Search Abstracts | Symposia | Slide Sessions | Poster Sessions
From wordforms to meaning: understanding the neurobiology of word learning using experimental pain.
Poster Session C, Friday, October 25, 4:30 - 6:00 pm, Great Hall 3 and 4
Natalia Egorova-Brumley1, Marie Spehar1, Bradford Moffat1; 1The University of Melbourne
Abstract words, denoting ideas, qualities, or states such as courage, education, or pain, have been notoriously difficult to study due to the subjectivity and variability of their meanings across individuals. Yet, understanding how these words are represented in the brain once just learned would shed light on the extent to which their neural representations are disembodied in heteromodal language cortices or embodied in the physical experiences that comprise their meanings. Twenty-seven right-handed participants (8 male, mean age = 25.4 ± 6.7 years) studied 6 pseudowords and underwent behavioural testing and 7 Tesla magnetic resonance imaging the following day. The novel word learning paradigm in this study, which utilised experimental pain methods, enabled the creation of pseudowords with controlled meanings denoting painful and non-painful sensations. It involved the presentation of written pseudowords with heat stimuli of varying intensities applied to the arm and leg. Presenting only words in the scanner (previously learned and completely novel), we were able to delineate brain correlates of (a) wordform learning, by comparing novel pseudowords vs. learned pseudowords; (b) the effect of meaning, by contrasting pseudowords learned with and without meaning; and (c) the representation of specific semantic features (e.g., painful vs. nonpainful sensations to the arm vs. to the leg). Accuracy and reaction times on the semantic judgement task, as well as free recall task accuracy, confirmed that all the pseudowords were learned, further revealing significantly better learning for words with meaning vs. no meaning and better retention of words denoting pain, with no difference between leg and arm sensation words. All pseudowords demonstrated activation in the bilateral fusiform gyrus. Learned wordforms elicited increased activation in the left posterior superior temporal sulcus and the angular gyrus. Pseudowords paired with sensory/painful stimulation showed increases in sensory-motor and insula activation, while the leg and arm relatedness was encoded in largely overlapping regions in the superior frontal and parietal cortex. In addition to demonstrating that word meanings can be acquired from sensory contexts, these results provide preliminary evidence of rapidly emerging cortical activation consistent with abstract word meanings being both disembodied and embodied in sensory experiences.
Topic Areas: Language Development/Acquisition, Meaning: Lexical Semantics