Load Required Libraries
library(dplyr)
##
## Attaching package: 'dplyr'
## The following objects are masked from 'package:stats':
##
## filter, lag
## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, union
library(tidyverse)
## ── Attaching packages ─────────────────────────────────────── tidyverse 1.3.1 ──
## ✔ ggplot2 3.3.6 ✔ purrr 0.3.4
## ✔ tibble 3.1.7 ✔ stringr 1.4.0
## ✔ tidyr 1.2.0 ✔ forcats 0.5.1
## ✔ readr 2.1.2
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag() masks stats::lag()
library(readxl)
library(ggplot2)
library(lmerTest)
## Loading required package: lme4
## Loading required package: Matrix
## Warning: package 'Matrix' was built under R version 4.2.1
##
## Attaching package: 'Matrix'
## The following objects are masked from 'package:tidyr':
##
## expand, pack, unpack
##
## Attaching package: 'lmerTest'
## The following object is masked from 'package:lme4':
##
## lmer
## The following object is masked from 'package:stats':
##
## step
library(tidyr)
# Begin data upload
full_cross_sectional_data <- read.csv('Z:/Brackins dissertation r script and dummy data frames/cross-sectional dummy data frame.csv')
head(full_cross_sectional_data)
## subject_id sub_study SNQ_Quantity SNQ_Quality snq_total RAVLT Age sex
## 1 1 Discoverysci 40 3 38 11 84 Female
## 2 2 Discoverysci 39 9 37 4 61 Male
## 3 3 Discoverysci 31 12 35 15 85 Male
## 4 4 Discoverysci 29 9 47 10 58 Female
## 5 5 Discoverysci 36 3 51 3 23 Male
## 6 6 Discoverysci 41 9 37 11 66 Male
## ethnicity race personality
## 1 Not Hispanic or Latino or Spanish white 1
## 2 Not Hispanic or Latino or Spanish Black or African American 2
## 3 Not Hispanic or Latino or Spanish white 3
## 4 Not Hispanic or Latino or Spanish white 2
## 5 Not Hispanic or Latino or Spanish Asian 3
## 6 Not Hispanic or Latino or Spanish white 2
## Left.Thalamus.Proper Right.Thalamus.Proper thalamus_total_volume Left.Caudate
## 1 5773.2 5897.8 11671.0 2954.1
## 2 7300.9 6999.6 14300.5 3105.6
## 3 6751.6 6128.8 12880.4 3276.8
## 4 6907.9 6661.7 13569.6 2787.9
## 5 8358.7 8224.3 16583.0 3645.5
## 6 6321.5 6158.5 12480.0 2612.1
## Right.Caudate caudate_total_volume Left.Putamen Right.Putamen
## 1 3044.6 5998.7 3373.0 3401.6
## 2 3129.6 6235.2 4493.3 4626.2
## 3 3456.4 6733.2 4336.0 4739.3
## 4 2848.7 5636.6 3585.7 3777.8
## 5 3579.6 7225.1 5324.1 5072.2
## 6 2910.9 5523.0 3837.8 3753.5
## putamen_total_volume Left.Hippocampus Right.Hippocampus
## 1 6774.6 3679.3 3643.8
## 2 9119.5 3694.1 3768.9
## 3 9075.3 3101.9 3732.1
## 4 7363.5 3790.6 3899.8
## 5 10396.3 4127.0 4731.2
## 6 7591.3 4064.9 4077.8
## hippocampus_total_volume Left.Amygdala Right.Amygdala amygdala_total_volume
## 1 7483 1307.3 1391.3 2383
## 2 6708 1486.8 1417.6 2529
## 3 8318 847.0 1350.2 2338
## 4 7763 1303.0 1468.0 2126
## 5 7106 2004.6 1920.3 2576
## 6 8536 1296.0 1666.3 3324
## lhCortexVol rhCortexVol CortexVol SubCortGrayVol TotalGrayVol
## 1 180099.9 180873.9 360973.8 46657 515272.8
## 2 209443.4 215176.8 424620.3 54962 595006.3
## 3 212829.6 223411.2 436240.8 50870 596354.8
## 4 235435.5 236103.3 471538.8 48988 625777.8
## 5 243773.6 246826.7 490600.3 62668 677188.3
## 6 203425.2 203020.1 406445.2 49865 563753.2
## EstimatedTotalIntraCranialVol lh_rostralmiddlefrontal_thickness
## 1 1158574 2.344
## 2 1374227 2.580
## 3 1479439 2.242
## 4 1514698 2.464
## 5 1553358 2.579
## 6 1410264 2.286
## rh_rostralmiddlefrontal_thickness rostral_middle_frontal_average_thickness
## 1 2.394 2.3690
## 2 2.758 2.6690
## 3 2.306 2.2740
## 4 2.385 2.4245
## 5 2.553 2.5660
## 6 2.251 2.2685
## lh_caudalmiddlefrontal_thickness rh_caudalmiddlefrontal_thickness
## 1 2.560 2.588
## 2 2.998 3.093
## 3 2.434 2.499
## 4 2.619 2.638
## 5 2.825 2.721
## 6 2.429 2.477
## caudal_middle_average_thickness lh_parsopercularis_thickness
## 1 2.5740 2.552
## 2 3.0455 2.944
## 3 2.4665 2.400
## 4 2.6285 2.615
## 5 2.7730 2.853
## 6 2.4530 2.557
## rh_parsopercularis_thickness lh_parsorbitalis_thickness
## 1 2.618 2.612
## 2 2.853 3.017
## 3 2.556 2.689
## 4 2.539 3.144
## 5 2.846 2.876
## 6 2.418 2.619
## rh_parsorbitalis_thickness lh_parstriangularis_thickness
## 1 2.932 2.621
## 2 3.102 2.779
## 3 2.473 2.258
## 4 2.528 2.507
## 5 2.738 2.602
## 6 2.556 2.117
## rh_parstriangularis_thickness left_inferior_pfc_average_thickness
## 1 2.680 1.94625
## 2 2.640 2.18500
## 3 2.404 1.83675
## 4 2.391 2.06650
## 5 2.594 2.08275
## 6 2.198 1.82325
## right_inferior_pfc_average_thickness inferior_pfc_average_thickness
## 1 2.743333 2.344792
## 2 2.865000 2.525000
## 3 2.477667 2.157208
## 4 2.486000 2.276250
## 5 2.726000 2.404375
## 6 2.390667 2.106958
## lh_superiorfrontal_thickness rh_superiorfrontal_thickness
## 1 2.825 2.693
## 2 2.825 3.040
## 3 2.825 2.512
## 4 2.825 2.687
## 5 2.825 2.871
## 6 2.825 2.489
## superior_frontal_average_thickness left_pfc_average_thickness
## 1 2.7590 1.93505
## 2 2.9325 2.11760
## 3 2.6685 1.86755
## 4 2.7560 1.99490
## 5 2.8480 2.06235
## 6 2.6570 1.87265
## right_pfc_average_thickness pfc_average_thickness
## 1 2.083667 2.009358
## 2 2.939000 2.528300
## 3 2.448667 2.158108
## 4 2.549000 2.271950
## 5 2.717750 2.390050
## 6 2.401917 2.137283
## lh_inferiorparietal_thickness rh_inferiorparietal_thickness
## 1 2.699 2.677
## 2 2.503 2.561
## 3 2.249 2.423
## 4 2.467 2.548
## 5 2.564 2.602
## 6 2.291 2.378
## inferior_parietal_average_thickness lh_superiorparietal_thickness
## 1 2.6880 2.397
## 2 2.5320 2.187
## 3 2.3360 2.042
## 4 2.5075 2.257
## 5 2.5830 2.315
## 6 2.3345 2.079
## rh_superiorparietal_thickness superior_parietal_average_thickness
## 1 2.291 2.344
## 2 2.507 2.347
## 3 2.226 2.134
## 4 2.329 2.293
## 5 2.085 2.200
## 6 2.235 2.157
## left_ppc_average_thickness right_ppc_average_thickness ppc_average_thickness
## 1 2.5480 2.4840 2.51600
## 2 2.3450 2.5340 2.43950
## 3 2.1455 2.3245 2.23500
## 4 2.3620 2.4385 2.40025
## 5 2.4395 2.3435 2.39150
## 6 2.1850 2.3065 2.24575
## lh_MeanThickness_thickness rh_MeanThickness_thickness mean_thickness_average
## 1 2.56971 2.55457 2.562140
## 2 2.63834 2.67018 2.654260
## 3 2.34325 2.40386 2.373555
## 4 2.54392 2.53097 2.537445
## 5 2.59484 2.61362 2.604230
## 6 2.38080 2.38929 2.385045
## lh_rostralmiddlefrontal_volume rh_rostralmiddlefrontal_volume
## 1 9516 10586
## 2 12052 14879
## 3 12786 13033
## 4 14226 15372
## 5 15189 15461
## 6 13507 12734
## rostral_middle_frontal_total_volume lh_caudalmiddlefrontal_volume
## 1 20102 4300
## 2 26931 4944
## 3 25819 5319
## 4 29598 7763
## 5 30650 8307
## 6 26241 5147
## rh_caudalmiddlefrontal_volume caudal_middle_total_volume
## 1 3776 8076
## 2 5189 10133
## 3 5555 10874
## 4 5910 13673
## 5 6075 14382
## 6 4194 9341
## lh_parsopercularis_volume rh_parsopercularis_volume lh_parsorbitalis_volume
## 1 3419 3000 1527
## 2 4911 4267 2110
## 3 4424 3968 2606
## 4 4094 3737 2416
## 5 8615 3568 2630
## 6 3737 2949 2423
## rh_parsorbitalis_volume lh_parstriangularis_volume rh_parstriangularis_volume
## 1 2310 3163 3935
## 2 2368 2813 3758
## 3 2074 2939 3782
## 4 2901 3257 3998
## 5 2621 4049 3546
## 6 2433 2715 2889
## left_inferior_pfc_total_volume right_inferior_pfc_total_volume
## 1 8109 9245
## 2 9834 10393
## 3 9969 9824
## 4 9767 10636
## 5 15294 9735
## 6 8875 8271
## inferior_pfc_total_volume lh_superiorfrontal_volume rh_superiorfrontal_volume
## 1 17354 18279 15867
## 2 20227 20659 19351
## 3 19793 21598 20955
## 4 20403 23104 21690
## 5 25029 24464 18694
## 6 17146 19484 19499
## superior_frontal_total_volume left_pfc_total_volume right_pfc_total_volume
## 1 34146 40204 39474
## 2 40010 47489 49812
## 3 42553 49672 49367
## 4 44794 54860 53608
## 5 43158 63254 49965
## 6 38983 47013 44698
## pfc_total_volume lh.rostralanteriorcingulate rh.rostralanteriorcingulate
## 1 79678 8280 10421
## 2 97301 11336 13624
## 3 99039 12039 12930
## 4 108468 12213 14344
## 5 113219 14096 14063
## 6 91711 9078 10673
## lh.posteriorcingulate rh.posteriorcingulate lh.caudalanteriorcingulate
## 1 18701 9684 9665
## 2 24960 9812 11474
## 3 24969 9826 10409
## 4 26557 11910 12520
## 5 28159 10029 13735
## 6 19751 11365 9813
## rh.caudalanteriorcingulate lh.isthmuscingulate rh.isthmuscingulate
## 1 19349 17964 20086
## 2 21286 21148 25098
## 3 20235 21865 23339
## 4 24430 24123 26864
## 5 23764 24125 27798
## 6 21178 20443 20486
## ppc_total_volume
## 1 38050
## 2 46246
## 3 45204
## 4 50987
## 5 51923
## 6 40929
# Select specific columns
selected_cross_sectional_data <- full_cross_sectional_data %>%
select(subject_id, sub_study, SNQ_Quality, SNQ_Quantity, snq_total, RAVLT, Age, sex, personality, ethnicity, race,
hippocampus_total_volume, amygdala_total_volume, EstimatedTotalIntraCranialVol,
lh.caudalanteriorcingulate, rh.caudalanteriorcingulate,
lh.rostralanteriorcingulate, rh.rostralanteriorcingulate,
lh.posteriorcingulate, rh.posteriorcingulate, lh.isthmuscingulate, rh.isthmuscingulate)
head(selected_cross_sectional_data)
## subject_id sub_study SNQ_Quality SNQ_Quantity snq_total RAVLT Age sex
## 1 1 Discoverysci 3 40 38 11 84 Female
## 2 2 Discoverysci 9 39 37 4 61 Male
## 3 3 Discoverysci 12 31 35 15 85 Male
## 4 4 Discoverysci 9 29 47 10 58 Female
## 5 5 Discoverysci 3 36 51 3 23 Male
## 6 6 Discoverysci 9 41 37 11 66 Male
## personality ethnicity race
## 1 1 Not Hispanic or Latino or Spanish white
## 2 2 Not Hispanic or Latino or Spanish Black or African American
## 3 3 Not Hispanic or Latino or Spanish white
## 4 2 Not Hispanic or Latino or Spanish white
## 5 3 Not Hispanic or Latino or Spanish Asian
## 6 2 Not Hispanic or Latino or Spanish white
## hippocampus_total_volume amygdala_total_volume EstimatedTotalIntraCranialVol
## 1 7483 2383 1158574
## 2 6708 2529 1374227
## 3 8318 2338 1479439
## 4 7763 2126 1514698
## 5 7106 2576 1553358
## 6 8536 3324 1410264
## lh.caudalanteriorcingulate rh.caudalanteriorcingulate
## 1 9665 19349
## 2 11474 21286
## 3 10409 20235
## 4 12520 24430
## 5 13735 23764
## 6 9813 21178
## lh.rostralanteriorcingulate rh.rostralanteriorcingulate lh.posteriorcingulate
## 1 8280 10421 18701
## 2 11336 13624 24960
## 3 12039 12930 24969
## 4 12213 14344 26557
## 5 14096 14063 28159
## 6 9078 10673 19751
## rh.posteriorcingulate lh.isthmuscingulate rh.isthmuscingulate
## 1 9684 17964 20086
## 2 9812 21148 25098
## 3 9826 21865 23339
## 4 11910 24123 26864
## 5 10029 24125 27798
## 6 11365 20443 20486
# Create new column for 'total_cingulate_volume'
selected_cross_sectional_data$total_cingulate_volume <-
selected_cross_sectional_data$lh.caudalanteriorcingulate +
selected_cross_sectional_data$rh.caudalanteriorcingulate +
selected_cross_sectional_data$lh.rostralanteriorcingulate +
selected_cross_sectional_data$rh.rostralanteriorcingulate +
selected_cross_sectional_data$lh.posteriorcingulate +
selected_cross_sectional_data$rh.posteriorcingulate +
selected_cross_sectional_data$lh.isthmuscingulate +
selected_cross_sectional_data$rh.isthmuscingulate
# Check the new data frame
head(selected_cross_sectional_data)
## subject_id sub_study SNQ_Quality SNQ_Quantity snq_total RAVLT Age sex
## 1 1 Discoverysci 3 40 38 11 84 Female
## 2 2 Discoverysci 9 39 37 4 61 Male
## 3 3 Discoverysci 12 31 35 15 85 Male
## 4 4 Discoverysci 9 29 47 10 58 Female
## 5 5 Discoverysci 3 36 51 3 23 Male
## 6 6 Discoverysci 9 41 37 11 66 Male
## personality ethnicity race
## 1 1 Not Hispanic or Latino or Spanish white
## 2 2 Not Hispanic or Latino or Spanish Black or African American
## 3 3 Not Hispanic or Latino or Spanish white
## 4 2 Not Hispanic or Latino or Spanish white
## 5 3 Not Hispanic or Latino or Spanish Asian
## 6 2 Not Hispanic or Latino or Spanish white
## hippocampus_total_volume amygdala_total_volume EstimatedTotalIntraCranialVol
## 1 7483 2383 1158574
## 2 6708 2529 1374227
## 3 8318 2338 1479439
## 4 7763 2126 1514698
## 5 7106 2576 1553358
## 6 8536 3324 1410264
## lh.caudalanteriorcingulate rh.caudalanteriorcingulate
## 1 9665 19349
## 2 11474 21286
## 3 10409 20235
## 4 12520 24430
## 5 13735 23764
## 6 9813 21178
## lh.rostralanteriorcingulate rh.rostralanteriorcingulate lh.posteriorcingulate
## 1 8280 10421 18701
## 2 11336 13624 24960
## 3 12039 12930 24969
## 4 12213 14344 26557
## 5 14096 14063 28159
## 6 9078 10673 19751
## rh.posteriorcingulate lh.isthmuscingulate rh.isthmuscingulate
## 1 9684 17964 20086
## 2 9812 21148 25098
## 3 9826 21865 23339
## 4 11910 24123 26864
## 5 10029 24125 27798
## 6 11365 20443 20486
## total_cingulate_volume
## 1 114150
## 2 138738
## 3 135612
## 4 152961
## 5 155769
## 6 122787
# ---- begin loading longitudinal time one data----
# Read in longitudinal data frame
longitudinal_data <- read.csv("Z:/Brackins dissertation r script and dummy data frames/longitudinal dummy data frame.csv")
head(longitudinal_data)
## subject_id visit sub_study SNQ_Quantity SNQ_Quality snq_total RAVLT
## 1 1 Time 1 adultlongitudinal 21 10 53 4
## 2 2 Time 2 adultlongitudinal 22 12 37 8
## 3 3 Time 1 adultlongitudinal 44 11 55 1
## 4 4 Time 2 adultlongitudinal 30 12 31 15
## 5 5 Time 1 adultlongitudinal 37 7 40 3
## 6 6 Time 2 adultlongitudinal 42 10 40 3
## Age sex personality ethnicity
## 1 78 Female 1 Not Hispanic or Latino or Spanish
## 2 46 Male 5 Not Hispanic or Latino or Spanish
## 3 64 Male 3 Not Hispanic or Latino or Spanish
## 4 70 Female 2 Not Hispanic or Latino or Spanish
## 5 67 Male 3 Not Hispanic or Latino or Spanish
## 6 62 Male 3 Not Hispanic or Latino or Spanish
## race Left.Thalamus.Proper Right.Thalamus.Proper
## 1 white 5773.2 5897.8
## 2 Black or African American 7300.9 6999.6
## 3 white 6751.6 6128.8
## 4 white 6907.9 6661.7
## 5 Asian 8358.7 8224.3
## 6 white 6321.5 6158.5
## thalamus_total_volume Left.Caudate Right.Caudate caudate_total_volume
## 1 11671.0 2954.1 3044.6 5998.7
## 2 14300.5 3105.6 3129.6 6235.2
## 3 12880.4 3276.8 3456.4 6733.2
## 4 13569.6 2787.9 2848.7 5636.6
## 5 16583.0 3645.5 3579.6 7225.1
## 6 12480.0 2612.1 2910.9 5523.0
## Left.Putamen Right.Putamen putamen_total_volume Left.Hippocampus
## 1 3373.0 3401.6 6774.6 3679.3
## 2 4493.3 4626.2 9119.5 3694.1
## 3 4336.0 4739.3 9075.3 3101.9
## 4 3585.7 3777.8 7363.5 3790.6
## 5 5324.1 5072.2 10396.3 4127.0
## 6 3837.8 3753.5 7591.3 4064.9
## Right.Hippocampus hippocampus_total_volume Left.Amygdala Right.Amygdala
## 1 3643.8 7414 1307.3 1391.3
## 2 3768.9 5356 1486.8 1417.6
## 3 3732.1 7785 847.0 1350.2
## 4 3899.8 7081 1303.0 1468.0
## 5 4731.2 5143 2004.6 1920.3
## 6 4077.8 5975 1296.0 1666.3
## amygdala_total_volume lhCortexVol rhCortexVol CortexVol SubCortGrayVol
## 1 3256 180099.9 180873.9 360973.8 46657
## 2 2345 209443.4 215176.8 424620.3 54962
## 3 2772 212829.6 223411.2 436240.8 50870
## 4 3119 235435.5 236103.3 471538.8 48988
## 5 3261 243773.6 246826.7 490600.3 62668
## 6 3006 203425.2 203020.1 406445.2 49865
## TotalGrayVol EstimatedTotalIntraCranialVol lh_rostralmiddlefrontal_thickness
## 1 515272.8 1158574 2.344
## 2 595006.3 1374227 2.580
## 3 596354.8 1479439 2.242
## 4 625777.8 1514698 2.464
## 5 677188.3 1553358 2.579
## 6 563753.2 1410264 2.286
## rh_rostralmiddlefrontal_thickness rostral_middle_frontal_average_thickness
## 1 2.394 2.3690
## 2 2.758 2.6690
## 3 2.306 2.2740
## 4 2.385 2.4245
## 5 2.553 2.5660
## 6 2.251 2.2685
## lh_caudalmiddlefrontal_thickness rh_caudalmiddlefrontal_thickness
## 1 2.560 2.588
## 2 2.998 3.093
## 3 2.434 2.499
## 4 2.619 2.638
## 5 2.825 2.721
## 6 2.429 2.477
## caudal_middle_average_thickness lh_parsopercularis_thickness
## 1 2.5740 2.552
## 2 3.0455 2.944
## 3 2.4665 2.400
## 4 2.6285 2.615
## 5 2.7730 2.853
## 6 2.4530 2.557
## rh_parsopercularis_thickness lh_parsorbitalis_thickness
## 1 2.618 2.612
## 2 2.853 3.017
## 3 2.556 2.689
## 4 2.539 3.144
## 5 2.846 2.876
## 6 2.418 2.619
## rh_parsorbitalis_thickness lh_parstriangularis_thickness
## 1 2.932 2.621
## 2 3.102 2.779
## 3 2.473 2.258
## 4 2.528 2.507
## 5 2.738 2.602
## 6 2.556 2.117
## rh_parstriangularis_thickness left_inferior_pfc_average_thickness
## 1 2.680 1.94625
## 2 2.640 2.18500
## 3 2.404 1.83675
## 4 2.391 2.06650
## 5 2.594 2.08275
## 6 2.198 1.82325
## right_inferior_pfc_average_thickness inferior_pfc_average_thickness
## 1 2.743333 2.344792
## 2 2.865000 2.525000
## 3 2.477667 2.157208
## 4 2.486000 2.276250
## 5 2.726000 2.404375
## 6 2.390667 2.106958
## lh_superiorfrontal_thickness rh_superiorfrontal_thickness
## 1 2.825 2.693
## 2 2.825 3.040
## 3 2.825 2.512
## 4 2.825 2.687
## 5 2.825 2.871
## 6 2.825 2.489
## superior_frontal_average_thickness left_pfc_average_thickness
## 1 2.7590 1.93505
## 2 2.9325 2.11760
## 3 2.6685 1.86755
## 4 2.7560 1.99490
## 5 2.8480 2.06235
## 6 2.6570 1.87265
## right_pfc_average_thickness pfc_average_thickness
## 1 2.083667 2.009358
## 2 2.939000 2.528300
## 3 2.448667 2.158108
## 4 2.549000 2.271950
## 5 2.717750 2.390050
## 6 2.401917 2.137283
## lh_inferiorparietal_thickness rh_inferiorparietal_thickness
## 1 2.699 2.677
## 2 2.503 2.561
## 3 2.249 2.423
## 4 2.467 2.548
## 5 2.564 2.602
## 6 2.291 2.378
## inferior_parietal_average_thickness lh_superiorparietal_thickness
## 1 2.6880 2.397
## 2 2.5320 2.187
## 3 2.3360 2.042
## 4 2.5075 2.257
## 5 2.5830 2.315
## 6 2.3345 2.079
## rh_superiorparietal_thickness superior_parietal_average_thickness
## 1 2.291 2.344
## 2 2.507 2.347
## 3 2.226 2.134
## 4 2.329 2.293
## 5 2.085 2.200
## 6 2.235 2.157
## left_ppc_average_thickness right_ppc_average_thickness ppc_average_thickness
## 1 2.5480 2.4840 2.51600
## 2 2.3450 2.5340 2.43950
## 3 2.1455 2.3245 2.23500
## 4 2.3620 2.4385 2.40025
## 5 2.4395 2.3435 2.39150
## 6 2.1850 2.3065 2.24575
## lh_MeanThickness_thickness rh_MeanThickness_thickness mean_thickness_average
## 1 2.56971 2.55457 2.562140
## 2 2.63834 2.67018 2.654260
## 3 2.34325 2.40386 2.373555
## 4 2.54392 2.53097 2.537445
## 5 2.59484 2.61362 2.604230
## 6 2.38080 2.38929 2.385045
## lh_rostralmiddlefrontal_volume rh_rostralmiddlefrontal_volume
## 1 9516 10586
## 2 12052 14879
## 3 12786 13033
## 4 14226 15372
## 5 15189 15461
## 6 13507 12734
## rostral_middle_frontal_total_volume lh_caudalmiddlefrontal_volume
## 1 20102 4300
## 2 26931 4944
## 3 25819 5319
## 4 29598 7763
## 5 30650 8307
## 6 26241 5147
## rh_caudalmiddlefrontal_volume caudal_middle_total_volume
## 1 3776 8076
## 2 5189 10133
## 3 5555 10874
## 4 5910 13673
## 5 6075 14382
## 6 4194 9341
## lh_parsopercularis_volume rh_parsopercularis_volume lh_parsorbitalis_volume
## 1 3419 3000 1527
## 2 4911 4267 2110
## 3 4424 3968 2606
## 4 4094 3737 2416
## 5 8615 3568 2630
## 6 3737 2949 2423
## rh_parsorbitalis_volume lh_parstriangularis_volume rh_parstriangularis_volume
## 1 2310 3163 3935
## 2 2368 2813 3758
## 3 2074 2939 3782
## 4 2901 3257 3998
## 5 2621 4049 3546
## 6 2433 2715 2889
## left_inferior_pfc_total_volume right_inferior_pfc_total_volume
## 1 8109 9245
## 2 9834 10393
## 3 9969 9824
## 4 9767 10636
## 5 15294 9735
## 6 8875 8271
## inferior_pfc_total_volume lh_superiorfrontal_volume rh_superiorfrontal_volume
## 1 17354 18279 15867
## 2 20227 20659 19351
## 3 19793 21598 20955
## 4 20403 23104 21690
## 5 25029 24464 18694
## 6 17146 19484 19499
## superior_frontal_total_volume left_pfc_total_volume right_pfc_total_volume
## 1 34146 40204 39474
## 2 40010 47489 49812
## 3 42553 49672 49367
## 4 44794 54860 53608
## 5 43158 63254 49965
## 6 38983 47013 44698
## lh.rostralanteriorcingulate rh.rostralanteriorcingulate lh.posteriorcingulate
## 1 79678 8280 10421
## 2 97301 11336 13624
## 3 99039 12039 12930
## 4 108468 12213 14344
## 5 113219 14096 14063
## 6 91711 9078 10673
## rh.posteriorcingulate lh.caudalanteriorcingulate rh.caudalanteriorcingulate
## 1 18701 9684 9665
## 2 24960 9812 11474
## 3 24969 9826 10409
## 4 26557 11910 12520
## 5 28159 10029 13735
## 6 19751 11365 9813
## lh.isthmuscingulate rh.isthmuscingulate right_ppc_total_volume
## 1 19349 17964 20086
## 2 21286 21148 25098
## 3 20235 21865 23339
## 4 24430 24123 26864
## 5 23764 24125 27798
## 6 21178 20443 20486
## ppc_total_volume
## 1 38050
## 2 46246
## 3 45204
## 4 50987
## 5 51923
## 6 40929
# Filter for Time 1
longitudinal_data_time1 <- longitudinal_data %>%
filter(visit == "Time 1")
head(longitudinal_data_time1)
## subject_id visit sub_study SNQ_Quantity SNQ_Quality snq_total RAVLT
## 1 1 Time 1 adultlongitudinal 21 10 53 4
## 2 3 Time 1 adultlongitudinal 44 11 55 1
## 3 5 Time 1 adultlongitudinal 37 7 40 3
## 4 7 Time 1 adultlongitudinal 41 10 38 7
## 5 9 Time 1 adultlongitudinal 38 11 49 3
## 6 11 Time 1 adultlongitudinal 28 11 53 2
## Age sex personality ethnicity
## 1 78 Female 1 Not Hispanic or Latino or Spanish
## 2 64 Male 3 Not Hispanic or Latino or Spanish
## 3 67 Male 3 Not Hispanic or Latino or Spanish
## 4 70 Male 4 Hispanic or Latino or Spanish
## 5 60 Female 4 Not Hispanic or Latino or Spanish
## 6 44 Male 5 Not Hispanic or Latino or Spanish
## race Left.Thalamus.Proper Right.Thalamus.Proper
## 1 white 5773.2 5897.8
## 2 white 6751.6 6128.8
## 3 Asian 8358.7 8224.3
## 4 white 7268.4 7087.7
## 5 white 7765.0 7022.6
## 6 Black or African American 7485.7 7533.6
## thalamus_total_volume Left.Caudate Right.Caudate caudate_total_volume
## 1 11671.0 2954.1 3044.6 5998.7
## 2 12880.4 3276.8 3456.4 6733.2
## 3 16583.0 3645.5 3579.6 7225.1
## 4 14356.1 2981.0 3013.1 5994.1
## 5 14787.6 3420.9 3299.2 6720.1
## 6 15019.3 3071.1 3272.4 6343.5
## Left.Putamen Right.Putamen putamen_total_volume Left.Hippocampus
## 1 3373.0 3401.6 6774.6 3679.3
## 2 4336.0 4739.3 9075.3 3101.9
## 3 5324.1 5072.2 10396.3 4127.0
## 4 5140.4 5231.6 10372.0 3530.7
## 5 4200.5 4220.1 8420.6 3761.4
## 6 4185.5 4366.5 8552.0 3561.2
## Right.Hippocampus hippocampus_total_volume Left.Amygdala Right.Amygdala
## 1 3643.8 7414 1307.3 1391.3
## 2 3732.1 7785 847.0 1350.2
## 3 4731.2 5143 2004.6 1920.3
## 4 3704.1 5467 1511.2 1521.0
## 5 3865.7 5107 1399.7 1706.0
## 6 3655.6 7906 1608.5 1479.0
## amygdala_total_volume lhCortexVol rhCortexVol CortexVol SubCortGrayVol
## 1 3256 180099.9 180873.9 360973.8 46657
## 2 2772 212829.6 223411.2 436240.8 50870
## 3 3261 243773.6 246826.7 490600.3 62668
## 4 3504 234988.0 236805.8 471793.8 54672
## 5 3576 201729.7 203926.7 405656.4 53984
## 6 2674 208727.0 210406.0 419133.0 53227
## TotalGrayVol EstimatedTotalIntraCranialVol lh_rostralmiddlefrontal_thickness
## 1 515272.8 1158574 2.344
## 2 596354.8 1479439 2.242
## 3 677188.3 1553358 2.579
## 4 636791.8 1455158 2.384
## 5 582896.4 1321438 2.463
## 6 571305.0 1434374 2.459
## rh_rostralmiddlefrontal_thickness rostral_middle_frontal_average_thickness
## 1 2.394 2.3690
## 2 2.306 2.2740
## 3 2.553 2.5660
## 4 2.511 2.4475
## 5 2.382 2.4225
## 6 2.315 2.3870
## lh_caudalmiddlefrontal_thickness rh_caudalmiddlefrontal_thickness
## 1 2.560 2.588
## 2 2.434 2.499
## 3 2.825 2.721
## 4 2.815 2.836
## 5 2.710 2.550
## 6 2.658 2.767
## caudal_middle_average_thickness lh_parsopercularis_thickness
## 1 2.5740 2.552
## 2 2.4665 2.400
## 3 2.7730 2.853
## 4 2.8255 2.828
## 5 2.6300 2.648
## 6 2.7125 2.622
## rh_parsopercularis_thickness lh_parsorbitalis_thickness
## 1 2.618 2.612
## 2 2.556 2.689
## 3 2.846 2.876
## 4 2.885 2.765
## 5 2.557 2.758
## 6 2.646 2.632
## rh_parsorbitalis_thickness lh_parstriangularis_thickness
## 1 2.932 2.621
## 2 2.473 2.258
## 3 2.738 2.602
## 4 2.994 2.752
## 5 2.885 2.514
## 6 2.822 2.361
## rh_parstriangularis_thickness left_inferior_pfc_average_thickness
## 1 2.680 1.94625
## 2 2.404 1.83675
## 3 2.594 2.08275
## 4 2.726 2.08625
## 5 2.636 1.98000
## 6 2.433 1.90375
## right_inferior_pfc_average_thickness inferior_pfc_average_thickness
## 1 2.743333 2.344792
## 2 2.477667 2.157208
## 3 2.726000 2.404375
## 4 2.868333 2.477292
## 5 2.692667 2.336333
## 6 2.633667 2.268708
## lh_superiorfrontal_thickness rh_superiorfrontal_thickness
## 1 2.825 2.693
## 2 2.825 2.512
## 3 2.825 2.871
## 4 2.825 3.108
## 5 2.825 2.807
## 6 2.825 2.754
## superior_frontal_average_thickness left_pfc_average_thickness
## 1 2.7590 1.93505
## 2 2.6685 1.86755
## 3 2.8480 2.06235
## 4 2.9665 2.02205
## 5 2.8160 1.99560
## 6 2.7895 1.96915
## right_pfc_average_thickness pfc_average_thickness
## 1 2.083667 2.009358
## 2 2.448667 2.158108
## 3 2.717750 2.390050
## 4 2.830833 2.426442
## 5 2.607917 2.301758
## 6 2.617417 2.293283
## lh_inferiorparietal_thickness rh_inferiorparietal_thickness
## 1 2.699 2.677
## 2 2.249 2.423
## 3 2.564 2.602
## 4 2.584 2.621
## 5 2.759 2.683
## 6 2.443 2.498
## inferior_parietal_average_thickness lh_superiorparietal_thickness
## 1 2.6880 2.397
## 2 2.3360 2.042
## 3 2.5830 2.315
## 4 2.6025 2.407
## 5 2.7210 2.346
## 6 2.4705 2.280
## rh_superiorparietal_thickness superior_parietal_average_thickness
## 1 2.291 2.3440
## 2 2.226 2.1340
## 3 2.085 2.2000
## 4 2.293 2.3500
## 5 2.509 2.4275
## 6 2.315 2.2975
## left_ppc_average_thickness right_ppc_average_thickness ppc_average_thickness
## 1 2.5480 2.4840 2.51600
## 2 2.1455 2.3245 2.23500
## 3 2.4395 2.3435 2.39150
## 4 2.4955 2.4570 2.47625
## 5 2.5525 2.5960 2.57425
## 6 2.3615 2.4065 2.38400
## lh_MeanThickness_thickness rh_MeanThickness_thickness mean_thickness_average
## 1 2.56971 2.55457 2.562140
## 2 2.34325 2.40386 2.373555
## 3 2.59484 2.61362 2.604230
## 4 2.67279 2.70045 2.686620
## 5 2.61234 2.57711 2.594725
## 6 2.45544 2.44891 2.452175
## lh_rostralmiddlefrontal_volume rh_rostralmiddlefrontal_volume
## 1 9516 10586
## 2 12786 13033
## 3 15189 15461
## 4 13206 14113
## 5 12949 12628
## 6 13978 13324
## rostral_middle_frontal_total_volume lh_caudalmiddlefrontal_volume
## 1 20102 4300
## 2 25819 5319
## 3 30650 8307
## 4 27319 6711
## 5 25577 4973
## 6 27302 5303
## rh_caudalmiddlefrontal_volume caudal_middle_total_volume
## 1 3776 8076
## 2 5555 10874
## 3 6075 14382
## 4 5632 12343
## 5 4713 9686
## 6 5161 10464
## lh_parsopercularis_volume rh_parsopercularis_volume lh_parsorbitalis_volume
## 1 3419 3000 1527
## 2 4424 3968 2606
## 3 8615 3568 2630
## 4 4279 3617 2299
## 5 3423 3309 2075
## 6 4054 3204 2414
## rh_parsorbitalis_volume lh_parstriangularis_volume rh_parstriangularis_volume
## 1 2310 3163 3935
## 2 2074 2939 3782
## 3 2621 4049 3546
## 4 2693 3740 4111
## 5 2431 3482 4088
## 6 2979 3647 4240
## left_inferior_pfc_total_volume right_inferior_pfc_total_volume
## 1 8109 9245
## 2 9969 9824
## 3 15294 9735
## 4 10318 10421
## 5 8980 9828
## 6 10115 10423
## inferior_pfc_total_volume lh_superiorfrontal_volume rh_superiorfrontal_volume
## 1 17354 18279 15867
## 2 19793 21598 20955
## 3 25029 24464 18694
## 4 20739 22566 23294
## 5 18808 19376 19232
## 6 20538 20825 19395
## superior_frontal_total_volume left_pfc_total_volume right_pfc_total_volume
## 1 34146 40204 39474
## 2 42553 49672 49367
## 3 43158 63254 49965
## 4 45860 52801 53460
## 5 38608 46278 46401
## 6 40220 50221 48303
## lh.rostralanteriorcingulate rh.rostralanteriorcingulate lh.posteriorcingulate
## 1 79678 8280 10421
## 2 99039 12039 12930
## 3 113219 14096 14063
## 4 106261 10021 16784
## 5 92679 11782 12599
## 6 98524 8149 12012
## rh.posteriorcingulate lh.caudalanteriorcingulate rh.caudalanteriorcingulate
## 1 18701 9684 9665
## 2 24969 9826 10409
## 3 28159 10029 13735
## 4 26805 13974 13568
## 5 24381 11031 10449
## 6 20161 11918 10642
## lh.isthmuscingulate rh.isthmuscingulate right_ppc_total_volume
## 1 19349 17964 20086
## 2 20235 21865 23339
## 3 23764 24125 27798
## 4 27542 23995 30352
## 5 21480 22813 23048
## 6 22560 20067 22654
## ppc_total_volume
## 1 38050
## 2 45204
## 3 51923
## 4 54347
## 5 45861
## 6 42721
# Select specific columns
longitudinal_data_time1_specific_columns <- longitudinal_data_time1 %>%
select(subject_id, sub_study, SNQ_Quantity, SNQ_Quality, snq_total, RAVLT, Age, sex, personality, ethnicity, race,
hippocampus_total_volume, amygdala_total_volume, EstimatedTotalIntraCranialVol,
lh.caudalanteriorcingulate, rh.caudalanteriorcingulate,
lh.rostralanteriorcingulate, rh.rostralanteriorcingulate,
lh.posteriorcingulate, rh.posteriorcingulate, lh.isthmuscingulate, rh.isthmuscingulate)
head(longitudinal_data_time1_specific_columns)
## subject_id sub_study SNQ_Quantity SNQ_Quality snq_total RAVLT Age
## 1 1 adultlongitudinal 21 10 53 4 78
## 2 3 adultlongitudinal 44 11 55 1 64
## 3 5 adultlongitudinal 37 7 40 3 67
## 4 7 adultlongitudinal 41 10 38 7 70
## 5 9 adultlongitudinal 38 11 49 3 60
## 6 11 adultlongitudinal 28 11 53 2 44
## sex personality ethnicity
## 1 Female 1 Not Hispanic or Latino or Spanish
## 2 Male 3 Not Hispanic or Latino or Spanish
## 3 Male 3 Not Hispanic or Latino or Spanish
## 4 Male 4 Hispanic or Latino or Spanish
## 5 Female 4 Not Hispanic or Latino or Spanish
## 6 Male 5 Not Hispanic or Latino or Spanish
## race hippocampus_total_volume amygdala_total_volume
## 1 white 7414 3256
## 2 white 7785 2772
## 3 Asian 5143 3261
## 4 white 5467 3504
## 5 white 5107 3576
## 6 Black or African American 7906 2674
## EstimatedTotalIntraCranialVol lh.caudalanteriorcingulate
## 1 1158574 9684
## 2 1479439 9826
## 3 1553358 10029
## 4 1455158 13974
## 5 1321438 11031
## 6 1434374 11918
## rh.caudalanteriorcingulate lh.rostralanteriorcingulate
## 1 9665 79678
## 2 10409 99039
## 3 13735 113219
## 4 13568 106261
## 5 10449 92679
## 6 10642 98524
## rh.rostralanteriorcingulate lh.posteriorcingulate rh.posteriorcingulate
## 1 8280 10421 18701
## 2 12039 12930 24969
## 3 14096 14063 28159
## 4 10021 16784 26805
## 5 11782 12599 24381
## 6 8149 12012 20161
## lh.isthmuscingulate rh.isthmuscingulate
## 1 19349 17964
## 2 20235 21865
## 3 23764 24125
## 4 27542 23995
## 5 21480 22813
## 6 22560 20067
# Create new column for 'total_cingulate_volume'
longitudinal_data_time1_specific_columns$total_cingulate_volume <-
longitudinal_data_time1_specific_columns$lh.caudalanteriorcingulate +
longitudinal_data_time1_specific_columns$rh.caudalanteriorcingulate +
longitudinal_data_time1_specific_columns$lh.rostralanteriorcingulate +
longitudinal_data_time1_specific_columns$rh.rostralanteriorcingulate +
longitudinal_data_time1_specific_columns$lh.posteriorcingulate +
longitudinal_data_time1_specific_columns$rh.posteriorcingulate +
longitudinal_data_time1_specific_columns$lh.isthmuscingulate +
longitudinal_data_time1_specific_columns$rh.isthmuscingulate
# Merge the two data frames
full_data_frame_merged <- rbind(selected_cross_sectional_data, longitudinal_data_time1_specific_columns)
head(full_data_frame_merged)
## subject_id sub_study SNQ_Quality SNQ_Quantity snq_total RAVLT Age sex
## 1 1 Discoverysci 3 40 38 11 84 Female
## 2 2 Discoverysci 9 39 37 4 61 Male
## 3 3 Discoverysci 12 31 35 15 85 Male
## 4 4 Discoverysci 9 29 47 10 58 Female
## 5 5 Discoverysci 3 36 51 3 23 Male
## 6 6 Discoverysci 9 41 37 11 66 Male
## personality ethnicity race
## 1 1 Not Hispanic or Latino or Spanish white
## 2 2 Not Hispanic or Latino or Spanish Black or African American
## 3 3 Not Hispanic or Latino or Spanish white
## 4 2 Not Hispanic or Latino or Spanish white
## 5 3 Not Hispanic or Latino or Spanish Asian
## 6 2 Not Hispanic or Latino or Spanish white
## hippocampus_total_volume amygdala_total_volume EstimatedTotalIntraCranialVol
## 1 7483 2383 1158574
## 2 6708 2529 1374227
## 3 8318 2338 1479439
## 4 7763 2126 1514698
## 5 7106 2576 1553358
## 6 8536 3324 1410264
## lh.caudalanteriorcingulate rh.caudalanteriorcingulate
## 1 9665 19349
## 2 11474 21286
## 3 10409 20235
## 4 12520 24430
## 5 13735 23764
## 6 9813 21178
## lh.rostralanteriorcingulate rh.rostralanteriorcingulate lh.posteriorcingulate
## 1 8280 10421 18701
## 2 11336 13624 24960
## 3 12039 12930 24969
## 4 12213 14344 26557
## 5 14096 14063 28159
## 6 9078 10673 19751
## rh.posteriorcingulate lh.isthmuscingulate rh.isthmuscingulate
## 1 9684 17964 20086
## 2 9812 21148 25098
## 3 9826 21865 23339
## 4 11910 24123 26864
## 5 10029 24125 27798
## 6 11365 20443 20486
## total_cingulate_volume
## 1 114150
## 2 138738
## 3 135612
## 4 152961
## 5 155769
## 6 122787
# ---- filter data to keep >54 ----
# Filter for Age > 54
full_data_frame_merged_over_55 <- full_data_frame_merged %>%
filter(Age > 54)
head(full_data_frame_merged_over_55)
## subject_id sub_study SNQ_Quality SNQ_Quantity snq_total RAVLT Age sex
## 1 1 Discoverysci 3 40 38 11 84 Female
## 2 2 Discoverysci 9 39 37 4 61 Male
## 3 3 Discoverysci 12 31 35 15 85 Male
## 4 4 Discoverysci 9 29 47 10 58 Female
## 5 6 Discoverysci 9 41 37 11 66 Male
## 6 7 Discoverysci 12 28 40 15 84 Male
## personality ethnicity race
## 1 1 Not Hispanic or Latino or Spanish white
## 2 2 Not Hispanic or Latino or Spanish Black or African American
## 3 3 Not Hispanic or Latino or Spanish white
## 4 2 Not Hispanic or Latino or Spanish white
## 5 2 Not Hispanic or Latino or Spanish white
## 6 3 Hispanic or Latino or Spanish white
## hippocampus_total_volume amygdala_total_volume EstimatedTotalIntraCranialVol
## 1 7483 2383 1158574
## 2 6708 2529 1374227
## 3 8318 2338 1479439
## 4 7763 2126 1514698
## 5 8536 3324 1410264
## 6 8844 2505 1455158
## lh.caudalanteriorcingulate rh.caudalanteriorcingulate
## 1 9665 19349
## 2 11474 21286
## 3 10409 20235
## 4 12520 24430
## 5 9813 21178
## 6 13568 27542
## lh.rostralanteriorcingulate rh.rostralanteriorcingulate lh.posteriorcingulate
## 1 8280 10421 18701
## 2 11336 13624 24960
## 3 12039 12930 24969
## 4 12213 14344 26557
## 5 9078 10673 19751
## 6 10021 16784 26805
## rh.posteriorcingulate lh.isthmuscingulate rh.isthmuscingulate
## 1 9684 17964 20086
## 2 9812 21148 25098
## 3 9826 21865 23339
## 4 11910 24123 26864
## 5 11365 20443 20486
## 6 13974 23995 30352
## total_cingulate_volume
## 1 114150
## 2 138738
## 3 135612
## 4 152961
## 5 122787
## 6 163041
# Correct for ICV
full_data_frame_merged_over_55_corrected_for_icv <- full_data_frame_merged_over_55 %>%
mutate(
hippocampus_total_volume_corrected = hippocampus_total_volume / EstimatedTotalIntraCranialVol,
amygdala_total_volume_corrected = amygdala_total_volume / EstimatedTotalIntraCranialVol,
total_cingulate_volume_corrected = total_cingulate_volume / EstimatedTotalIntraCranialVol
)
head(full_data_frame_merged_over_55_corrected_for_icv)
## subject_id sub_study SNQ_Quality SNQ_Quantity snq_total RAVLT Age sex
## 1 1 Discoverysci 3 40 38 11 84 Female
## 2 2 Discoverysci 9 39 37 4 61 Male
## 3 3 Discoverysci 12 31 35 15 85 Male
## 4 4 Discoverysci 9 29 47 10 58 Female
## 5 6 Discoverysci 9 41 37 11 66 Male
## 6 7 Discoverysci 12 28 40 15 84 Male
## personality ethnicity race
## 1 1 Not Hispanic or Latino or Spanish white
## 2 2 Not Hispanic or Latino or Spanish Black or African American
## 3 3 Not Hispanic or Latino or Spanish white
## 4 2 Not Hispanic or Latino or Spanish white
## 5 2 Not Hispanic or Latino or Spanish white
## 6 3 Hispanic or Latino or Spanish white
## hippocampus_total_volume amygdala_total_volume EstimatedTotalIntraCranialVol
## 1 7483 2383 1158574
## 2 6708 2529 1374227
## 3 8318 2338 1479439
## 4 7763 2126 1514698
## 5 8536 3324 1410264
## 6 8844 2505 1455158
## lh.caudalanteriorcingulate rh.caudalanteriorcingulate
## 1 9665 19349
## 2 11474 21286
## 3 10409 20235
## 4 12520 24430
## 5 9813 21178
## 6 13568 27542
## lh.rostralanteriorcingulate rh.rostralanteriorcingulate lh.posteriorcingulate
## 1 8280 10421 18701
## 2 11336 13624 24960
## 3 12039 12930 24969
## 4 12213 14344 26557
## 5 9078 10673 19751
## 6 10021 16784 26805
## rh.posteriorcingulate lh.isthmuscingulate rh.isthmuscingulate
## 1 9684 17964 20086
## 2 9812 21148 25098
## 3 9826 21865 23339
## 4 11910 24123 26864
## 5 11365 20443 20486
## 6 13974 23995 30352
## total_cingulate_volume hippocampus_total_volume_corrected
## 1 114150 0.006458799
## 2 138738 0.004881289
## 3 135612 0.005622402
## 4 152961 0.005125116
## 5 122787 0.006052769
## 6 163041 0.006077691
## amygdala_total_volume_corrected total_cingulate_volume_corrected
## 1 0.002056838 0.09852625
## 2 0.001840307 0.10095710
## 3 0.001580329 0.09166449
## 4 0.001403581 0.10098452
## 5 0.002357006 0.08706670
## 6 0.001721463 0.11204351
# ---- begin running correlations ----
alpha_corrected <- 0.05 / 14
print(paste("Bonferroni-corrected alpha level:", alpha_corrected))
## [1] "Bonferroni-corrected alpha level: 0.00357142857142857"
# Correlation 1: Age vs RAVLT
cor_test_1 <- cor.test(full_data_frame_merged_over_55_corrected_for_icv$Age,
full_data_frame_merged_over_55_corrected_for_icv$RAVLT)
print(cor_test_1)
##
## Pearson's product-moment correlation
##
## data: full_data_frame_merged_over_55_corrected_for_icv$Age and full_data_frame_merged_over_55_corrected_for_icv$RAVLT
## t = 1.1235, df = 253, p-value = 0.2623
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.05284149 0.19164227
## sample estimates:
## cor
## 0.07045843
# Plot for Age and RAVLT
ggplot(full_data_frame_merged_over_55_corrected_for_icv, aes(x = Age, y = RAVLT)) +
geom_point() +
geom_smooth(method = "lm") +
labs(title = "Age vs RAVLT", x = "Age", y = "Episodic Memory Performance (RAVLT)") +
geom_hline(yintercept = alpha_corrected, linetype = "dashed", color = "red")
## `geom_smooth()` using formula 'y ~ x'
## Warning: Removed 2 rows containing non-finite values (stat_smooth).
## Warning: Removed 2 rows containing missing values (geom_point).
Correlations Between Age, Memory, and Social Connection
Quality/Quantity
### Correlation for Age and Memory
# ---- correlations age and memory ----
# Correlation 1: Age negatively associated with episodic memory performance (RAVLT)
cor_test_1 <- cor.test(full_data_frame_merged_over_55_corrected_for_icv$Age,
full_data_frame_merged_over_55_corrected_for_icv$RAVLT)
p_value_1 <- cor_test_1$p.value
sig_1 <- ifelse(p_value_1 < alpha_corrected, "Significant", "Not Significant")
print(paste("Correlation 1: Age vs RAVLT - p-value:", p_value_1, "-", sig_1))
## [1] "Correlation 1: Age vs RAVLT - p-value: 0.262289216389112 - Not Significant"
print(paste("Correlation coefficient:", cor_test_1$estimate))
## [1] "Correlation coefficient: 0.0704584305118598"
# Plot for Age and RAVLT
ggplot(full_data_frame_merged_over_55_corrected_for_icv, aes(x = Age, y = RAVLT)) +
geom_point() +
geom_smooth(method = "lm") +
labs(title = "Age vs RAVLT", x = "Age", y = "Episodic Memory Performance (RAVLT)") +
geom_hline(yintercept = alpha_corrected, linetype = "dashed", color = "red")
## `geom_smooth()` using formula 'y ~ x'
## Warning: Removed 2 rows containing non-finite values (stat_smooth).
## Warning: Removed 2 rows containing missing values (geom_point).
# ---- correlation age and snq ----
# Correlation 2: Age positively associated with social connection quality (SNQ_Quality)
cor_test_2 <- cor.test(full_data_frame_merged_over_55_corrected_for_icv$Age,
full_data_frame_merged_over_55_corrected_for_icv$SNQ_Quality)
p_value_2 <- cor_test_2$p.value
sig_2 <- ifelse(p_value_2 < alpha_corrected, "Significant", "Not Significant")
print(paste("Correlation 2: Age vs SNQ_Quality - p-value:", p_value_2, "-", sig_2))
## [1] "Correlation 2: Age vs SNQ_Quality - p-value: 0.316914826833802 - Not Significant"
print(paste("Correlation coefficient:", cor_test_2$estimate))
## [1] "Correlation coefficient: -0.0629206607749498"
# Plot for Age and Social Connection Quality
ggplot(full_data_frame_merged_over_55_corrected_for_icv, aes(x = Age, y = SNQ_Quality)) +
geom_point() +
geom_smooth(method = "lm") +
labs(title = "Age vs Social Connection Quality", x = "Age", y = "Social Connection Quality (SNQ_Quality)") +
geom_hline(yintercept = alpha_corrected, linetype = "dashed", color = "red")
## `geom_smooth()` using formula 'y ~ x'
## Warning: Removed 2 rows containing non-finite values (stat_smooth).
## Warning: Removed 2 rows containing missing values (geom_point).
# Correlation 3: Age negatively associated with social connection quantity (SNQ_Quantity)
cor_test_3 <- cor.test(full_data_frame_merged_over_55_corrected_for_icv$Age,
full_data_frame_merged_over_55_corrected_for_icv$SNQ_Quantity)
p_value_3 <- cor_test_3$p.value
sig_3 <- ifelse(p_value_3 < alpha_corrected, "Significant", "Not Significant")
print(paste("Correlation 3: Age vs SNQ_Quantity - p-value:", p_value_3, "-", sig_3))
## [1] "Correlation 3: Age vs SNQ_Quantity - p-value: 0.0577298477258219 - Not Significant"
print(paste("Correlation coefficient:", cor_test_3$estimate))
## [1] "Correlation coefficient: -0.119003351004473"
# Plot for Age and Social Connection Quantity
ggplot(full_data_frame_merged_over_55_corrected_for_icv, aes(x = Age, y = SNQ_Quantity)) +
geom_point() +
geom_smooth(method = "lm") +
labs(title = "Age vs Social Connection Quantity", x = "Age", y = "Social Connection Quantity (SNQ_Quantity)") +
geom_hline(yintercept = alpha_corrected, linetype = "dashed", color = "red")
## `geom_smooth()` using formula 'y ~ x'
## Warning: Removed 2 rows containing non-finite values (stat_smooth).
## Warning: Removed 2 rows containing missing values (geom_point).
# ---- correlations ravlt and snq ----
# Correlation 4: RAVLT positively associated with social connection quality (SNQ_Quality)
cor_test_4 <- cor.test(full_data_frame_merged_over_55_corrected_for_icv$RAVLT,
full_data_frame_merged_over_55_corrected_for_icv$SNQ_Quality)
p_value_4 <- cor_test_4$p.value
sig_4 <- ifelse(p_value_4 < alpha_corrected, "Significant", "Not Significant")
print(paste("Correlation 4: RAVLT vs SNQ_Quality - p-value:", p_value_4, "-", sig_4))
## [1] "Correlation 4: RAVLT vs SNQ_Quality - p-value: 0.577351810707077 - Not Significant"
print(paste("Correlation coefficient:", cor_test_4$estimate))
## [1] "Correlation coefficient: -0.0350583039623054"
# Plot for RAVLT and Social Connection Quality
ggplot(full_data_frame_merged_over_55_corrected_for_icv, aes(x = SNQ_Quality, y = RAVLT)) +
geom_point() +
geom_smooth(method = "lm") +
labs(title = "Social Connection Quality vs RAVLT", x = "Social Connection Quality", y = "Episodic Memory Performance (RAVLT)") +
geom_hline(yintercept = alpha_corrected, linetype = "dashed", color = "red")
## `geom_smooth()` using formula 'y ~ x'
## Warning: Removed 2 rows containing non-finite values (stat_smooth).
## Warning: Removed 2 rows containing missing values (geom_point).
# Correlation 5: RAVLT positively associated with social connection quantity (SNQ_Quantity)
cor_test_5 <- cor.test(full_data_frame_merged_over_55_corrected_for_icv$RAVLT,
full_data_frame_merged_over_55_corrected_for_icv$SNQ_Quantity)
p_value_5 <- cor_test_5$p.value
sig_5 <- ifelse(p_value_5 < alpha_corrected, "Significant", "Not Significant")
print(paste("Correlation 5: RAVLT vs SNQ_Quantity - p-value:", p_value_5, "-", sig_5))
## [1] "Correlation 5: RAVLT vs SNQ_Quantity - p-value: 0.212570927903375 - Not Significant"
print(paste("Correlation coefficient:", cor_test_5$estimate))
## [1] "Correlation coefficient: 0.0783253981252475"
# Plot for RAVLT and Social Connection Quantity
ggplot(full_data_frame_merged_over_55_corrected_for_icv, aes(x = SNQ_Quantity, y = RAVLT)) +
geom_point() +
geom_smooth(method = "lm") +
labs(title = "Social Connection Quantity vs RAVLT", x = "Social Connection Quantity", y = "Episodic Memory Performance (RAVLT)") +
geom_hline(yintercept = alpha_corrected, linetype = "dashed", color = "red")
## `geom_smooth()` using formula 'y ~ x'
## Warning: Removed 2 rows containing non-finite values (stat_smooth).
## Warning: Removed 2 rows containing missing values (geom_point).
Correlations Between Age and ROI Corrected Volumes
### Correlation Between Age and Hippocampus Total Volume Corrected
# ---- correlations age and roi ----
# Correlation 1: Age vs Hippocampus Total Volume Corrected
cor_test_1 <- cor.test(full_data_frame_merged_over_55_corrected_for_icv$Age,
full_data_frame_merged_over_55_corrected_for_icv$hippocampus_total_volume_corrected)
p_value_1 <- cor_test_1$p.value
sig_1 <- ifelse(p_value_1 < alpha_corrected, "Significant", "Not Significant")
print(paste("Correlation 1: Age vs Hippocampus Total Volume Corrected - p-value:", p_value_1, "-", sig_1))
## [1] "Correlation 1: Age vs Hippocampus Total Volume Corrected - p-value: 0.766031338640573 - Not Significant"
print(paste("Correlation coefficient:", cor_test_1$estimate))
## [1] "Correlation coefficient: 0.0187249216776731"
ggplot(full_data_frame_merged_over_55_corrected_for_icv, aes(x = Age, y = hippocampus_total_volume_corrected)) +
geom_point() +
geom_smooth(method = "lm") +
labs(title = "Age vs Hippocampus Total Volume Corrected", x = "Age", y = "Hippocampus Total Volume Corrected") +
geom_hline(yintercept = alpha_corrected, linetype = "dashed", color = "red")
## `geom_smooth()` using formula 'y ~ x'
## Warning: Removed 2 rows containing non-finite values (stat_smooth).
## Warning: Removed 2 rows containing missing values (geom_point).
# Correlation 2: Age vs Amygdala Total Volume Corrected
cor_test_2 <- cor.test(full_data_frame_merged_over_55_corrected_for_icv$Age,
full_data_frame_merged_over_55_corrected_for_icv$amygdala_total_volume_corrected)
p_value_2 <- cor_test_2$p.value
sig_2 <- ifelse(p_value_2 < alpha_corrected, "Significant", "Not Significant")
print(paste("Correlation 2: Age vs Amygdala Total Volume Corrected - p-value:", p_value_2, "-", sig_2))
## [1] "Correlation 2: Age vs Amygdala Total Volume Corrected - p-value: 0.872157736529027 - Not Significant"
print(paste("Correlation coefficient:", cor_test_2$estimate))
## [1] "Correlation coefficient: 0.0101265899081824"
ggplot(full_data_frame_merged_over_55_corrected_for_icv, aes(x = Age, y = amygdala_total_volume_corrected)) +
geom_point() +
geom_smooth(method = "lm") +
labs(title = "Age vs Amygdala Total Volume Corrected", x = "Age", y = "Amygdala Total Volume Corrected") +
geom_hline(yintercept = alpha_corrected, linetype = "dashed", color = "red")
## `geom_smooth()` using formula 'y ~ x'
## Warning: Removed 2 rows containing non-finite values (stat_smooth).
## Warning: Removed 2 rows containing missing values (geom_point).
# Correlation 3: Age vs Total Cingulate Volume Corrected
cor_test_3 <- cor.test(full_data_frame_merged_over_55_corrected_for_icv$Age,
full_data_frame_merged_over_55_corrected_for_icv$total_cingulate_volume_corrected)
p_value_3 <- cor_test_3$p.value
sig_3 <- ifelse(p_value_3 < alpha_corrected, "Significant", "Not Significant")
print(paste("Correlation 3: Age vs Total Cingulate Volume Corrected - p-value:", p_value_3, "-", sig_3))
## [1] "Correlation 3: Age vs Total Cingulate Volume Corrected - p-value: 0.247871336182177 - Not Significant"
print(paste("Correlation coefficient:", cor_test_3$estimate))
## [1] "Correlation coefficient: -0.0726236318387365"
ggplot(full_data_frame_merged_over_55_corrected_for_icv, aes(x = Age, y = total_cingulate_volume_corrected)) +
geom_point() +
geom_smooth(method = "lm") +
labs(title = "Age vs Total Cingulate Volume Corrected", x = "Age", y = "Total Cingulate Volume Corrected") +
geom_hline(yintercept = alpha_corrected, linetype = "dashed", color = "red")
## `geom_smooth()` using formula 'y ~ x'
## Warning: Removed 2 rows containing non-finite values (stat_smooth).
## Warning: Removed 2 rows containing missing values (geom_point).
# ---- correlations roi and snq quality ----
# Correlation 4: Hippocampus Total Volume Corrected vs Social Connection Quality
cor_test_4 <- cor.test(full_data_frame_merged_over_55_corrected_for_icv$hippocampus_total_volume_corrected,
full_data_frame_merged_over_55_corrected_for_icv$SNQ_Quality)
p_value_4 <- cor_test_4$p.value
sig_4 <- ifelse(p_value_4 < alpha_corrected, "Significant", "Not Significant")
print(paste("Correlation 4: Hippocampus Total Volume Corrected vs SNQ_Quality - p-value:", p_value_4, "-", sig_4))
## [1] "Correlation 4: Hippocampus Total Volume Corrected vs SNQ_Quality - p-value: 0.280902553853503 - Not Significant"
print(paste("Correlation coefficient:", cor_test_4$estimate))
## [1] "Correlation coefficient: 0.0677807165227571"
ggplot(full_data_frame_merged_over_55_corrected_for_icv, aes(x = SNQ_Quality, y = hippocampus_total_volume_corrected)) +
geom_point() +
geom_smooth(method = "lm") +
labs(title = "Social Connection Quality vs Hippocampus Total Volume Corrected", x = "Social Connection Quality", y = "Hippocampus Total Volume Corrected") +
geom_hline(yintercept = alpha_corrected, linetype = "dashed", color = "red")
## `geom_smooth()` using formula 'y ~ x'
## Warning: Removed 2 rows containing non-finite values (stat_smooth).
## Warning: Removed 2 rows containing missing values (geom_point).
# Correlation 5: Amygdala Total Volume Corrected vs Social Connection Quality
cor_test_5 <- cor.test(full_data_frame_merged_over_55_corrected_for_icv$amygdala_total_volume_corrected,
full_data_frame_merged_over_55_corrected_for_icv$SNQ_Quality)
p_value_5 <- cor_test_5$p.value
sig_5 <- ifelse(p_value_5 < alpha_corrected, "Significant", "Not Significant")
print(paste("Correlation 5: Amygdala Total Volume Corrected vs SNQ_Quality - p-value:", p_value_5, "-", sig_5))
## [1] "Correlation 5: Amygdala Total Volume Corrected vs SNQ_Quality - p-value: 0.134765722696597 - Not Significant"
print(paste("Correlation coefficient:", cor_test_5$estimate))
## [1] "Correlation coefficient: 0.0939105248853412"
ggplot(full_data_frame_merged_over_55_corrected_for_icv, aes(x = SNQ_Quality, y = amygdala_total_volume_corrected)) +
geom_point() +
geom_smooth(method = "lm") +
labs(title = "Social Connection Quality vs Amygdala Total Volume Corrected", x = "Social Connection Quality", y = "Amygdala Total Volume Corrected") +
geom_hline(yintercept = alpha_corrected, linetype = "dashed", color = "red")
## `geom_smooth()` using formula 'y ~ x'
## Warning: Removed 2 rows containing non-finite values (stat_smooth).
## Warning: Removed 2 rows containing missing values (geom_point).
# Correlation 6: Total Cingulate Volume Corrected vs Social Connection Quality
cor_test_6 <- cor.test(full_data_frame_merged_over_55_corrected_for_icv$total_cingulate_volume_corrected,
full_data_frame_merged_over_55_corrected_for_icv$SNQ_Quality)
p_value_6 <- cor_test_6$p.value
sig_6 <- ifelse(p_value_6 < alpha_corrected, "Significant", "Not Significant")
print(paste("Correlation 6: Total Cingulate Volume Corrected vs SNQ_Quality - p-value:", p_value_6, "-", sig_6))
## [1] "Correlation 6: Total Cingulate Volume Corrected vs SNQ_Quality - p-value: 7.10618050219819e-08 - Significant"
print(paste("Correlation coefficient:", cor_test_6$estimate))
## [1] "Correlation coefficient: 0.329582862168436"
ggplot(full_data_frame_merged_over_55_corrected_for_icv, aes(x = SNQ_Quality, y = total_cingulate_volume_corrected)) +
geom_point() +
geom_smooth(method = "lm") +
labs(title = "Social Connection Quality vs Total Cingulate Volume Corrected", x = "Social Connection Quality", y = "Total Cingulate Volume Corrected") +
geom_hline(yintercept = alpha_corrected, linetype = "dashed", color = "red")
## `geom_smooth()` using formula 'y ~ x'
## Warning: Removed 2 rows containing non-finite values (stat_smooth).
## Warning: Removed 2 rows containing missing values (geom_point).
Correlations Between ROI Corrected Volumes and Social Connection
Quantity
### Correlation Between Hippocampus Total Volume Corrected and Social Connection Quantity
# ---- correlations roi and snq quantity ----
# Correlation 7: Hippocampus Total Volume Corrected vs Social Connection Quantity
cor_test_7 <- cor.test(full_data_frame_merged_over_55_corrected_for_icv$hippocampus_total_volume_corrected,
full_data_frame_merged_over_55_corrected_for_icv$SNQ_Quantity)
p_value_7 <- cor_test_7$p.value
sig_7 <- ifelse(p_value_7 < alpha_corrected, "Significant", "Not Significant")
print(paste("Correlation 7: Hippocampus Total Volume Corrected vs SNQ_Quantity - p-value:", p_value_7, "-", sig_7))
## [1] "Correlation 7: Hippocampus Total Volume Corrected vs SNQ_Quantity - p-value: 0.631916339872348 - Not Significant"
print(paste("Correlation coefficient:", cor_test_7$estimate))
## [1] "Correlation coefficient: 0.0301393655582844"
ggplot(full_data_frame_merged_over_55_corrected_for_icv, aes(x = SNQ_Quantity, y = hippocampus_total_volume_corrected)) +
geom_point() +
geom_smooth(method = "lm") +
labs(title = "Social Connection Quantity vs Hippocampus Total Volume Corrected", x = "Social Connection Quantity", y = "Hippocampus Total Volume Corrected") +
geom_hline(yintercept = alpha_corrected, linetype = "dashed", color = "red")
## `geom_smooth()` using formula 'y ~ x'
## Warning: Removed 2 rows containing non-finite values (stat_smooth).
## Warning: Removed 2 rows containing missing values (geom_point).
# Correlation 8: Amygdala Total Volume Corrected vs Social Connection Quantity
cor_test_8 <- cor.test(full_data_frame_merged_over_55_corrected_for_icv$amygdala_total_volume_corrected,
full_data_frame_merged_over_55_corrected_for_icv$SNQ_Quantity)
p_value_8 <- cor_test_8$p.value
sig_8 <- ifelse(p_value_8 < alpha_corrected, "Significant", "Not Significant")
print(paste("Correlation 8: Amygdala Total Volume Corrected vs SNQ_Quantity - p-value:", p_value_8, "-", sig_8))
## [1] "Correlation 8: Amygdala Total Volume Corrected vs SNQ_Quantity - p-value: 0.786258896578505 - Not Significant"
print(paste("Correlation coefficient:", cor_test_8$estimate))
## [1] "Correlation coefficient: 0.0170640371229014"
ggplot(full_data_frame_merged_over_55_corrected_for_icv, aes(x = SNQ_Quantity, y = amygdala_total_volume_corrected)) +
geom_point() +
geom_smooth(method = "lm") +
labs(title = "Social Connection Quantity vs Amygdala Total Volume Corrected", x = "Social Connection Quantity", y = "Amygdala Total Volume Corrected") +
geom_hline(yintercept = alpha_corrected, linetype = "dashed", color = "red")
## `geom_smooth()` using formula 'y ~ x'
## Warning: Removed 2 rows containing non-finite values (stat_smooth).
## Warning: Removed 2 rows containing missing values (geom_point).
# Correlation 9: Total Cingulate Volume Corrected vs Social Connection Quantity
cor_test_9 <- cor.test(full_data_frame_merged_over_55_corrected_for_icv$total_cingulate_volume_corrected,
full_data_frame_merged_over_55_corrected_for_icv$SNQ_Quantity)
p_value_9 <- cor_test_9$p.value
sig_9 <- ifelse(p_value_9 < alpha_corrected, "Significant", "Not Significant")
print(paste("Correlation 9: Total Cingulate Volume Corrected vs SNQ_Quantity - p-value:", p_value_9, "-", sig_9))
## [1] "Correlation 9: Total Cingulate Volume Corrected vs SNQ_Quantity - p-value: 0.602219854888563 - Not Significant"
print(paste("Correlation coefficient:", cor_test_9$estimate))
## [1] "Correlation coefficient: -0.0327918214248807"
ggplot(full_data_frame_merged_over_55_corrected_for_icv, aes(x = SNQ_Quantity, y = total_cingulate_volume_corrected)) +
geom_point() +
geom_smooth(method = "lm") +
labs(title = "Social Connection Quantity vs Total Cingulate Volume Corrected", x = "Social Connection Quantity", y = "Total Cingulate Volume Corrected") +
geom_hline(yintercept = alpha_corrected, linetype = "dashed", color = "red")
## `geom_smooth()` using formula 'y ~ x'
## Warning: Removed 2 rows containing non-finite values (stat_smooth).
## Warning: Removed 2 rows containing missing values (geom_point).
Linear Mixed Models for RAVLT and SNQ
### Linear Mixed Models for Age, RAVLT, and SNQ
# ---- linear mixed models ravlt and snq ----
# Model 1: Interaction between Social Connection Quality and Quantity Predicting Episodic Memory Performance
# Add covariates Sex + Personality
model_1 <- lmer(RAVLT ~ SNQ_Quality * SNQ_Quantity + Age + sex + personality +
(1 | subject_id) + (1 | sub_study), data = full_data_frame_merged_over_55_corrected_for_icv)
## boundary (singular) fit: see help('isSingular')
# Display the summary of Model 1
summary(model_1)
## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: RAVLT ~ SNQ_Quality * SNQ_Quantity + Age + sex + personality +
## (1 | subject_id) + (1 | sub_study)
## Data: full_data_frame_merged_over_55_corrected_for_icv
##
## REML criterion at convergence: 1497
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.00317 -0.81597 -0.00711 0.76243 1.93361
##
## Random effects:
## Groups Name Variance Std.Dev.
## subject_id (Intercept) 0.0000 0.000
## sub_study (Intercept) 0.1467 0.383
## Residual 19.5702 4.424
## Number of obs: 255, groups: subject_id, 202; sub_study, 3
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.973013 4.070500 247.926537 0.239 0.811
## SNQ_Quality 0.286852 0.412374 247.995042 0.696 0.487
## SNQ_Quantity 0.120122 0.093989 247.616984 1.278 0.202
## Age 0.039252 0.031849 247.622401 1.232 0.219
## sexMale 0.498687 0.585491 247.938817 0.852 0.395
## personality -0.130724 0.199014 247.102297 -0.657 0.512
## SNQ_Quality:SNQ_Quantity -0.009348 0.012148 247.412386 -0.770 0.442
##
## Correlation of Fixed Effects:
## (Intr) SNQ_Ql SNQ_Qn Age sexMal prsnlt
## SNQ_Quality -0.735
## SNQ_Quantty -0.805 0.889
## Age -0.598 0.032 0.069
## sexMale -0.020 -0.044 -0.028 -0.022
## personality -0.156 0.050 0.047 -0.038 0.005
## SNQ_Q:SNQ_Q 0.714 -0.977 -0.912 -0.022 0.049 -0.057
## optimizer (nloptwrap) convergence code: 0 (OK)
## boundary (singular) fit: see help('isSingular')
# Model 2: Age Moderating the Relationship between Social Connection Quality and Quantity
# Including three-way interaction: Age * Social Connection Quality * Social Connection Quantity
# Add covariates Sex + Personality
model_2 <- lmer(RAVLT ~ Age * SNQ_Quality * SNQ_Quantity + sex + personality +
(1 | subject_id) + (1 | sub_study), data = full_data_frame_merged_over_55_corrected_for_icv)
## Warning: Some predictor variables are on very different scales: consider
## rescaling
## boundary (singular) fit: see help('isSingular')
## Warning: Some predictor variables are on very different scales: consider
## rescaling
# Display the summary of Model 2
summary(model_2)
## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: RAVLT ~ Age * SNQ_Quality * SNQ_Quantity + sex + personality +
## (1 | subject_id) + (1 | sub_study)
## Data: full_data_frame_merged_over_55_corrected_for_icv
##
## REML criterion at convergence: 1524.3
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.96508 -0.82432 -0.06898 0.77285 1.92284
##
## Random effects:
## Groups Name Variance Std.Dev.
## subject_id (Intercept) 0.000 0.0000
## sub_study (Intercept) 0.142 0.3768
## Residual 19.771 4.4464
## Number of obs: 255, groups: subject_id, 202; sub_study, 3
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 1.013e+01 2.505e+01 2.436e+02 0.405 0.686
## Age -9.164e-02 3.546e-01 2.435e+02 -0.258 0.796
## SNQ_Quality -1.663e+00 3.337e+00 2.436e+02 -0.498 0.619
## SNQ_Quantity -1.435e-01 7.353e-01 2.435e+02 -0.195 0.845
## sexMale 4.922e-01 5.905e-01 2.450e+02 0.834 0.405
## personality -1.350e-01 2.004e-01 2.442e+02 -0.674 0.501
## Age:SNQ_Quality 2.809e-02 4.746e-02 2.438e+02 0.592 0.554
## Age:SNQ_Quantity 3.780e-03 1.046e-02 2.435e+02 0.361 0.718
## SNQ_Quality:SNQ_Quantity 4.757e-02 9.811e-02 2.437e+02 0.485 0.628
## Age:SNQ_Quality:SNQ_Quantity -8.213e-04 1.401e-03 2.439e+02 -0.586 0.558
##
## Correlation of Fixed Effects:
## (Intr) Age SNQ_Ql SNQ_Qn sexMal prsnlt Ag:SNQ_Ql Ag:SNQ_Qn
## Age -0.991
## SNQ_Quality -0.907 0.901
## SNQ_Quantty -0.978 0.973 0.889
## sexMale 0.034 -0.040 -0.021 -0.052
## personality 0.002 -0.031 0.002 -0.018 0.007
## Ag:SNQ_Qlty 0.896 -0.904 -0.992 -0.881 0.016 0.004
## Ag:SNQ_Qntt 0.967 -0.978 -0.881 -0.992 0.049 0.024 0.887
## SNQ_Q:SNQ_Q 0.887 -0.883 -0.980 -0.907 0.033 -0.007 0.974 0.901
## A:SNQ_Q:SNQ -0.876 0.886 0.971 0.898 -0.027 0.001 -0.980 -0.905
## SNQ_Q:
## Age
## SNQ_Quality
## SNQ_Quantty
## sexMale
## personality
## Ag:SNQ_Qlty
## Ag:SNQ_Qntt
## SNQ_Q:SNQ_Q
## A:SNQ_Q:SNQ -0.992
## fit warnings:
## Some predictor variables are on very different scales: consider rescaling
## optimizer (nloptwrap) convergence code: 0 (OK)
## boundary (singular) fit: see help('isSingular')
# Model comparison using ANOVA to test if Model 2 significantly improves fit over Model 1
anova_comparison <- anova(model_1, model_2)
## refitting model(s) with ML (instead of REML)
# Display the ANOVA comparison between Model 1 and Model 2
print(anova_comparison)
## Data: full_data_frame_merged_over_55_corrected_for_icv
## Models:
## model_1: RAVLT ~ SNQ_Quality * SNQ_Quantity + Age + sex + personality + (1 | subject_id) + (1 | sub_study)
## model_2: RAVLT ~ Age * SNQ_Quality * SNQ_Quantity + sex + personality + (1 | subject_id) + (1 | sub_study)
## npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
## model_1 10 1495.9 1531.3 -737.97 1475.9
## model_2 13 1501.4 1547.4 -737.70 1475.4 0.529 3 0.9125
# Extract and display the p-values for fixed effects for both models
# Comment out model that does not win
anova(model_1) # Get significance of fixed effects for Model 1
## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## SNQ_Quality 9.470 9.470 1 248.00 0.4839 0.4873
## SNQ_Quantity 31.966 31.966 1 247.62 1.6334 0.2024
## Age 29.726 29.726 1 247.62 1.5190 0.2189
## sex 14.197 14.197 1 247.94 0.7255 0.3952
## personality 8.444 8.444 1 247.10 0.4315 0.5119
## SNQ_Quality:SNQ_Quantity 11.589 11.589 1 247.41 0.5922 0.4423
anova(model_2) # Get significance of fixed effects for Model 2
## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Age 1.3206 1.3206 1 243.48 0.0668 0.7963
## SNQ_Quality 4.9112 4.9112 1 243.61 0.2484 0.6187
## SNQ_Quantity 0.7530 0.7530 1 243.48 0.0381 0.8454
## sex 13.7368 13.7368 1 244.96 0.6948 0.4053
## personality 8.9776 8.9776 1 244.24 0.4541 0.5010
## Age:SNQ_Quality 6.9267 6.9267 1 243.79 0.3503 0.5545
## Age:SNQ_Quantity 2.5817 2.5817 1 243.48 0.1306 0.7181
## SNQ_Quality:SNQ_Quantity 4.6472 4.6472 1 243.72 0.2351 0.6282
## Age:SNQ_Quality:SNQ_Quantity 6.7965 6.7965 1 243.85 0.3438 0.5582
# Model 1: Interaction plot for Social Connection Quality * Quantity
ggplot(full_data_frame_merged_over_55_corrected_for_icv, aes(x = SNQ_Quantity, y = RAVLT, color = SNQ_Quality)) +
geom_point() +
geom_smooth(method = "lm", formula = y ~ x) +
labs(title = "Interaction: Social Connection Quality and Quantity on Episodic Memory",
x = "Social Connection Quantity", y = "Episodic Memory Performance (RAVLT)")
## Warning: Removed 2 rows containing non-finite values (stat_smooth).
## Warning: Removed 2 rows containing missing values (geom_point).
# Model 2: Interaction plot for Age * Social Connection Quality * Quantity
ggplot(full_data_frame_merged_over_55_corrected_for_icv, aes(x = SNQ_Quantity, y = RAVLT, color = as.factor(Age))) +
facet_wrap(~ SNQ_Quality) +
geom_point() +
geom_smooth(method = "lm", formula = y ~ x) +
labs(title = "Age Moderating the Effect of Social Connection Quality and Quantity on Episodic Memory",
x = "Social Connection Quantity", y = "Episodic Memory Performance (RAVLT)")
## Warning: Removed 2 rows containing non-finite values (stat_smooth).
## Warning in qt((1 - level)/2, df): NaNs produced
## Warning in qt((1 - level)/2, df): NaNs produced
## Warning in qt((1 - level)/2, df): NaNs produced
## Warning in qt((1 - level)/2, df): NaNs produced
## Warning in qt((1 - level)/2, df): NaNs produced
## Warning in qt((1 - level)/2, df): NaNs produced
## Warning in qt((1 - level)/2, df): NaNs produced
## Warning in qt((1 - level)/2, df): NaNs produced
## Warning in qt((1 - level)/2, df): NaNs produced
## Warning in qt((1 - level)/2, df): NaNs produced
## Warning in qt((1 - level)/2, df): NaNs produced
## Warning in qt((1 - level)/2, df): NaNs produced
## Warning in qt((1 - level)/2, df): NaNs produced
## Warning in qt((1 - level)/2, df): NaNs produced
## Warning in qt((1 - level)/2, df): NaNs produced
## Warning in qt((1 - level)/2, df): NaNs produced
## Warning in qt((1 - level)/2, df): NaNs produced
## Warning in qt((1 - level)/2, df): NaNs produced
## Warning in qt((1 - level)/2, df): NaNs produced
## Warning in qt((1 - level)/2, df): NaNs produced
## Warning in qt((1 - level)/2, df): NaNs produced
## Warning in qt((1 - level)/2, df): NaNs produced
## Warning in qt((1 - level)/2, df): NaNs produced
## Warning in qt((1 - level)/2, df): NaNs produced
## Warning in qt((1 - level)/2, df): NaNs produced
## Warning in qt((1 - level)/2, df): NaNs produced
## Warning in qt((1 - level)/2, df): NaNs produced
## Warning in qt((1 - level)/2, df): NaNs produced
## Warning in qt((1 - level)/2, df): NaNs produced
## Warning in qt((1 - level)/2, df): NaNs produced
## Warning in qt((1 - level)/2, df): NaNs produced
## Warning in qt((1 - level)/2, df): NaNs produced
## Warning in qt((1 - level)/2, df): NaNs produced
## Warning in qt((1 - level)/2, df): NaNs produced
## Warning: Removed 2 rows containing missing values (geom_point).
## Warning in max(ids, na.rm = TRUE): no non-missing arguments to max; returning
## -Inf
## Warning in max(ids, na.rm = TRUE): no non-missing arguments to max; returning
## -Inf
## Warning in max(ids, na.rm = TRUE): no non-missing arguments to max; returning
## -Inf
## Warning in max(ids, na.rm = TRUE): no non-missing arguments to max; returning
## -Inf
## Warning in max(ids, na.rm = TRUE): no non-missing arguments to max; returning
## -Inf
## Warning in max(ids, na.rm = TRUE): no non-missing arguments to max; returning
## -Inf
## Warning in max(ids, na.rm = TRUE): no non-missing arguments to max; returning
## -Inf
## Warning in max(ids, na.rm = TRUE): no non-missing arguments to max; returning
## -Inf
## Warning in max(ids, na.rm = TRUE): no non-missing arguments to max; returning
## -Inf
## Warning in max(ids, na.rm = TRUE): no non-missing arguments to max; returning
## -Inf
## Warning in max(ids, na.rm = TRUE): no non-missing arguments to max; returning
## -Inf
## Warning in max(ids, na.rm = TRUE): no non-missing arguments to max; returning
## -Inf
## Warning in max(ids, na.rm = TRUE): no non-missing arguments to max; returning
## -Inf
## Warning in max(ids, na.rm = TRUE): no non-missing arguments to max; returning
## -Inf
## Warning in max(ids, na.rm = TRUE): no non-missing arguments to max; returning
## -Inf
## Warning in max(ids, na.rm = TRUE): no non-missing arguments to max; returning
## -Inf
## Warning in max(ids, na.rm = TRUE): no non-missing arguments to max; returning
## -Inf
## Warning in max(ids, na.rm = TRUE): no non-missing arguments to max; returning
## -Inf
## Warning in max(ids, na.rm = TRUE): no non-missing arguments to max; returning
## -Inf
## Warning in max(ids, na.rm = TRUE): no non-missing arguments to max; returning
## -Inf
## Warning in max(ids, na.rm = TRUE): no non-missing arguments to max; returning
## -Inf
## Warning in max(ids, na.rm = TRUE): no non-missing arguments to max; returning
## -Inf
## Warning in max(ids, na.rm = TRUE): no non-missing arguments to max; returning
## -Inf
## Warning in max(ids, na.rm = TRUE): no non-missing arguments to max; returning
## -Inf
## Warning in max(ids, na.rm = TRUE): no non-missing arguments to max; returning
## -Inf
## Warning in max(ids, na.rm = TRUE): no non-missing arguments to max; returning
## -Inf
## Warning in max(ids, na.rm = TRUE): no non-missing arguments to max; returning
## -Inf
## Warning in max(ids, na.rm = TRUE): no non-missing arguments to max; returning
## -Inf
## Warning in max(ids, na.rm = TRUE): no non-missing arguments to max; returning
## -Inf
## Warning in max(ids, na.rm = TRUE): no non-missing arguments to max; returning
## -Inf
## Warning in max(ids, na.rm = TRUE): no non-missing arguments to max; returning
## -Inf
## Warning in max(ids, na.rm = TRUE): no non-missing arguments to max; returning
## -Inf
## Warning in max(ids, na.rm = TRUE): no non-missing arguments to max; returning
## -Inf
## Warning in max(ids, na.rm = TRUE): no non-missing arguments to max; returning
## -Inf
Linear Mixed Models for ROI and SNQ
Linear Mixed Models for Age, ROI, and SNQ
# ---- linear mixed models roi and snq ----
# Model 4: Interaction between Social Connection Quality and Quantity Predicting HPC
# Add covariates Sex + Personality
model_hpc_1 <- lmer(hippocampus_total_volume_corrected ~ SNQ_Quality * SNQ_Quantity + Age + sex + personality +
(1 | subject_id) + (1 | sub_study), data = full_data_frame_merged_over_55_corrected_for_icv)
## boundary (singular) fit: see help('isSingular')
# Display the summary of Model 1 (HPC)
summary(model_hpc_1)
## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: hippocampus_total_volume_corrected ~ SNQ_Quality * SNQ_Quantity +
## Age + sex + personality + (1 | subject_id) + (1 | sub_study)
## Data: full_data_frame_merged_over_55_corrected_for_icv
##
## REML criterion at convergence: -2706.8
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.61365 -0.63593 0.00581 0.71330 2.45051
##
## Random effects:
## Groups Name Variance Std.Dev.
## subject_id (Intercept) 2.545e-07 0.0005045
## sub_study (Intercept) 0.000e+00 0.0000000
## Residual 6.155e-07 0.0007845
## Number of obs: 255, groups: subject_id, 202; sub_study, 3
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 4.596e-03 8.501e-04 2.471e+02 5.406 1.52e-07 ***
## SNQ_Quality 1.776e-05 8.522e-05 2.404e+02 0.208 0.835110
## SNQ_Quantity 2.573e-06 1.949e-05 2.439e+02 0.132 0.895086
## Age 3.804e-06 6.606e-06 2.432e+02 0.576 0.565254
## sexMale -4.919e-04 1.295e-04 1.904e+02 -3.797 0.000197 ***
## personality -2.904e-05 4.108e-05 2.361e+02 -0.707 0.480275
## SNQ_Quality:SNQ_Quantity -4.186e-08 2.513e-06 2.407e+02 -0.017 0.986724
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) SNQ_Ql SNQ_Qn Age sexMal prsnlt
## SNQ_Quality -0.742
## SNQ_Quantty -0.810 0.896
## Age -0.608 0.049 0.084
## sexMale -0.026 -0.036 -0.031 -0.017
## personality -0.152 0.045 0.050 -0.045 -0.001
## SNQ_Q:SNQ_Q 0.720 -0.978 -0.916 -0.035 0.046 -0.052
## optimizer (nloptwrap) convergence code: 0 (OK)
## boundary (singular) fit: see help('isSingular')
# Model 5: Age Moderating the Relationship between Social Connection Quality and Quantity on HPC
# Including three-way interaction: Age * Social Connection Quality * Social Connection Quantity
# Add covariates Sex + Personality
model_hpc_2 <- lmer(hippocampus_total_volume_corrected ~ Age * SNQ_Quality * SNQ_Quantity + sex + personality +
(1 | subject_id) + (1 | sub_study), data = full_data_frame_merged_over_55_corrected_for_icv)
## Warning: Some predictor variables are on very different scales: consider
## rescaling
## boundary (singular) fit: see help('isSingular')
## Warning: Some predictor variables are on very different scales: consider
## rescaling
# Display the summary of Model 2 (HPC)
summary(model_hpc_2)
## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula:
## hippocampus_total_volume_corrected ~ Age * SNQ_Quality * SNQ_Quantity +
## sex + personality + (1 | subject_id) + (1 | sub_study)
## Data: full_data_frame_merged_over_55_corrected_for_icv
##
## REML criterion at convergence: -2634.4
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.7264 -0.6576 0.0111 0.6535 2.4106
##
## Random effects:
## Groups Name Variance Std.Dev.
## subject_id (Intercept) 2.335e-07 0.0004833
## sub_study (Intercept) 0.000e+00 0.0000000
## Residual 6.228e-07 0.0007892
## Number of obs: 255, groups: subject_id, 202; sub_study, 3
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 1.200e-02 5.169e-03 2.446e+02 2.321 0.021119
## Age -1.004e-04 7.310e-05 2.442e+02 -1.374 0.170816
## SNQ_Quality -9.850e-04 6.877e-04 2.436e+02 -1.432 0.153371
## SNQ_Quantity -1.541e-04 1.515e-04 2.441e+02 -1.018 0.309846
## sexMale -4.986e-04 1.285e-04 1.890e+02 -3.880 0.000144
## personality -2.547e-05 4.095e-05 2.352e+02 -0.622 0.534633
## Age:SNQ_Quality 1.413e-05 9.782e-06 2.433e+02 1.445 0.149770
## Age:SNQ_Quantity 2.203e-06 2.153e-06 2.435e+02 1.023 0.307346
## SNQ_Quality:SNQ_Quantity 2.081e-05 2.019e-05 2.428e+02 1.031 0.303644
## Age:SNQ_Quality:SNQ_Quantity -2.933e-07 2.883e-07 2.428e+02 -1.017 0.310108
##
## (Intercept) *
## Age
## SNQ_Quality
## SNQ_Quantity
## sexMale ***
## personality
## Age:SNQ_Quality
## Age:SNQ_Quantity
## SNQ_Quality:SNQ_Quantity
## Age:SNQ_Quality:SNQ_Quantity
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) Age SNQ_Ql SNQ_Qn sexMal prsnlt Ag:SNQ_Ql Ag:SNQ_Qn
## Age -0.991
## SNQ_Quality -0.907 0.902
## SNQ_Quantty -0.978 0.973 0.890
## sexMale 0.032 -0.038 -0.020 -0.049
## personality 0.003 -0.032 0.002 -0.018 0.001
## Ag:SNQ_Qlty 0.896 -0.905 -0.992 -0.882 0.016 0.003
## Ag:SNQ_Qntt 0.967 -0.979 -0.882 -0.992 0.046 0.024 0.888
## SNQ_Q:SNQ_Q 0.888 -0.884 -0.980 -0.908 0.030 -0.007 0.975 0.902
## A:SNQ_Q:SNQ -0.876 0.886 0.971 0.897 -0.025 0.001 -0.981 -0.906
## SNQ_Q:
## Age
## SNQ_Quality
## SNQ_Quantty
## sexMale
## personality
## Ag:SNQ_Qlty
## Ag:SNQ_Qntt
## SNQ_Q:SNQ_Q
## A:SNQ_Q:SNQ -0.992
## fit warnings:
## Some predictor variables are on very different scales: consider rescaling
## optimizer (nloptwrap) convergence code: 0 (OK)
## boundary (singular) fit: see help('isSingular')
# Model comparison using ANOVA to test if Model 2 significantly improves fit over Model 1
anova_comparison_hpc <- anova(model_hpc_1, model_hpc_2)
## refitting model(s) with ML (instead of REML)
# Display the ANOVA comparison between Model 1 and Model 2 (HPC)
print(anova_comparison_hpc)
## Data: full_data_frame_merged_over_55_corrected_for_icv
## Models:
## model_hpc_1: hippocampus_total_volume_corrected ~ SNQ_Quality * SNQ_Quantity + Age + sex + personality + (1 | subject_id) + (1 | sub_study)
## model_hpc_2: hippocampus_total_volume_corrected ~ Age * SNQ_Quality * SNQ_Quantity + sex + personality + (1 | subject_id) + (1 | sub_study)
## npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
## model_hpc_1 10 -2826.7 -2791.2 1423.3 -2846.7
## model_hpc_2 13 -2827.1 -2781.1 1426.6 -2853.1 6.4966 3 0.0898 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
# Extract and display the p-values for fixed effects for both models
# Comment out the model that does not win
anova(model_hpc_1) # Get significance of fixed effects for Model 1 (HPC)
## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## SNQ_Quality 2.6700e-08 2.6700e-08 1 240.45 0.0434 0.8351105
## SNQ_Quantity 1.0700e-08 1.0700e-08 1 243.85 0.0174 0.8950857
## Age 2.0410e-07 2.0410e-07 1 243.16 0.3316 0.5652539
## sex 8.8753e-06 8.8753e-06 1 190.43 14.4197 0.0001966
## personality 3.0760e-07 3.0760e-07 1 236.06 0.4998 0.4802752
## SNQ_Quality:SNQ_Quantity 2.0000e-10 2.0000e-10 1 240.69 0.0003 0.9867239
##
## SNQ_Quality
## SNQ_Quantity
## Age
## sex ***
## personality
## SNQ_Quality:SNQ_Quantity
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
anova(model_hpc_2) # Get significance of fixed effects for Model 2 (HPC)
## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value
## Age 1.1751e-06 1.1751e-06 1 244.19 1.8869
## SNQ_Quality 1.2774e-06 1.2774e-06 1 243.55 2.0512
## SNQ_Quantity 6.4500e-07 6.4500e-07 1 244.06 1.0356
## sex 9.3740e-06 9.3740e-06 1 188.99 15.0517
## personality 2.4080e-07 2.4080e-07 1 235.17 0.3867
## Age:SNQ_Quality 1.3002e-06 1.3002e-06 1 243.32 2.0878
## Age:SNQ_Quantity 6.5170e-07 6.5170e-07 1 243.49 1.0464
## SNQ_Quality:SNQ_Quantity 6.6180e-07 6.6180e-07 1 242.84 1.0626
## Age:SNQ_Quality:SNQ_Quantity 6.4430e-07 6.4430e-07 1 242.75 1.0345
## Pr(>F)
## Age 0.1708156
## SNQ_Quality 0.1533710
## SNQ_Quantity 0.3098461
## sex 0.0001444 ***
## personality 0.5346334
## Age:SNQ_Quality 0.1497696
## Age:SNQ_Quantity 0.3073458
## SNQ_Quality:SNQ_Quantity 0.3036436
## Age:SNQ_Quality:SNQ_Quantity 0.3101078
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Linear Mixed Models for Amygdala and SNQ
Linear Mixed Models for Age, Amygdala, and SNQ
# Model 6: Interaction between Social Connection Quality and Quantity Predicting Amygdala Total Volume
# Add covariates Sex + Personality
model_amygdala_1 <- lmer(amygdala_total_volume_corrected ~ SNQ_Quality * SNQ_Quantity + Age + sex + personality +
(1 | subject_id) + (1 | sub_study), data = full_data_frame_merged_over_55_corrected_for_icv)
# Display the summary of Model 1 (Amygdala)
summary(model_amygdala_1)
## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: amygdala_total_volume_corrected ~ SNQ_Quality * SNQ_Quantity +
## Age + sex + personality + (1 | subject_id) + (1 | sub_study)
## Data: full_data_frame_merged_over_55_corrected_for_icv
##
## REML criterion at convergence: -3162.8
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.3119 -0.6635 -0.1147 0.6637 2.7559
##
## Random effects:
## Groups Name Variance Std.Dev.
## subject_id (Intercept) 1.724e-08 1.313e-04
## sub_study (Intercept) 6.831e-09 8.265e-05
## Residual 1.175e-07 3.428e-04
## Number of obs: 255, groups: subject_id, 202; sub_study, 3
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 1.851e-03 3.408e-04 2.392e+02 5.431 1.38e-07 ***
## SNQ_Quality 8.332e-06 3.422e-05 2.465e+02 0.243 0.80786
## SNQ_Quantity 2.507e-06 7.789e-06 2.464e+02 0.322 0.74784
## Age 1.822e-06 2.641e-06 2.465e+02 0.690 0.49091
## sexMale -1.493e-04 4.987e-05 1.765e+02 -2.995 0.00314 **
## personality -2.991e-05 1.646e-05 2.438e+02 -1.817 0.07041 .
## SNQ_Quality:SNQ_Quantity -1.502e-07 1.006e-06 2.456e+02 -0.149 0.88145
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) SNQ_Ql SNQ_Qn Age sexMal prsnlt
## SNQ_Quality -0.728
## SNQ_Quantty -0.799 0.889
## Age -0.596 0.035 0.076
## sexMale -0.023 -0.046 -0.030 -0.015
## personality -0.152 0.044 0.046 -0.041 0.004
## SNQ_Q:SNQ_Q 0.709 -0.976 -0.914 -0.028 0.049 -0.052
# Model 7: Age Moderating the Relationship between Social Connection Quality and Quantity on Amygdala Total Volume
# Including three-way interaction: Age * Social Connection Quality * Social Connection Quantity
# Add covariates Sex + Personality
model_amygdala_2 <- lmer(amygdala_total_volume_corrected ~ Age * SNQ_Quality * SNQ_Quantity + sex + personality +
(1 | subject_id) + (1 | sub_study), data = full_data_frame_merged_over_55_corrected_for_icv)
## Warning: Some predictor variables are on very different scales: consider
## rescaling
## Warning: Some predictor variables are on very different scales: consider
## rescaling
# Display the summary of Model 2 (Amygdala)
summary(model_amygdala_2)
## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: amygdala_total_volume_corrected ~ Age * SNQ_Quality * SNQ_Quantity +
## sex + personality + (1 | subject_id) + (1 | sub_study)
## Data: full_data_frame_merged_over_55_corrected_for_icv
##
## REML criterion at convergence: -3081.6
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.26072 -0.68290 -0.04703 0.66826 2.80519
##
## Random effects:
## Groups Name Variance Std.Dev.
## subject_id (Intercept) 1.632e-08 1.278e-04
## sub_study (Intercept) 6.719e-09 8.197e-05
## Residual 1.184e-07 3.440e-04
## Number of obs: 255, groups: subject_id, 202; sub_study, 3
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 4.316e-03 2.066e-03 2.437e+02 2.089 0.03775
## Age -3.358e-05 2.923e-05 2.434e+02 -1.149 0.25178
## SNQ_Quality -4.149e-04 2.752e-04 2.435e+02 -1.508 0.13293
## SNQ_Quantity -7.489e-05 6.059e-05 2.433e+02 -1.236 0.21766
## sexMale -1.480e-04 4.997e-05 1.752e+02 -2.962 0.00348
## personality -3.051e-05 1.650e-05 2.413e+02 -1.849 0.06563
## Age:SNQ_Quality 6.109e-06 3.916e-06 2.436e+02 1.560 0.12004
## Age:SNQ_Quantity 1.114e-06 8.618e-07 2.432e+02 1.292 0.19751
## SNQ_Quality:SNQ_Quantity 1.303e-05 8.086e-06 2.433e+02 1.612 0.10831
## Age:SNQ_Quality:SNQ_Quantity -1.904e-07 1.155e-07 2.435e+02 -1.649 0.10052
##
## (Intercept) *
## Age
## SNQ_Quality
## SNQ_Quantity
## sexMale **
## personality .
## Age:SNQ_Quality
## Age:SNQ_Quantity
## SNQ_Quality:SNQ_Quantity
## Age:SNQ_Quality:SNQ_Quantity
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) Age SNQ_Ql SNQ_Qn sexMal prsnlt Ag:SNQ_Ql Ag:SNQ_Qn
## Age -0.991
## SNQ_Quality -0.907 0.901
## SNQ_Quantty -0.978 0.973 0.889
## sexMale 0.033 -0.039 -0.020 -0.051
## personality 0.002 -0.031 0.003 -0.018 0.007
## Ag:SNQ_Qlty 0.895 -0.904 -0.992 -0.880 0.014 0.002
## Ag:SNQ_Qntt 0.967 -0.978 -0.881 -0.992 0.048 0.024 0.887
## SNQ_Q:SNQ_Q 0.887 -0.883 -0.980 -0.907 0.030 -0.009 0.975 0.901
## A:SNQ_Q:SNQ -0.875 0.886 0.971 0.897 -0.024 0.003 -0.981 -0.905
## SNQ_Q:
## Age
## SNQ_Quality
## SNQ_Quantty
## sexMale
## personality
## Ag:SNQ_Qlty
## Ag:SNQ_Qntt
## SNQ_Q:SNQ_Q
## A:SNQ_Q:SNQ -0.992
## fit warnings:
## Some predictor variables are on very different scales: consider rescaling
# Model comparison using ANOVA to test if Model 2 significantly improves fit over Model 1
anova_comparison_amygdala <- anova(model_amygdala_1, model_amygdala_2)
## refitting model(s) with ML (instead of REML)
# Display the ANOVA comparison between Model 1 and Model 2 (Amygdala)
print(anova_comparison_amygdala)
## Data: full_data_frame_merged_over_55_corrected_for_icv
## Models:
## model_amygdala_1: amygdala_total_volume_corrected ~ SNQ_Quality * SNQ_Quantity + Age + sex + personality + (1 | subject_id) + (1 | sub_study)
## model_amygdala_2: amygdala_total_volume_corrected ~ Age * SNQ_Quality * SNQ_Quantity + sex + personality + (1 | subject_id) + (1 | sub_study)
## npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
## model_amygdala_1 10 -3294.0 -3258.6 1657.0 -3314.0
## model_amygdala_2 13 -3291.1 -3245.1 1658.5 -3317.1 3.131 3 0.3719
# Extract and display the p-values for fixed effects for both models
# Comment out the model that does not win
anova(model_amygdala_1) # Get significance of fixed effects for Model 1 (Amygdala)
## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## SNQ_Quality 6.9600e-09 6.9600e-09 1 246.51 0.0593 0.807861
## SNQ_Quantity 1.2170e-08 1.2170e-08 1 246.36 0.1036 0.747838
## Age 5.5920e-08 5.5920e-08 1 246.47 0.4760 0.490911
## sex 1.0538e-06 1.0538e-06 1 176.46 8.9683 0.003142 **
## personality 3.8802e-07 3.8802e-07 1 243.76 3.3024 0.070407 .
## SNQ_Quality:SNQ_Quantity 2.6200e-09 2.6200e-09 1 245.56 0.0223 0.881453
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
anova(model_amygdala_2) # Get significance of fixed effects for Model 2 (Amygdala)
## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value
## Age 1.5621e-07 1.5621e-07 1 243.41 1.3197
## SNQ_Quality 2.6907e-07 2.6907e-07 1 243.50 2.2732
## SNQ_Quantity 1.8083e-07 1.8083e-07 1 243.33 1.5277
## sex 1.0386e-06 1.0386e-06 1 175.24 8.7747
## personality 4.0484e-07 4.0484e-07 1 241.29 3.4202
## Age:SNQ_Quality 2.8808e-07 2.8808e-07 1 243.63 2.4338
## Age:SNQ_Quantity 1.9765e-07 1.9765e-07 1 243.24 1.6698
## SNQ_Quality:SNQ_Quantity 3.0750e-07 3.0750e-07 1 243.27 2.5978
## Age:SNQ_Quality:SNQ_Quantity 3.2171e-07 3.2171e-07 1 243.49 2.7179
## Pr(>F)
## Age 0.251777
## SNQ_Quality 0.132928
## SNQ_Quantity 0.217655
## sex 0.003478 **
## personality 0.065627 .
## Age:SNQ_Quality 0.120044
## Age:SNQ_Quantity 0.197510
## SNQ_Quality:SNQ_Quantity 0.108306
## Age:SNQ_Quality:SNQ_Quantity 0.100519
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Linear Mixed Models for Cingulate and SNQ
Linear Mixed Models for Age, Cingulate, and SNQ
# Model 8: Interaction between Social Connection Quality and Quantity Predicting Total Cingulate Volume
# Add covariates Sex + Personality
model_cingulate_1 <- lmer(total_cingulate_volume_corrected ~ SNQ_Quality * SNQ_Quantity + Age + sex + personality +
(1 | subject_id) + (1 | sub_study), data = full_data_frame_merged_over_55_corrected_for_icv)
# Display the summary of Model 1 (Cingulate)
summary(model_cingulate_1)
## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: total_cingulate_volume_corrected ~ SNQ_Quality * SNQ_Quantity +
## Age + sex + personality + (1 | subject_id) + (1 | sub_study)
## Data: full_data_frame_merged_over_55_corrected_for_icv
##
## REML criterion at convergence: -1563
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.7159 -0.3415 -0.0086 0.3215 2.5299
##
## Random effects:
## Groups Name Variance Std.Dev.
## subject_id (Intercept) 9.614e-05 0.009805
## sub_study (Intercept) 8.741e-04 0.029565
## Residual 1.334e-05 0.003653
## Number of obs: 255, groups: subject_id, 202; sub_study, 3
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 1.036e-01 1.839e-02 2.690e+00 5.633 0.01474 *
## SNQ_Quality 1.853e-03 6.531e-04 9.162e+01 2.836 0.00561 **
## SNQ_Quantity 3.913e-04 1.521e-04 9.753e+01 2.572 0.01161 *
## Age 5.723e-05 5.156e-05 9.857e+01 1.110 0.26967
## sexMale -2.994e-03 1.543e-03 1.939e+02 -1.941 0.05374 .
## personality -9.990e-06 2.988e-04 8.118e+01 -0.033 0.97341
## SNQ_Quality:SNQ_Quantity -5.622e-05 1.915e-05 9.164e+01 -2.936 0.00421 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) SNQ_Ql SNQ_Qn Age sexMal prsnlt
## SNQ_Quality -0.277
## SNQ_Quantty -0.302 0.905
## Age -0.237 0.087 0.145
## sexMale -0.023 -0.030 -0.024 0.001
## personality -0.063 0.024 0.060 -0.003 -0.005
## SNQ_Q:SNQ_Q 0.267 -0.972 -0.933 -0.072 0.032 -0.030
# Model 9: Age Moderating the Relationship between Social Connection Quality and Quantity on Total Cingulate Volume
# Including three-way interaction: Age * Social Connection Quality * Social Connection Quantity
# Add covariates Sex + Personality
model_cingulate_2 <- lmer(total_cingulate_volume_corrected ~ Age * SNQ_Quality * SNQ_Quantity + sex + personality +
(1 | subject_id) + (1 | sub_study), data = full_data_frame_merged_over_55_corrected_for_icv)
## Warning: Some predictor variables are on very different scales: consider
## rescaling
## Warning: Some predictor variables are on very different scales: consider
## rescaling
# Display the summary of Model 2 (Cingulate)
summary(model_cingulate_2)
## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: total_cingulate_volume_corrected ~ Age * SNQ_Quality * SNQ_Quantity +
## sex + personality + (1 | subject_id) + (1 | sub_study)
## Data: full_data_frame_merged_over_55_corrected_for_icv
##
## REML criterion at convergence: -1500.8
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.70161 -0.32313 -0.00319 0.30781 2.53827
##
## Random effects:
## Groups Name Variance Std.Dev.
## subject_id (Intercept) 9.441e-05 0.009716
## sub_study (Intercept) 8.753e-04 0.029586
## Residual 1.375e-05 0.003709
## Number of obs: 255, groups: subject_id, 202; sub_study, 3
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 9.929e-02 4.761e-02 7.287e+01 2.085 0.0405
## Age 1.321e-04 6.178e-04 1.300e+02 0.214 0.8310
## SNQ_Quality 4.505e-03 5.628e-03 1.141e+02 0.800 0.4251
## SNQ_Quantity 9.010e-04 1.284e-03 1.312e+02 0.702 0.4842
## sexMale -3.026e-03 1.536e-03 1.933e+02 -1.970 0.0503
## personality 3.244e-05 3.035e-04 8.052e+01 0.107 0.9151
## Age:SNQ_Quality -4.107e-05 7.908e-05 1.085e+02 -0.519 0.6046
## Age:SNQ_Quantity -7.652e-06 1.794e-05 1.236e+02 -0.427 0.6705
## SNQ_Quality:SNQ_Quantity -1.879e-04 1.636e-04 1.104e+02 -1.148 0.2533
## Age:SNQ_Quality:SNQ_Quantity 1.967e-06 2.315e-06 1.058e+02 0.850 0.3975
##
## (Intercept) *
## Age
## SNQ_Quality
## SNQ_Quantity
## sexMale .
## personality
## Age:SNQ_Quality
## Age:SNQ_Quantity
## SNQ_Quality:SNQ_Quantity
## Age:SNQ_Quality:SNQ_Quantity
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) Age SNQ_Ql SNQ_Qn sexMal prsnlt Ag:SNQ_Ql Ag:SNQ_Qn
## Age -0.927
## SNQ_Quality -0.855 0.914
## SNQ_Quantty -0.915 0.977 0.901
## sexMale 0.025 -0.037 -0.021 -0.045
## personality 0.030 -0.058 -0.018 -0.042 -0.001
## Ag:SNQ_Qlty 0.840 -0.911 -0.993 -0.888 0.017 0.018
## Ag:SNQ_Qntt 0.907 -0.981 -0.898 -0.993 0.042 0.048 0.899
## SNQ_Q:SNQ_Q 0.834 -0.893 -0.982 -0.913 0.027 0.006 0.978 0.911
## A:SNQ_Q:SNQ -0.818 0.888 0.973 0.897 -0.022 -0.007 -0.983 -0.909
## SNQ_Q:
## Age
## SNQ_Quality
## SNQ_Quantty
## sexMale
## personality
## Ag:SNQ_Qlty
## Ag:SNQ_Qntt
## SNQ_Q:SNQ_Q
## A:SNQ_Q:SNQ -0.993
## fit warnings:
## Some predictor variables are on very different scales: consider rescaling
# Model comparison using ANOVA to test if Model 2 significantly improves fit over Model 1
anova_comparison_cingulate <- anova(model_cingulate_1, model_cingulate_2)
## refitting model(s) with ML (instead of REML)
# Display the ANOVA comparison between Model 1 and Model 2 (Cingulate)
print(anova_comparison_cingulate)
## Data: full_data_frame_merged_over_55_corrected_for_icv
## Models:
## model_cingulate_1: total_cingulate_volume_corrected ~ SNQ_Quality * SNQ_Quantity + Age + sex + personality + (1 | subject_id) + (1 | sub_study)
## model_cingulate_2: total_cingulate_volume_corrected ~ Age * SNQ_Quality * SNQ_Quantity + sex + personality + (1 | subject_id) + (1 | sub_study)
## npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
## model_cingulate_1 10 -1646.6 -1611.2 833.29 -1666.6
## model_cingulate_2 13 -1644.8 -1598.8 835.39 -1670.8 4.2127 3 0.2394
# Extract and display the p-values for fixed effects for both models
# Comment out the model that does not win
anova(model_cingulate_1) # Get significance of fixed effects for Model 1 (Cingulate)
## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## SNQ_Quality 1.0735e-04 1.0735e-04 1 91.617 8.0457 0.005614
## SNQ_Quantity 8.8299e-05 8.8299e-05 1 97.526 6.6176 0.011605
## Age 1.6442e-05 1.6442e-05 1 98.571 1.2322 0.269672
## sex 5.0254e-05 5.0254e-05 1 193.941 3.7663 0.053744
## personality 1.5000e-08 1.5000e-08 1 81.177 0.0011 0.973414
## SNQ_Quality:SNQ_Quantity 1.1500e-04 1.1500e-04 1 91.644 8.6188 0.004206
##
## SNQ_Quality **
## SNQ_Quantity *
## Age
## sex .
## personality
## SNQ_Quality:SNQ_Quantity **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
anova(model_cingulate_2) # Get significance of fixed effects for Model 2 (Cingulate)
## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value
## Age 6.2900e-07 6.2900e-07 1 129.994 0.0457
## SNQ_Quality 8.8120e-06 8.8120e-06 1 114.141 0.6406
## SNQ_Quantity 6.7710e-06 6.7710e-06 1 131.169 0.4922
## sex 5.3384e-05 5.3384e-05 1 193.338 3.8812
## personality 1.5700e-07 1.5700e-07 1 80.523 0.0114
## Age:SNQ_Quality 3.7100e-06 3.7100e-06 1 108.538 0.2697
## Age:SNQ_Quantity 2.5020e-06 2.5020e-06 1 123.573 0.1819
## SNQ_Quality:SNQ_Quantity 1.8142e-05 1.8142e-05 1 110.363 1.3190
## Age:SNQ_Quality:SNQ_Quantity 9.9280e-06 9.9280e-06 1 105.757 0.7218
## Pr(>F)
## Age 0.83100
## SNQ_Quality 0.42514
## SNQ_Quantity 0.48417
## sex 0.05026 .
## personality 0.91514
## Age:SNQ_Quality 0.60458
## Age:SNQ_Quantity 0.67047
## SNQ_Quality:SNQ_Quantity 0.25326
## Age:SNQ_Quality:SNQ_Quantity 0.39746
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Testing the Order of Fixed Effects
Testing if the Order of Fixed Effects Alters Results
# Model A: original order with interaction between Social Connection Quality and Quantity Predicting RAVLT
# Add covariates in the order: SNQ_Quality * SNQ_Quantity + age + sex + personality
model_A <- lmer(RAVLT ~ SNQ_Quality * SNQ_Quantity + Age + sex + personality +
(1 | subject_id) + (1 | sub_study), data = full_data_frame_merged_over_55_corrected_for_icv)
## boundary (singular) fit: see help('isSingular')
# Display the summary of Model A
summary(model_A)
## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: RAVLT ~ SNQ_Quality * SNQ_Quantity + Age + sex + personality +
## (1 | subject_id) + (1 | sub_study)
## Data: full_data_frame_merged_over_55_corrected_for_icv
##
## REML criterion at convergence: 1497
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.00317 -0.81597 -0.00711 0.76243 1.93361
##
## Random effects:
## Groups Name Variance Std.Dev.
## subject_id (Intercept) 0.0000 0.000
## sub_study (Intercept) 0.1467 0.383
## Residual 19.5702 4.424
## Number of obs: 255, groups: subject_id, 202; sub_study, 3
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.973013 4.070500 247.926537 0.239 0.811
## SNQ_Quality 0.286852 0.412374 247.995042 0.696 0.487
## SNQ_Quantity 0.120122 0.093989 247.616984 1.278 0.202
## Age 0.039252 0.031849 247.622401 1.232 0.219
## sexMale 0.498687 0.585491 247.938817 0.852 0.395
## personality -0.130724 0.199014 247.102297 -0.657 0.512
## SNQ_Quality:SNQ_Quantity -0.009348 0.012148 247.412386 -0.770 0.442
##
## Correlation of Fixed Effects:
## (Intr) SNQ_Ql SNQ_Qn Age sexMal prsnlt
## SNQ_Quality -0.735
## SNQ_Quantty -0.805 0.889
## Age -0.598 0.032 0.069
## sexMale -0.020 -0.044 -0.028 -0.022
## personality -0.156 0.050 0.047 -0.038 0.005
## SNQ_Q:SNQ_Q 0.714 -0.977 -0.912 -0.022 0.049 -0.057
## optimizer (nloptwrap) convergence code: 0 (OK)
## boundary (singular) fit: see help('isSingular')
# Model B: altered order changing the order of fixed effects
# Rearrange the order of fixed effects: personality + sex + age + SNQ_Quality * SNQ_Quantity
model_B <- lmer(RAVLT ~ personality + sex + Age + SNQ_Quality * SNQ_Quantity +
(1 | subject_id) + (1 | sub_study), data = full_data_frame_merged_over_55_corrected_for_icv)
## boundary (singular) fit: see help('isSingular')
# Display the summary of Model B
summary(model_B)
## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: RAVLT ~ personality + sex + Age + SNQ_Quality * SNQ_Quantity +
## (1 | subject_id) + (1 | sub_study)
## Data: full_data_frame_merged_over_55_corrected_for_icv
##
## REML criterion at convergence: 1497
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.00317 -0.81597 -0.00711 0.76243 1.93361
##
## Random effects:
## Groups Name Variance Std.Dev.
## subject_id (Intercept) 0.0000 0.000
## sub_study (Intercept) 0.1467 0.383
## Residual 19.5702 4.424
## Number of obs: 255, groups: subject_id, 202; sub_study, 3
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.973013 4.070500 247.926538 0.239 0.811
## personality -0.130724 0.199014 247.102297 -0.657 0.512
## sexMale 0.498687 0.585491 247.938817 0.852 0.395
## Age 0.039252 0.031849 247.622401 1.232 0.219
## SNQ_Quality 0.286852 0.412374 247.995042 0.696 0.487
## SNQ_Quantity 0.120122 0.093989 247.616982 1.278 0.202
## SNQ_Quality:SNQ_Quantity -0.009348 0.012148 247.412383 -0.770 0.442
##
## Correlation of Fixed Effects:
## (Intr) prsnlt sexMal Age SNQ_Ql SNQ_Qn
## personality -0.156
## sexMale -0.020 0.005
## Age -0.598 -0.038 -0.022
## SNQ_Quality -0.735 0.050 -0.044 0.032
## SNQ_Quantty -0.805 0.047 -0.028 0.069 0.889
## SNQ_Q:SNQ_Q 0.714 -0.057 0.049 -0.022 -0.977 -0.912
## optimizer (nloptwrap) convergence code: 0 (OK)
## boundary (singular) fit: see help('isSingular')
# Model comparison using ANOVA to test if changing the order affects model fit
anova_comparison_order <- anova(model_A, model_B)
## refitting model(s) with ML (instead of REML)
# Display the ANOVA comparison between Model A and Model B
print(anova_comparison_order)
## Data: full_data_frame_merged_over_55_corrected_for_icv
## Models:
## model_A: RAVLT ~ SNQ_Quality * SNQ_Quantity + Age + sex + personality + (1 | subject_id) + (1 | sub_study)
## model_B: RAVLT ~ personality + sex + Age + SNQ_Quality * SNQ_Quantity + (1 | subject_id) + (1 | sub_study)
## npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
## model_A 10 1495.9 1531.3 -737.97 1475.9
## model_B 10 1495.9 1531.3 -737.97 1475.9 0 0
# Extract and display the p-values for fixed effects for both models to compare significance
anova(model_A) # Get significance of fixed effects for Model A
## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## SNQ_Quality 9.470 9.470 1 248.00 0.4839 0.4873
## SNQ_Quantity 31.966 31.966 1 247.62 1.6334 0.2024
## Age 29.726 29.726 1 247.62 1.5190 0.2189
## sex 14.197 14.197 1 247.94 0.7255 0.3952
## personality 8.444 8.444 1 247.10 0.4315 0.5119
## SNQ_Quality:SNQ_Quantity 11.589 11.589 1 247.41 0.5922 0.4423
anova(model_B) # Get significance of fixed effects for Model B
## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## personality 8.444 8.444 1 247.10 0.4315 0.5119
## sex 14.197 14.197 1 247.94 0.7255 0.3952
## Age 29.726 29.726 1 247.62 1.5190 0.2189
## SNQ_Quality 9.470 9.470 1 248.00 0.4839 0.4873
## SNQ_Quantity 31.966 31.966 1 247.62 1.6334 0.2024
## SNQ_Quality:SNQ_Quantity 11.589 11.589 1 247.41 0.5922 0.4423
# Model with both random effects
model_1 <- lmer(RAVLT ~ SNQ_Quality * SNQ_Quantity + Age + sex + personality +
(1 | subject_id) + (1 | sub_study), data = full_data_frame_merged_over_55_corrected_for_icv)
## boundary (singular) fit: see help('isSingular')
# Model without subject_id
model_no_subject <- lmer(RAVLT ~ SNQ_Quality * SNQ_Quantity + Age + sex + personality +
(1 | sub_study), data = full_data_frame_merged_over_55_corrected_for_icv)
# Model without sub-study
model_no_substudy <- lmer(RAVLT ~ SNQ_Quality * SNQ_Quantity + Age + sex + personality +
(1 | subject_id), data = full_data_frame_merged_over_55_corrected_for_icv)
## boundary (singular) fit: see help('isSingular')
# Compare models using ANOVA
anova_full_vs_no_subject <- anova(model_1, model_no_subject)
## refitting model(s) with ML (instead of REML)
anova_full_vs_no_substudy <- anova(model_1, model_no_substudy)
## refitting model(s) with ML (instead of REML)
# Display results
print("Comparison of full model vs. model without subject-level random effect:")
## [1] "Comparison of full model vs. model without subject-level random effect:"
print(anova_full_vs_no_subject)
## Data: full_data_frame_merged_over_55_corrected_for_icv
## Models:
## model_no_subject: RAVLT ~ SNQ_Quality * SNQ_Quantity + Age + sex + personality + (1 | sub_study)
## model_1: RAVLT ~ SNQ_Quality * SNQ_Quantity + Age + sex + personality + (1 | subject_id) + (1 | sub_study)
## npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
## model_no_subject 9 1493.9 1525.8 -737.97 1475.9
## model_1 10 1495.9 1531.3 -737.97 1475.9 0 1 1
print("Comparison of full model vs. model without sub-study-level random effect:")
## [1] "Comparison of full model vs. model without sub-study-level random effect:"
print(anova_full_vs_no_substudy)
## Data: full_data_frame_merged_over_55_corrected_for_icv
## Models:
## model_no_substudy: RAVLT ~ SNQ_Quality * SNQ_Quantity + Age + sex + personality + (1 | subject_id)
## model_1: RAVLT ~ SNQ_Quality * SNQ_Quantity + Age + sex + personality + (1 | subject_id) + (1 | sub_study)
## npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
## model_no_substudy 9 1493.9 1525.8 -737.97 1475.9
## model_1 10 1495.9 1531.3 -737.97 1475.9 0 1 1