A neural implementation of cognitive reserve: insights from a longitudinal fMRI study of set-switching in aging
Data files
Dec 13, 2024 version files 84.74 KB
-
A_Neural_Implementation_of_Cognitive_Re.zip
79.65 KB
-
README.md
5.10 KB
Abstract
The dataset (Data.mat) comprises variables extracted from demographic information, cognitive task performance, and MRI/fMRI imaging of a longitudinal study of 52 adults aged 60–71, evaluated at baseline and after a 5-year follow-up. The analysis code is structured into two main codes: (A) Main Analysis in MATLAB (Code1_MainAnalysis_Matlab.m), which utilizes the primary dataset (Data.mat) to examine brain variables, model cognitive trajectories, and prepare data for further interaction analysis. (B) Interaction Visualization in R (Code2_Johnson-Neyman_R.R), which produces interaction effect visualizations using the output from MATLAB analysis.
This document provides documentation for the dataset and code used in the paper by Fatemeh Hasanzadeh, Christian Habeck, Yunglin Gazes, and Yaakov Stern.
Dataset: Data.mat
The following section describes the variables contained in the Data.mat file, which includes longitudinal measurements from 52 older adults at baseline and 5-year follow-up. The dataset contains demographic, behavioral, and neuroimaging data collected as part of the study.
Demographic and Study Variables
| Variable | Description |
|---|---|
| subid | Unique participant identifier |
| TimePoint | Study timepoint (1 = baseline, 2 = 5-year follow-up) |
| scannerID | MRI scanner identifier (1= Philips/ 3= PRISMA) |
| Age | Participant age at each time point |
| Sex | Participant biological sex (1=Male , 2=Female) |
| IntraCranialVolume | intracranial volume |
Behavioral Measures
| Variable | Description |
|---|---|
| ECF_Sing_propCor | Proportion of correct responses in single-task condition |
| ECF_Dual_propCor | Proportion of correct responses in dual-task condition |
| ECF_Sing_CORmedRespTime | Median response time* for correct responses in single-task (seconds) |
| ECF_Dual_CORmedRespTime | Median response time for correct responses in dual-task (seconds) |
| SwitchCost_RT | Switch cost (difference between dual and single-task response time) |
*response time or reaction time are the same.
Functional MRI Measures
| Variable | Description |
|---|---|
| OrTSingle | Ordinal Trend pattern expression score for single-task condition |
| OrTDual | Ordinal Trend pattern expression score for dual-task condition |
Structural MRI Measures
Naming convention:
- lh_* : Left hemisphere regions
- rh_* : Right hemisphere regions
For each cortical region, the dataset includes two types of measurements:
- Volume measurements (e.g., 'lh_bankssts_volume')
- Thickness measurements (e.g., 'lh_bankssts_thickness')
Important Notes
- All volume measurements are in cubic millimeters (mm³)
- All thickness measurements are in millimeters (mm)
- Brain regions are segmented based on the Desikan-Killiany Atlas
- Structural measures were processed using FreeSurfer v5.1.3
- Scanner harmonization was performed using the COMBAT method
- The dataset includes measurements from two different scanners:
- Baseline: 3.0T Philips Achieva scanner (all participants)
- Follow-up:
- 39 participants in Philips Achieva
- 13 participants on 3.0T Siemens PRISMA
Code Structure and Analysis Pipeline
The analysis consists of two main codes:
A. Code1_MainAnalysis_Matlab.m– MATLAB Code including Main Analysis
1-Data Preparation
- Loads and organizes longitudinal data from 52 older adults
- Extracts baseline and follow-up measurements
2-Behavioral Analysis
- Performs repeated measures ANOVA on reaction times
- Analyzes single and dual-task conditions
- Calculates and visualizes individual and population trajectories of switch costs across time points
3-Longitudinal Change Analysis
- Obtain change scores between baseline and follow-up
- Calculates differential Ordinal Trend (dOrT) scores
- Processes demographic and clinical variables
4-Brain Structure Analysis
- Implements elastic net regression on brain measures
- Combines volume and thickness measurements from multiple regions
- Generates a brain change index predicting switch cost changes
5-Main Statistical Model
- Fits linear model examining:
- Brain change index and dOrT effects on task performance
- dOrT moderation effects
- Control variables (age, sex, scanner, etc.)
- Includes outlier removal using the IQR method
- Generates data for the Johnson-Neyman plot
Input: Data.mat (main dataset)
Generated Outputs:
- Selected_BrainVariables_ElasticNet.csv (selected brain regions)
- Trajectories of Switch Costs.tif (visualization)
- Model_Main.csv (model coefficients)
- Data4JNplot.csv (data for Interaction Analysis)
B. Code2_Johnson-Neyman_R.R – R code including Interaction visualization
Secondary analysis in R
Creates simple slopes and Johnson-Neyman plots for interaction visualization
Input: 'Data4JNplot.csv’ (output from MATLAB analysis)
Generated Outputs:
- SimpleSlopesPlot.tif (visualization)
- Johnson-NeymanPlot.tif (visualization)
This dataset does not contain any direct fMRI data. The fMRI data in our study were analyzed using a specific technique called Ordinal Trend Canonical Variates Analysis (OrT CVA). Through this approach, we derived a measure known as the OrT score, which was calculated for both single and dual-task conditions. This OrT score is the only fMRI-related variable included in the dataset that we uploaded.
To provide further details, OrT CVA is a multivariate data-driven technique that identifies patterns of regional functional activation that show a monotonic change across multiple experimental conditions (in the current study single and dual conditions). The extracted functional activation patterns, called ordinal trends (OrT), indicate sustained activity across graduated increase in task demand (Habeck, Krakauer, et al., 2005; Habeck, Rakitin, et al., 2005). The technique utilizes a specialized design matrix to enhance variance contributions from patterns that exhibit within-subject increases in pattern scores from single to dual conditions. The test statistic that is used to assess the significance of the task condition relationship of the derived activation pattern is the number of exceptions i.e. the number of individuals showing decreased pattern expression from single to dual condition and thus violate the majority rule of an increase. A null distribution is generated using a permutation test with 1,000 iterations, where condition assignments are randomized within participants. The p-value is determined by the fraction of times the permutation test yields a number of exceptions as low or lower than the point estimate. To ensure the robustness of voxel loadings in the derived pattern, a simple bootstrap technique is employed. The data are resampled with replacement (without randomizing subject and condition assignments), and the analytic point-estimate process is repeated 500 times. Z-values for the voxel loadings are computed as the ratio of the point estimate of the loading divided by the bootstrap standard deviation around this point estimate. The OrT score which is the average level of activation of the identified patterns is computed for both single and dual conditions at baseline.