How Aging Shapes Semantic Memory: Exploring the Relationships between Language Abilities, Network Construction, and Word Characteristics on the Structure of Semantic Networks in Younger and Older Adults
Restricted (Penn State Only)
- Author:
- Cosgrove, Abigail
- Graduate Program:
- Psychology
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- June 15, 2023
- Committee Members:
- Kristin Buss (she/her), Program Head/Chair
Chaleece Sandberg (she/her), Outside Unit & Field Member
Roger Beaty, Major Field Member
Michele Diaz, Chair & Dissertation Advisor
Janet van Hell, Major Field Member - Keywords:
- Aging
Semantic Memory
Network Science - Abstract:
- Healthy aging is associated with declines across a variety of cognitive domains, including memory, processing speed, executive functioning, and language production ability (Burke & Shafto, 2008; Salthouse, 2010). Life experiences and conceptual knowledge, however, tend to increase with age (Park et al., 2002). Individuals continuously acquire and retain new words, concepts, and ideas that require representation in the semantic system. Computationally modeled networks allow us to analyze the interactions within that system (Siew et al., 2019). Previous work focused on semantic networks and aging have found that with increased age, semantic memory becomes less efficient, less organized, and sparsely connected (Cosgrove et al., 2021; Wulff et al., 2019). Yet most studies examining semantic networks have focused exclusively on concrete concepts using aggregated, group-based analyses. Abstract concepts comprise a significant portion of our knowledge, and group-based analyses, while important, limit our ability to look at individual differences which may be even more important for older adults. Moreover, language production abilities could be influenced by differences in individual network properties, local structural characteristics, or linguistic features of the word (i.e., semantic diversity, concreteness). This dissertation addressed three primary research questions: are there specific individual network properties that influence language abilities, are age-related differences in semantic networks related to age-related differences in vocabulary (larger, more diverse) or language production (age-related declines), and do specific word level characteristics play a critical role in the structure of semantic memory organization? Our first study explored individual differences in the aging lexicon through individual-based semantic memory networks of younger and older adults. This study’s results replicated previous group-based findings. In addition, vocabulary knowledge was significantly related to the semantic memory network measures. Corresponding with the age effects, increases in vocabulary were associated with more segregated, less robust, and less efficient networks, which may reflect the critical role that the accumulation of knowledge has on semantic memory structure (Cosgrove et al., 2023). The second study examined age-related differences in semantic networks that contain both abstract and concrete words. Network science provides the tools to quantitatively examine the differences between abstract and concrete words in semantic memory. Adding more abstract words to semantic memory estimations significantly altered the network structures and played a critical role in age effects. Abstract word networks were more efficient and more interconnected than concrete word networks, and there were no age-related individual differences with semantic memory networks containing both concrete and abstract words. The third study hones in on the specific lexical features (i.e., semantic diversity and concreteness) of words in semantic memory and assessed their role in the local structural organization of younger and older adult networks. Findings suggest that the semantically diverse and abstract words had stronger relationships and were more interconnected in the networks, regardless of age. Overall, these studies advance the field of network science by broadening the scope of the aging mental lexicon through individual differences measures, the inclusion of abstract words, and by analyzing effects of specific word characteristics on network structure.