original cue word as an associate (i.e., if presented with cow, chicken, or barn, what is the probability that someone will respond with farm?). QRSG is a measure of how large and strongly connected a given word’s associative network is. A region-of-interest (ROI) analysis was conducted to examine the relationship between brain activity and these associative measures within four bilateral ROIs: ATL, AG, PCC, and MTL, with each of these ROIs being broken into smaller subregions defined by the Human Connectome Project atlas. Within each subregion, two-tailed t-tests were conducted against zero, and Bonferroni corrected for multiple comparisons (p<.05). QSS results: Set size was significantly negatively associated with activity in left MTL and ATL, including entorhinal and perirhinal cortices and the superior temporal gyrus pole (STGp; all p<.05), especially for abstract words. No other areas were significantly associated with QSS. QPR: Probability resonance was significantly positively associated with right hemisphere AG subregions (anterior inferior parietal), as well as bilateral ATL (left STGp, right superior temporal sulcus). No other areas were significantly associated. QRSG: Resonance strength was significantly positively associated with activity in bilateral PCC subregions, as well as left entorhinal cortex and left temporoparietal junction. No other areas were significantly associated. Conclusions: As set size decreases, activity in left MTL and ATL increases, suggesting that MTL and ATL may help represent words without many strong associates. High QPR and QRSG were associated with increased bilateral AG, PCC, and ATL activity, suggesting that words with strong resonant associations are represented bilaterally.
Topic Areas: Meaning: Lexical Semantics, Reading