Abstract

Producing a word requires selecting among a set of similar alternatives. When many semantically related items become activated, the difficulty of the selection process is increased. Experiment 1 tested naming of items with either multiple synonymous labels (“Alternate Names,” e.g., gift/present) or closely semantically related but non-equivalent responses (“Near Semantic Neighbors,” e.g., jam/jelly). Picture naming was fastest and most accurate for pictures with only one label (“High Name Agreement”), slower and less accurate in the Alternate Names condition, and slowest and least accurate in the Near Semantic Neighbors condition. These results suggest that selection mechanisms in picture naming operate at two distinct levels of processing: selecting between similar but non-equivalent names requires two selection processes (semantic and lexical), whereas selecting among equivalent names only requires one selection at the lexical level. Experiment 2 examined how these selection mechanisms are affected by normal aging and found that older adults had significantly more difficulty in the Near Semantic Neighbors condition, but not in the Alternate Names condition. This suggests that aging affects semantic processing and selection more strongly than it affects lexical selection. Experiment 3 examined the role of the left inferior frontal gyrus (LIFG) in these selection processes by testing individuals with aphasia secondary to stroke lesions that either affected the LIFG or spared it. Surprisingly, there was no interaction between condition and lesion group: the presence of LIFG damage was not associated with substantively worse naming performance for pictures with multiple acceptable labels. These results are not consistent with a simple view of LIFG as the locus of lexical selection and suggest a more nuanced view of the neural basis of lexical and semantic selection.

Highlights

  • Producing a word requires selecting a single candidate word from a set of similar and related alternatives (e.g., Levelt et al, 1999; Howard et al, 2006; Dell et al, 2013; Walker and Hickok, 2015)

  • Naming latency was faster in the Alternate Names condition compared to the Near Semantic Neighbors condition

  • The same overall pattern of results was observed (Table 3, middle column), with latency significantly increasing from the High Name Agreement to the Alternate Names to the Near Semantic Neighbors condition

Read more

Summary

Introduction

Producing a word requires selecting a single candidate word from a set of similar and related alternatives (e.g., Levelt et al, 1999; Howard et al, 2006; Dell et al, 2013; Walker and Hickok, 2015) These alternatives may include words that describe the concept at different levels in a taxonomic hierarchy (e.g., beagle–dog–mammal–animal), synonyms (e.g., sofa–couch), and words that refer to similar objects or concepts Numerous studies have shown that activation of semantically related alternatives makes selection more difficult as reflected by slower response times and lower accuracy (e.g., Schnur et al, 2006; Bormann, 2011; Mirman, 2011; Roelofs et al, 2012; Shao et al, 2015) This selection process appears to engage ventro-lateral prefrontal cortex (VLPFC, sometimes described as inferior frontal gyrus, IFG), typically left-lateralized for word-based tasks. We used an “underdetermined object naming” task – naming of pictures that have multiple correct or acceptable names

Objectives
Methods
Results
Discussion
Conclusion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call