Slide Slam D1
Neural Correlates of Picture Naming versus Definitional Naming
Juliana Baldo1, Kelsey Spalding-Wilson2, Krista Schendel1, Timothy Herron1, Nina Dronkers3, Brian Yochim2; 1VA Northern California Health Care System, 2VA St. Louis Health Care System, 3University of California, Berkeley
Anomia, a disruption in naming, is one of the most common language symptoms following brain injury. Naming is often assessed with picture naming tests such as the Boston Naming Test (BNT). More recently, alternative naming tasks have been developed that require patients to name an item in response to a brief definition (e.g., What is the name of the part of your shirt that goes around your neck?). Such definitional naming tasks have the advantage of avoiding potential confounds in picture naming, such as poor performance due to visual processing deficits. It is unclear, however, to what extent picture naming and definitional naming tap the same cognitive and neural mechanisms. In the current study, we evaluated a group of chronic left hemisphere stroke patients on a newly developed definitional naming test, the Verbal Naming Test (VNT; Yochim et al., 2015, 2019). Our goals were: 1) to characterize stroke patient performance on this new naming task, 2) to identify its neural underpinnings, and 3) to compare the VNT to traditional tasks of visual naming. Stroke patients with a single left hemisphere stroke (n = 19) were administered the VNT, the Boston Naming Test-15 (BNT-15), and the Western Aphasia Battery (WAB) Object Naming subtest in different sessions. To determine the anatomic correlates of naming, we related patients’ naming scores to ROIs using quasi-logistic regression analyses with lesion volume, months post-stroke, and aphasia severity as covariates. Due to the small number of patients, the analysis was focused on ROIs in the left temporal lobe that have been previously implicated in naming: the superior temporal gyrus, middle temporal gyrus, and inferior temporal gyrus. These regions were further subdivided into an anterior and posterior half for each gyrus. Mean performance on the VNT without cueing was 29.4 items out of 50 (SD = 18.2, range: 0-50), and with phonemic cueing was 34.7 items out of 50 (SD = 19.3, range: 0-50). VNT performance was highly correlated with the BNT-15, r(16) = .94 and with WAB Object Naming, r(17) = .87. We found that both definitional and picture/object naming tasks were most strongly associated with the left middle temporal gyrus. We also found that all tasks were more strongly associated with posterior temporal cortex than anterior temporal cortex. These findings are consistent with our previous findings in a larger sample of stroke patients showing that left mid-posterior middle temporal cortex and underlying white matter was most critical to picture naming. However, other findings have suggested that left anterior, and sometimes ventral, temporal cortex, are most strongly associated with naming, particularly in degenerative cases. This association with anterior temporal cortex has also been noted for definitional naming tasks in previous studies. Additional, larger-scale studies with more comprehensive lesion coverage are needed to further investigate these naming dissociations in both stroke-related and degenerative disease.