Dynamics of second-language learners' semantic memory networks: Evidence from a snowball sampling paradigm

Area of Science:

• Cognitive Psychology
• Neuroscience of Language
• Network Science

Background:

• Semantic memory networks represent word relationships in the brain.
• Understanding how these networks change during second language (L2) acquisition is crucial for L2 pedagogy and cognitive science.
• Previous research has not quantitatively analyzed the structural evolution of L2 semantic networks over time.

Purpose of the Study:

• To analyze the structural changes in second-language (L2)-based semantic memory networks during L2 learning.
• To quantitatively assess how the duration of L2 learning influences the topology of L2 semantic networks.

Main Methods:

• Employed a snowball sampling paradigm to construct individual L2 semantic networks.
• Compared structural characteristics of these networks cross-sectionally between participants with differing L2 learning durations.
• Analyzed network metrics including average node degree, shortest path length, modularity, centralization, density, clustering coefficient, and small-world indices.

Main Results:

• Increased average node degree, indicating greater network connectivity.
• Decreased average shortest path length, suggesting enhanced network efficiency.
• Reduced modularity and centralization, implying less fragmentation and decentralization.
• Decreased network density.
• No significant change in average clustering coefficients or small-world indices.

Conclusions:

• L2 learning progressively restructures semantic memory networks, making them more interconnected and efficient.
• The observed quantitative changes provide evidence for the dynamic nature of semantic memory during L2 acquisition.
• Findings contribute to understanding the cognitive mechanisms underlying long-term L2 learning.