Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Neural bases of differences between older and younger adults in the learning of novel word forms and meanings
Poster Session E, Sunday, September 14, 11:00 am - 12:30 pm, Field House
This poster is part of the Sandbox Series.
Sartaj Singh1, Yuan Tao2, Brenda Rapp2, Robert Wiley1; 1University of North Carolina Greensboro, 2Johns Hopkins University
Introduction While research shows a wide range of cognitive changes associated with aging, there is limited research on the impact of aging on novel word learning. Word learning involves learning multiple components–spelling, pronunciation, and meaning. Previously, in Rapp et al. (2024), we reported that older adults exhibited significant difficulties in learning the spoken and written forms of novel words, while, in contrast, word meanings were learned comparably to younger adults. Here, we report the behavioral results from the complete set of participants and, building on these robust behavioral effects, we focus on understanding the neural bases of the strengths and weaknesses exhibited by older adults. Specifically, we seek to answer the following: Question 1: Do cortical (language network) and subcortical (hippocampal) mechanisms contribute similarly to learning word forms and meanings in older and younger adults? Question 2: Do older and younger adults differ in the extent of neural differentiation/de-differentiation as indexed by BOLD peak magnitude and spread (Gordon et al., 2018)? If so, do these account for the observed learning differences? Methods Behavioral: Twenty-three older adults (OAs; mean age = 67.4 years, SD = 5.4) and twenty younger adults (YAs; mean age = 24.0 years, SD = 10.6), all native English speakers. Participants attempted learning the spellings, pronunciations, and meanings of 36 novel words within a fantasy story, over several weeks. Results: For word spellings, YAs exhibited significantly higher accuracy than OAs (p = 0.024), with no significant differences in semantic learning between groups (p = 0.14). Analysis of phonological word-form learning is underway. Neuroimaging: fMRI scanning administered prior to learning allows for identification of predictors of learning effects; data collection is complete. Language Localizer: Language network responsivity indexed by the contrast of STORY>BACKWARD SPEECH. Learning Localizer: Learning responsivity indexed by the contrast LEARN>VISUAL BASELINE, in bilateral hippocampi. Planned Analyses: Analysis 1: Identify brain areas where responsiveness predicts subsequent word learning Step 1: Compute language and learning responsiveness measures per voxel, per participant. Step 2: For specific ROIs, use LME modelling to evaluate the relationship between the out-of-scanner word learning scores (one model for each of the three learning dimensions) and both language and learning neural responsiveness, across OA and YA groups. The ROIs will include: bilateral hippocampi, key nodes of the language network (ATL, IFG, STG, MTG). Analysis 2: Evaluate if neural differentiation/de-differentiation predicts later word learning Step 1: For each ROI and participant, extract locations corresponding to activation peaks for STORY>BACKWARD SPEECH in the language network ROIs, and the LEARN>VISUAL BASELINE for the bilateral hippocampi. Step 2: For each peak, extract unsmoothed percent-signal-change within spherical kernels, starting from a radius of 3mm and up to 10mm. Compute the incremental signal per unit volume added, then fit a 1/radius² decay function across radii. Step 3: For each ROI, use LME modelling to evaluate the relationship between out-of-scanner learning (one model per learning dimension), and both BOLD peak magnitudes and spreads, across OA and YA groups.
Topic Areas: Writing and Spelling,