merge_nameability_data_and_plot
Yuguang Duan, Ellise Suffill
2/21/2021
A basic look number of moves across low first / high first nameability
Pilot_ 1 (second characteristic of color/shape non-varied)
Has issue of stack history was visible so task too easy.
## Automatically converting the following non-factors to factors: name_order, seriesNo, shape_or_color## ------------------------------------------------------------------------------## You have loaded plyr after dplyr - this is likely to cause problems.
## If you need functions from both plyr and dplyr, please load plyr first, then dplyr:
## library(plyr); library(dplyr)## ------------------------------------------------------------------------------
## `stat_bindot()` using `bins = 30`. Pick better value with `binwidth`.
## Automatically converting the following non-factors to factors: name_order, episodeNo, shape_or_color## `geom_smooth()` using method = 'loess' and formula 'y ~ x'
## `geom_smooth()` using method = 'loess' and formula 'y ~ x'
## `geom_smooth()` using method = 'loess' and formula 'y ~ x'
## Linear mixed model fit by REML ['lmerMod']
## Formula: uniMove ~ ruleId + episodeNo + (ruleId | playerId)
## Data: .
##
## REML criterion at convergence: -23.6
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.1392 -0.6092 -0.1099 0.4482 4.6943
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## playerId (Intercept) 0.02238 0.1496
## ruleIdcolor_naming_low 0.01046 0.1023 -0.76
## Residual 0.03592 0.1895
## Number of obs: 140, groups: playerId, 19
##
## Fixed effects:
## Estimate Std. Error t value
## (Intercept) 1.334776 0.045566 29.294
## ruleIdcolor_naming_low -0.003324 0.040107 -0.083
## episodeNo -0.092089 0.013509 -6.817
##
## Correlation of Fixed Effects:
## (Intr) rlId__
## rlIdclr_nm_ -0.634
## episodeNo -0.413 0.006## boundary (singular) fit: see ?isSingular## Linear mixed model fit by REML ['lmerMod']
## Formula: uniMove ~ ruleId + episodeNo + (ruleId | playerId)
## Data: .
##
## REML criterion at convergence: -55.6
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.7082 -0.6577 -0.2156 0.4457 3.2716
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## playerId (Intercept) 0.0141946 0.119141
## ruleIdshape_naming_low 0.0000513 0.007162 -1.00
## Residual 0.0274303 0.165621
## Number of obs: 128, groups: playerId, 19
##
## Fixed effects:
## Estimate Std. Error t value
## (Intercept) 1.270530 0.038263 33.205
## ruleIdshape_naming_low 0.009619 0.029551 0.326
## episodeNo -0.078515 0.014030 -5.596
##
## Correlation of Fixed Effects:
## (Intr) rlId__
## rlIdshp_nm_ -0.412
## episodeNo -0.428 -0.054
## convergence code: 0
## boundary (singular) fit: see ?isSingular## boundary (singular) fit: see ?isSingular## Linear mixed model fit by REML ['lmerMod']
## Formula: uniMove ~ 1 + name_order + seriesNo + episodeNo + ((1 | playerId) +
## (0 + ruleId | playerId))
## Data: .
##
## REML criterion at convergence: -108.5
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.2422 -0.6895 -0.1202 0.4712 5.0284
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## playerId (Intercept) 3.061e-11 5.533e-06
## playerId.1 ruleIdcolor_naming_high 2.387e-02 1.545e-01
## ruleIdcolor_naming_low 1.025e-02 1.013e-01 0.71
## ruleIdshape_naming_high 1.163e-02 1.078e-01 0.99 0.80
## ruleIdshape_naming_low 1.288e-02 1.135e-01 0.87 0.97 0.93
## Residual 3.030e-02 1.741e-01
## Number of obs: 268, groups: playerId, 19
##
## Fixed effects:
## Estimate Std. Error t value
## (Intercept) 1.347953 0.038580 34.939
## name_orderlow_first 0.002386 0.026731 0.089
## seriesNo -0.028270 0.010087 -2.803
## episodeNo -0.090470 0.009501 -9.522
##
## Correlation of Fixed Effects:
## (Intr) nm_rd_ serisN
## nm_rdrlw_fr -0.461
## seriesNo -0.464 0.014
## episodeNo -0.344 -0.012 0.081
## convergence code: 0
## boundary (singular) fit: see ?isSingularPilot 2 (shape varies by color rule; color varies by shape rule)
Color findings seem flipped compared to expectation (i.e., high nameability = more moves than low nameability), but is probably due to shape interference.
## Automatically converting the following non-factors to factors: name_order, seriesNo, shape_or_color
## `stat_bindot()` using `bins = 30`. Pick better value with `binwidth`.
## Automatically converting the following non-factors to factors: name_order, episodeNo, shape_or_color## `geom_smooth()` using method = 'loess' and formula 'y ~ x'
## `geom_smooth()` using method = 'loess' and formula 'y ~ x'
## `geom_smooth()` using method = 'loess' and formula 'y ~ x'
## Linear mixed model fit by REML ['lmerMod']
## Formula: uniMove ~ ruleId + episodeNo + (ruleId | playerId)
## Data: .
##
## REML criterion at convergence: 64.2
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.5228 -0.6527 -0.1827 0.4117 3.8196
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## playerId (Intercept) 0.012901 0.11358
## ruleIdcolor_naming_low_varied 0.004217 0.06494 0.03
## Residual 0.072841 0.26989
## Number of obs: 148, groups: playerId, 20
##
## Fixed effects:
## Estimate Std. Error t value
## (Intercept) 1.46815 0.04758 30.854
## ruleIdcolor_naming_low_varied -0.06668 0.04731 -1.409
## episodeNo -0.10139 0.01598 -6.343
##
## Correlation of Fixed Effects:
## (Intr) rlI___
## rlIdclr_n__ -0.451
## episodeNo -0.516 0.030## boundary (singular) fit: see ?isSingular## Linear mixed model fit by REML ['lmerMod']
## Formula: uniMove ~ ruleId + episodeNo + (ruleId | playerId)
## Data: .
##
## REML criterion at convergence: 135.5
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.7919 -0.5407 -0.1781 0.4363 5.2198
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## playerId (Intercept) 0.03507 0.1873
## ruleIdshape_naming_low_varied 0.01228 0.1108 1.00
## Residual 0.10764 0.3281
## Number of obs: 151, groups: playerId, 20
##
## Fixed effects:
## Estimate Std. Error t value
## (Intercept) 1.56201 0.06441 24.251
## ruleIdshape_naming_low_varied 0.07899 0.05929 1.332
## episodeNo -0.16655 0.02108 -7.902
##
## Correlation of Fixed Effects:
## (Intr) rlI___
## rlIdshp_n__ -0.101
## episodeNo -0.474 0.001
## convergence code: 0
## boundary (singular) fit: see ?isSingular## boundary (singular) fit: see ?isSingular## Linear mixed model fit by REML ['lmerMod']
## Formula: uniMove ~ 1 + name_order + seriesNo + episodeNo + ((1 | playerId) +
## (0 + ruleId | playerId))
## Data: .
##
## REML criterion at convergence: 199.5
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.0347 -0.5717 -0.2107 0.3836 5.8818
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## playerId (Intercept) 0.009138 0.09559
## playerId.1 ruleIdcolor_naming_high_varied 0.009134 0.09557
## ruleIdcolor_naming_low_varied 0.006474 0.08046 0.39
## ruleIdshape_naming_high_varied 0.025878 0.16087 0.83 0.84
## ruleIdshape_naming_low_varied 0.090106 0.30018 0.74 0.90 0.99
## Residual 0.091176 0.30195
## Number of obs: 299, groups: playerId, 20
##
## Fixed effects:
## Estimate Std. Error t value
## (Intercept) 1.4776686 0.0469319 31.485
## name_orderlow_first -0.0062019 0.0407450 -0.152
## seriesNo -0.0007815 0.0203413 -0.038
## episodeNo -0.1324553 0.0130942 -10.116
##
## Correlation of Fixed Effects:
## (Intr) nm_rd_ serisN
## nm_rdrlw_fr -0.382
## seriesNo -0.396 0.252
## episodeNo -0.436 0.005 0.033
## convergence code: 0
## boundary (singular) fit: see ?isSingularPilot 3 (shape first, then color; second dimension varies)
## Automatically converting the following non-factors to factors: name_order, seriesNo, shape_or_color
## `stat_bindot()` using `bins = 30`. Pick better value with `binwidth`.
## Automatically converting the following non-factors to factors: name_order, episodeNo, shape_or_color## `geom_smooth()` using method = 'loess' and formula 'y ~ x'
## `geom_smooth()` using method = 'loess' and formula 'y ~ x'
## `geom_smooth()` using method = 'loess' and formula 'y ~ x'
## Linear mixed model fit by REML ['lmerMod']
## Formula: uniMove ~ ruleId + episodeNo + (ruleId | playerId)
## Data: .
##
## REML criterion at convergence: 84.1
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.8403 -0.5017 -0.0579 0.3031 4.6342
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## playerId (Intercept) 0.03924 0.1981
## ruleIdshape_naming_low_varied 0.04682 0.2164 -0.68
## Residual 0.07031 0.2652
## Number of obs: 165, groups: playerId, 20
##
## Fixed effects:
## Estimate Std. Error t value
## (Intercept) 1.53759 0.05968 25.766
## ruleIdshape_naming_low_varied 0.01529 0.06398 0.239
## episodeNo -0.13728 0.01651 -8.313
##
## Correlation of Fixed Effects:
## (Intr) rlI___
## rlIdshp_n__ -0.621
## episodeNo -0.463 0.044## Linear mixed model fit by REML ['lmerMod']
## Formula: uniMove ~ ruleId + episodeNo + (ruleId | playerId)
## Data: .
##
## REML criterion at convergence: 57.4
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.4698 -0.5925 -0.0974 0.3433 4.2470
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## playerId (Intercept) 0.05239 0.2289
## ruleIdcolor_naming_low_varied 0.04484 0.2118 -0.91
## Residual 0.06092 0.2468
## Number of obs: 163, groups: playerId, 20
##
## Fixed effects:
## Estimate Std. Error t value
## (Intercept) 1.46659 0.06322 23.198
## ruleIdcolor_naming_low_varied -0.02609 0.06129 -0.426
## episodeNo -0.13616 0.01599 -8.514
##
## Correlation of Fixed Effects:
## (Intr) rlI___
## rlIdclr_n__ -0.763
## episodeNo -0.395 0.000## boundary (singular) fit: see ?isSingular## Linear mixed model fit by REML ['lmerMod']
## Formula: uniMove ~ 1 + name_order + seriesNo + episodeNo + ((1 | playerId) +
## (0 + ruleId | playerId))
## Data: .
##
## REML criterion at convergence: 111
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -3.3060 -0.5473 -0.0782 0.3485 4.8993
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## playerId (Intercept) 0.00000 0.0000
## playerId.1 ruleIdcolor_naming_high_varied 0.05221 0.2285
## ruleIdcolor_naming_low_varied 0.01317 0.1148 0.40
## ruleIdshape_naming_high_varied 0.04266 0.2065 0.88 0.71
## ruleIdshape_naming_low_varied 0.02557 0.1599 0.37 0.77 0.42
## Residual 0.06301 0.2510
## Number of obs: 328, groups: playerId, 20
##
## Fixed effects:
## Estimate Std. Error t value
## (Intercept) 1.573374 0.056839 27.681
## name_orderlow_first -0.006522 0.049524 -0.132
## seriesNo -0.048468 0.014280 -3.394
## episodeNo -0.135835 0.011203 -12.125
##
## Correlation of Fixed Effects:
## (Intr) nm_rd_ serisN
## nm_rdrlw_fr -0.597
## seriesNo -0.312 -0.235
## episodeNo -0.329 0.022 0.016
## convergence code: 0
## boundary (singular) fit: see ?isSingularPilot 2 and 3 combined (looking at main effect of nameability)
##
## high_first low_first
## 318 309## fixed-effect model matrix is rank deficient so dropping 1 column / coefficient## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: uniMove ~ 1 + name_order + ruleId + seriesNo + episodeNo + (0 +
## ruleId | playerId)
## Data: transcriptsByRule
##
## REML criterion at convergence: 316.7
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.8581 -0.5601 -0.1414 0.3409 6.2728
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## playerId ruleIdcolor_naming_high_varied 0.03449 0.1857
## ruleIdcolor_naming_low_varied 0.01321 0.1149 0.51
## ruleIdshape_naming_high_varied 0.03634 0.1906 0.89 0.85
## ruleIdshape_naming_low_varied 0.05587 0.2364 0.43 0.83 0.73
## Residual 0.07703 0.2776
## Number of obs: 627, groups: playerId, 40
##
## Fixed effects:
## Estimate Std. Error df t value
## (Intercept) 1.513265 0.043225 76.155699 35.009
## name_orderlow_first 0.082944 0.048383 41.767266 1.714
## ruleIdcolor_naming_low_varied -0.124137 0.040277 45.292656 -3.082
## ruleIdshape_naming_high_varied 0.036733 0.034495 69.294691 1.065
## seriesNo -0.017670 0.011328 211.158825 -1.560
## episodeNo -0.133068 0.008652 560.686879 -15.379
## Pr(>|t|)
## (Intercept) < 2e-16 ***
## name_orderlow_first 0.09388 .
## ruleIdcolor_naming_low_varied 0.00349 **
## ruleIdshape_naming_high_varied 0.29063
## seriesNo 0.12028
## episodeNo < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) nm_rd_ rlIdc___ rlIds___ serisN
## nm_rdrlw_fr -0.440
## rlIdclr_n__ -0.103 -0.589
## rlIdshp_n__ -0.402 0.558 -0.111
## seriesNo -0.414 0.019 -0.023 0.032
## episodeNo -0.318 0.011 -0.016 -0.005 0.027
## fit warnings:
## fixed-effect model matrix is rank deficient so dropping 1 column / coefficient## GVIF Df GVIF^(1/(2*Df))
## name_order 2.455584 1 1.567030
## ruleId 2.458603 2 1.252195
## seriesNo 1.002370 1 1.001184
## episodeNo 1.001090 1 1.000545##
## 0 1 2 3
## 161 152 157 157##
## 0 1 2 3 4 5 6
## 160 160 146 104 43 9 5##
## color_naming_high_varied color_naming_low_varied shape_naming_high_varied
## 157 154 161
## shape_naming_low_varied
## 155##
## high_first low_first
## 318 309Pilot 4: high shape / low color vs high color/ low shape (disjunction)
## Joining, by = c("episodeId", "seriesNo")## Automatically converting the following non-factors to factors: ruleId, seriesNo, shape_or_color
## `stat_bindot()` using `bins = 30`. Pick better value with `binwidth`.
## Automatically converting the following non-factors to factors: ruleId, seriesNo, shape_or_color, episodeNo## `geom_smooth()` using method = 'loess' and formula 'y ~ x'
## `geom_smooth()` using method = 'loess' and formula 'y ~ x'
## Linear mixed model fit by REML ['lmerMod']
## Formula: uniMove ~ ruleId + episodeNo + (1 | playerId)
## Data: .
##
## REML criterion at convergence: 110
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.6655 -0.6678 -0.1285 0.5694 3.0361
##
## Random effects:
## Groups Name Variance Std.Dev.
## playerId (Intercept) 0.04406 0.2099
## Residual 0.14805 0.3848
## Number of obs: 92, groups: playerId, 17
##
## Fixed effects:
## Estimate Std. Error t value
## (Intercept) 1.74740 0.11620 15.038
## ruleIddisjunct_shape_high -0.05955 0.13234 -0.450
## episodeNo -0.11129 0.02340 -4.756
##
## Correlation of Fixed Effects:
## (Intr) rlId__
## rlIddsjnc__ -0.692
## episodeNo -0.514 0.091## Linear mixed model fit by REML ['lmerMod']
## Formula: uniMove ~ ruleId + episodeNo + (1 | playerId)
## Data: .
##
## REML criterion at convergence: 107
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.7943 -0.6220 -0.2391 0.5907 2.4669
##
## Random effects:
## Groups Name Variance Std.Dev.
## playerId (Intercept) 0.03423 0.1850
## Residual 0.12775 0.3574
## Number of obs: 104, groups: playerId, 17
##
## Fixed effects:
## Estimate Std. Error t value
## (Intercept) 1.570739 0.086996 18.055
## ruleIddisjunct_shape_high -0.005311 0.116942 -0.045
## episodeNo -0.059484 0.016208 -3.670
##
## Correlation of Fixed Effects:
## (Intr) rlId__
## rlIddsjnc__ -0.556
## episodeNo -0.494 -0.013## Linear mixed model fit by REML ['lmerMod']
## Formula: uniMove ~ 1 + shape_or_color + seriesNo + episodeNo + (1 | playerId)
## Data: .
##
## REML criterion at convergence: 210.2
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.0193 -0.7232 -0.1406 0.5762 3.3404
##
## Random effects:
## Groups Name Variance Std.Dev.
## playerId (Intercept) 0.03574 0.1891
## Residual 0.14226 0.3772
## Number of obs: 196, groups: playerId, 17
##
## Fixed effects:
## Estimate Std. Error t value
## (Intercept) 1.632542 0.093121 17.531
## shape_or_colorshape_high -0.013378 0.108449 -0.123
## seriesNo -0.002094 0.054878 -0.038
## episodeNo -0.075450 0.013558 -5.565
##
## Correlation of Fixed Effects:
## (Intr) shp___ serisN
## shp_r_clrs_ -0.684
## seriesNo -0.254 -0.018
## episodeNo -0.354 0.037 -0.117Look at pilot 4 disjunction guesses
Looking at prop of shape/color in bucket on basis of nameability by rule
## Joining, by = c("X.playerId", "ruleId", "shape_or_color", "seriesNo", "episodeNo", "bucketId")## Automatically converting the following non-factors to factors: shape_or_color## shape_or_color N color_prop color_prop_norm sd se ci
## 1 color_high 124 0.4565860 0.2810386 0.5647630 0.05071721 0.10039162
## 2 shape_high 152 0.1378289 0.2810386 0.3007001 0.02439000 0.04818973## Automatically converting the following non-factors to factors: shape_or_color## shape_or_color N shape_prop shape_prop_norm sd se ci
## 1 color_high 124 0.09986559 0.211715 0.2360919 0.02120168 0.04196742
## 2 shape_high 152 0.30296053 0.211715 0.4724550 0.03832116 0.07571491Pilot 4 (nameability disjunction) plots
## Automatically converting the following non-factors to factors: shape_or_color, episodeNo
## Automatically converting the following non-factors to factors: shape_or_color, episodeNo## `geom_smooth()` using method = 'loess' and formula 'y ~ x'
## `geom_smooth()` using method = 'loess' and formula 'y ~ x'
disjunction models
Averaged across buckets so we have use per episode per ppt
color use
## boundary (singular) fit: see ?isSingular## Linear mixed model fit by REML ['lmerMod']
## Formula: color_prop ~ shape_or_color + episodeNo + (1 + shape_or_color |
## X.playerId)
## Data: .
##
## REML criterion at convergence: -56.2
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.24043 -0.63212 -0.03231 0.48469 2.58162
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## X.playerId (Intercept) 0.04075 0.2019
## shape_or_colorshape_high 0.04075 0.2019 -1.00
## Residual 0.02274 0.1508
## Number of obs: 92, groups: X.playerId, 17
##
## Fixed effects:
## Estimate Std. Error t value
## (Intercept) 0.436646 0.083087 5.255
## shape_or_colorshape_high -0.343076 0.082805 -4.143
## episodeNo 0.022779 0.009024 2.524
##
## Correlation of Fixed Effects:
## (Intr) shp___
## shp_r_clrs_ -0.940
## episodeNo -0.273 0.043
## convergence code: 0
## boundary (singular) fit: see ?isSingularshape use
Can’t add random intercepts for shape/color nameability with current sample size (non-convergence)
## Linear mixed model fit by REML ['lmerMod']
## Formula: shape_prop ~ shape_or_color + episodeNo + (1 | X.playerId)
## Data: .
##
## REML criterion at convergence: -50
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.3189 -0.5631 -0.1865 0.4815 2.9256
##
## Random effects:
## Groups Name Variance Std.Dev.
## X.playerId (Intercept) 0.02248 0.1499
## Residual 0.02091 0.1446
## Number of obs: 92, groups: X.playerId, 17
##
## Fixed effects:
## Estimate Std. Error t value
## (Intercept) 0.1080106 0.0650973 1.659
## shape_or_colorshape_high 0.2231435 0.0801948 2.783
## episodeNo -0.0007647 0.0089287 -0.086
##
## Correlation of Fixed Effects:
## (Intr) shp___
## shp_r_clrs_ -0.734
## episodeNo -0.346 0.056