Mapping Intrinsic Timescales of the Elderly Brain and the Relationship with Mnemonic Discrimination

1168 

Abstract Submission 

Kaichao Wu¹, Leonardo Gollo¹

¹Monash University, Melbourne, Victoria

Kaichao Wu  
Monash University
Melbourne, Victoria

Leonardo Gollo  
Monash University
Melbourne, Victoria

Intrinsic timescales of brain regions indicate the duration of which neural information is likely stored in a given brain region, and may represent a fundamental property for understanding cognitive processes[1-2]. This study seeks to evaluate the alteration of intrinsic timescales in the elderly compared with young adults. Considering the neural timescale a basis of the functional hierarchy in the brain[3-4], we examine the intrinsic timescales of fMRI BOLD signals across brain networks. In addition, as a typical example of brain processing sensory inputs, the mnemonic distinguish ability (the ability to distinguish existing memories from input[5-6]) of the brain was investigated together with its association with intrinsic timescales.

Resting-state fMRI scans were obtained from the University of North Carolina samples at Greensboro[5]. The participants were 28 elderly adults (61–80 years old, mean: 69.82, SD:5.64) and 34 young (18–32 years old, mean: 22.21, SD: 3.65). Participants' mnemonic discrimination ability was measured by the lure discrimination index (LDI, the younger group: mean: 0.2630, SD: 0.1918; the elderly group: mean: 0.0992, SD: 0.1946), calculated as the difference in response probabilities if the participants give a similar response to lures and foils in the mnemonic discrimination task [5]. A spatial group ICA with a set of 100 components was performed on the preprocessed and denoised BOLD signal[7] ( see Fig. 1 A). After the removal of noise-related components, the 60 top components were retained. The time courses of non-noise components were then post-processed, and a band-pass filtered was applied (0.023–0.1 Hz). The intrinsic timescale was defined as the area under the curve of the autocorrelation function (ACF) from one to the time lag in which the autocorrelation first reaches a zero value (See Fig. 1 B). Repeating this procedure for all ICA components, an intrinsic timescale map of the whole brain was computed for each participant.

The spatial map of the 60 recognized ICA components where they were assigned to 6 functional networks (details can be seen in Fig. 1C). As an example, Fig.1 D illustrates representative autocorrelation functions of the Anterior cingulum cortex (ACC). Figure 2 A shows that the elderly population exhibits reduced whole-brain intrinsic timescales across all functional networks compared to younger adults. For the elderly cohort, the ANOVA test shows that the intrinsic timescale is different across networks (F = 15.76). In particular, a post-hoc t-test shows that the subcortical network has a significantly lower intrinsic timescale than the networks in the high-order functions (*p < 0.001, FDR corrected, see Fig. 2 B), demonstrating the hierarchical structure of intrinsic timescale in the elder brain. Furthermore, we found that the cuneus area in the VIS network, which is most known for its involvement in basic visual processing, has a significant correlation with the LDI (r = 0.2532, p< 0.05, FDR corrected. Fig. 2 C).

Our findings demonstrate that the intrinsic timescale of the elderly is significantly reduced. The reduced intrinsic timescales in the elderly brains could be associated with cognitive changes associated with aging, such as reduced Information Integration, and cognitive flexibility[8]. The elder brain exhibits a hierarchy of the intrinsic timescales with shorter intrinsic timescales at the subcortical regions than other functional networks. Finally, we found a significant association between the intrinsic timescales of the cuneus area and mnemonic discrimination ability. This finding indicates that decreased intrinsic timescale in the cuneus can be linked to challenges in the discrimination and accurate retrieval of memories in the elderly. By examining the temporal dynamics of the brain's functional networks, this study not only advances our understanding of the aging brain but also offers cues for investigating cognitive changes in the elderly.

Aging ¹

Other Methods

BOLD fMRI

Perception and Attention Other ²

Aging

Cognition

Modeling

Plasticity

Sub-Cortical

Other - Dynamics; Intrinsic timescale