INS NYC 2024 Program

Poster	Poster Session 04 Program Schedule 02/15/2024 12:00 pm - 01:15 pm Room: Majestic Complex (Posters 61-120) Poster Session 04: Neuroimaging \| Neurostimulation/Neuromodulation \| Teleneuropsychology/Technology Final Abstract #85 Functional Connectivity in Distributed Cortical Networks Associated with Semantic Processing in Healthy Older Adults Alexandria O'Neal, University of Florida, Gainesville, United States Kailey Langer, University of Florida, Gainesville, United States Amanda Garcia, James A Haley Veterans Hospital, Tampa, United States Alexa Chen, Florida Atlantic University, Boca Raton, United States Eric Porges, University of Florida, Gainesville, United States John Williamson, University of Florida, Gainesville, United States Adam Woods, University of Florida, Gainesville, United States Ronald Cohen, University of Florida, Gainesville, United States Category: Aging Keyword 1: semantic processing Keyword 2: aging (normal) Keyword 3: neuroimaging: functional connectivity Objective: Semantic processing is a complex function involving lexical knowledge, cognitive control, and sensorimotor integration. Beyond putative hubs for semantic processing, evidence suggests there are distributed brain regions involved in performing semantic tasks. While semantic knowledge increases with age and is typically well-preserved, cognitive aging is widely associated with decline in specialization and efficiency of pre-defined cortical networks. Within the present study, we aimed to characterize patterns of functional connectivity in broad, higher order cortical networks, along with a putative semantic hub network, during a semantic processing task. Participants and Methods: 34 healthy older adults (62% women, 94% White, 70.32 ± 11.24 years old, 16.41 ± 2.6 years of education) participated in task-based functional MRI within a study of successful cognitive aging. Participants performed semantic association decision tasks in a mixed event-related block fMRI paradigm involving binary judgements as to whether abstract and concrete words pairs were related (i.e., semantically associated). A rhyming condition using pseudo-word pairs for phonemic processing was also included. Within-network connectivity values from the cingulo-operculate network (CON), frontal-parietal control network (FPCN), default mode network (DMN), and dorsal attention network (DAN) (Yeo et al., 2011), as well as a manually defined semantic network (Garcia et al., 2022), were used for individual linear multiple regressions to assess task effects for each network, and significant effects were clarified through post-hoc t-tests. To investigate effects at the level of ROI-ROI connections within networks, follow-up analyses were conducted in CONN Toolbox for networks that demonstrated overall task effects. Results: Connectivity within the semantic network was associated with task, with significantly greater connectivity observed during concrete and abstract semantic conditions compared to the rhyme condition. Greater connectivity was observed for the rhyme condition relative to abstract and concrete conditions within DAN and relative to the abstract condition (but not concrete condition) within DMN. Connectivity within FPCN and CON did not demonstrate significant task effects. Within the semantic network, increased connectivity was observed between the inferior frontal gyrus (IFG) and other network ROIs (angular gyrus (AG), anterior medial temporal gyrus (aMTG)) during abstract and concrete conditions relative to rhyme. ROI-ROI analyses within the DAN revealed patterns of strengthened connectivity between bilateral pre-frontal and posterior regions during rhyme relative to semantic conditions. Patterns of altered connectivity between ROIs were mixed when contrasting semantic and rhyme conditions within the DMN. Conclusions: Our findings demonstrate strong task relevance via increased semantic network connectivity during semantic processing within healthy older adults. This also supports preservation of the semantic system, as an integral component of semantic processing, despite trends of decreased specialization of brain networks within aging. Further, these results demonstrate variable engagement of higher-order cortical networks as a function of task demand, with increased coherence within DAN and DMN networks during processing of phonemic information. This distinction may reflect relative novelty of the rhyming task, underscoring flexible recruitment of different networks to support task completion in healthy aging. Collectively, these findings contribute to our understanding of the underlying neural mechanisms involved in semantic processing in the context of cognitive aging.

Poster

02/15/2024
12:00 pm - 01:15 pm
Room: Majestic Complex (Posters 61-120)

Poster Session 04: Neuroimaging | Neurostimulation/Neuromodulation | Teleneuropsychology/Technology

Final Abstract #85

Functional Connectivity in Distributed Cortical Networks Associated with Semantic Processing in Healthy Older Adults

Alexandria O'Neal, University of Florida, Gainesville, United States
Kailey Langer, University of Florida, Gainesville, United States
Amanda Garcia, James A Haley Veterans Hospital, Tampa, United States
Alexa Chen, Florida Atlantic University, Boca Raton, United States
Eric Porges, University of Florida, Gainesville, United States
John Williamson, University of Florida, Gainesville, United States
Adam Woods, University of Florida, Gainesville, United States
Ronald Cohen, University of Florida, Gainesville, United States

Category: Aging

Keyword 1: semantic processing
Keyword 2: aging (normal)
Keyword 3: neuroimaging: functional connectivity

Objective:

Semantic processing is a complex function involving lexical knowledge, cognitive control, and sensorimotor integration. Beyond putative hubs for semantic processing, evidence suggests there are distributed brain regions involved in performing semantic tasks. While semantic knowledge increases with age and is typically well-preserved, cognitive aging is widely associated with decline in specialization and efficiency of pre-defined cortical networks. Within the present study, we aimed to characterize patterns of functional connectivity in broad, higher order cortical networks, along with a putative semantic hub network, during a semantic processing task.

Participants and Methods:

34 healthy older adults (62% women, 94% White, 70.32 ± 11.24 years old, 16.41 ± 2.6 years of education) participated in task-based functional MRI within a study of successful cognitive aging. Participants performed semantic association decision tasks in a mixed event-related block fMRI paradigm involving binary judgements as to whether abstract and concrete words pairs were related (i.e., semantically associated). A rhyming condition using pseudo-word pairs for phonemic processing was also included. Within-network connectivity values from the cingulo-operculate network (CON), frontal-parietal control network (FPCN), default mode network (DMN), and dorsal attention network (DAN) (Yeo et al., 2011), as well as a manually defined semantic network (Garcia et al., 2022), were used for individual linear multiple regressions to assess task effects for each network, and significant effects were clarified through post-hoc t-tests. To investigate effects at the level of ROI-ROI connections within networks, follow-up analyses were conducted in CONN Toolbox for networks that demonstrated overall task effects.

Results:

Connectivity within the semantic network was associated with task, with significantly greater connectivity observed during concrete and abstract semantic conditions compared to the rhyme condition. Greater connectivity was observed for the rhyme condition relative to abstract and concrete conditions within DAN and relative to the abstract condition (but not concrete condition) within DMN. Connectivity within FPCN and CON did not demonstrate significant task effects. Within the semantic network, increased connectivity was observed between the inferior frontal gyrus (IFG) and other network ROIs (angular gyrus (AG), anterior medial temporal gyrus (aMTG)) during abstract and concrete conditions relative to rhyme. ROI-ROI analyses within the DAN revealed patterns of strengthened connectivity between bilateral pre-frontal and posterior regions during rhyme relative to semantic conditions. Patterns of altered connectivity between ROIs were mixed when contrasting semantic and rhyme conditions within the DMN.

Conclusions:

Our findings demonstrate strong task relevance via increased semantic network connectivity during semantic processing within healthy older adults. This also supports preservation of the semantic system, as an integral component of semantic processing, despite trends of decreased specialization of brain networks within aging. Further, these results demonstrate variable engagement of higher-order cortical networks as a function of task demand, with increased coherence within DAN and DMN networks during processing of phonemic information. This distinction may reflect relative novelty of the rhyming task, underscoring flexible recruitment of different networks to support task completion in healthy aging. Collectively, these findings contribute to our understanding of the underlying neural mechanisms involved in semantic processing in the context of cognitive aging.