Mapping the effect of psycholinguistic variables on picture naming: insights from an FDG-PET study on neurodegenerative diseases

Picture naming involves mapping a visual input to a semantic representation and linking it to a word form for spoken output (Catricalà et al., 2020). Naming performance is affected by the properties of the words/pictures, but previous studies have failed to univocally identify the naming stage(s) and the brain regions affected by each variable. We explored the impact of the main variables on behavioural performance and brain metabolism in neurodegenerative patients.

A total of 178 neurodegenerative patients were administered the CaGi picture naming test (Catricalà et al., 2013) and underwent a FDG-PET scan. We extracted the components underlying the variables’ structure with principal component analysis. We calculated the impact of each variable/component on the patients’ performance, adopting an item-level procedure (de Marco et al., 2023) and correlating the obtained measures with the brain metabolism extracted from 11 regions of interest.

Four components emerged, namely word form, visual, lexical, and semantic. Stimuli with a simpler word form (short and with many phonological neighbours) and lexical structure (acquired early in life, frequent, and familiar) were named correctly more often, while more difficult ones induced semantic errors.

Gradual specialization emerged across regions, with the anterior temporal lobe contributing to multiple naming stages; the posterior fusiform gyrus supporting visual, semantic, and lexical properties; and the middle posterior temporal and inferior frontal gyri representing an interface between lexico-semantics and phonology.

These results go beyond one-to-one correspondence between a variable and a unique naming stage/brain area, suggesting that the same brain area can support relatively different naming stages depending on the network into which it is embedded.