0.1 Load Database

# stimulus_schedule = readxl::read_xlsx("/Users/robertoabreu/Documents/indiana/ams_task_data_082123/ams_task/imagePlaceList_5y_ram.xlsx")
# a = a %>%f
#   left_join(stimulus_schedule, by = "image")

#a = read.csv("/Volumes/gunderson_lab/projects/manynumbers/data/mn_amstask/lf_mn_ams_task_2023-09-12_9pm.csv")
a = read.csv("2024_ucd_pilot/csv_files/processed_data/ams/lf_files/lf_mn_ams_2024-07-08.csv")
a$participant = paste("ucd", a$participant, sep="")
b = read.csv("2024_lafayette_pilot/csv_files/processed_data/ams/lf_files/lf_mn_ams_2024-04-17.csv")
c = read.csv("2024_iu_pilot/csv_files/processed_data/ams/lf_files/lf_mn_ams_2024-05-17.csv")
a = rbind(a,b,c)

a$participant = as.factor(as.character(a$participant))
#incomplete = c("020624LB","020624LG")
#a = a[which(a$participant %!in% incomplete),]
#View(table(a$participant))

Path: Main/Studies/2023_manynumbers/2024_iu_pilot/scripts/mn_amstask_analyses_all_072524.Rmd

1.1 Data Cleaning - Trial Level

Anticipatory responses

df_acc=a
# df_acc$participant = paste(df_acc$participant, df_acc$date, df_acc$age, sep = "_")
# df_acc$participant = as.factor(as.character(df_acc$participant))
df_antcipatory=subset(df_acc, rt <= .25)
# unique(df_antcipatory$participant) #checks that all participants are included
kable(table(df_antcipatory$participant), table.attr = "style = \"color: white;\"")
Var1 Freq
36 0
37 0
38 0
39 0
40 0
41 0
42 0
43 0
44 0
45 0
46 0
47 0
48 0
49 0
50 0
51 0
53 0
54 0
55 0
56 0
57 0
59 0
60 0
61 0
mn001 0
mn002 0
mn003 0
mn004 0
mn005 0
mn006 0
mn007 0
mn008 0
mn009 0
mn010 0
mn011 0
mn012 0
mn013 0
mn014 0
mn015 0
mn017 0
mn018 0
mn019 0
mn020 0
mn022 0
ucd1 0
ucd10 0
ucd12 0
ucd13 0
ucd15 0
ucd16 0
ucd17 0
ucd18 0
ucd19 0
ucd2 0
ucd22 0
ucd25 0
ucd26 0
ucd27 0
ucd28 0
ucd3 0
ucd31 0
ucd32 0
ucd33 0
ucd34 0
ucd35 0
ucd36 0
ucd37 0
ucd38 2
ucd39 0
ucd4 0
ucd40 0
ucd41 0
ucd42 0
ucd43 0
ucd44 0
ucd45 0
ucd46 1
ucd47 0
ucd48 0
ucd49 0
ucd5 0
ucd50 0
ucd52 0
ucd53 0
ucd54 0
ucd55 0
ucd56 0
ucd6 0
ucd7 0
ucd9 0 
#table(dataset2$participant, dataset2$Block)

Slow responses

#check for missing values
length(which(is.na(df_acc$rt)))
## [1] 0
#Recodes missing trials (NA) to 99
df_acc$rt[is.na(df_acc$rt)] <- 99
#
na_trials=subset(df_acc, rt == 99)
#Select trials that had rt longer than 0.25 sec
#df_acc=subset(df_acc, rt > .25)
#Recodes missing trials from 99 to NA
df_acc["rt"]=replace(df_acc$rt, df_acc$rt==99, NA)
#Subset only valid (nonNA) trials
df_acc_na<-subset(df_acc, (!is.na(df_acc[,"rt"])))
#Calculate means and standard deviations by participants
summary_participant <- summarySE(df_acc_na, measurevar="rt", groupvars=c("participant"))
#Calculate upper(+3SD) lower (-3SD) thresholds
summary_participant["sd3_sup"]=summary_participant$rt+summary_participant$sd*3
summary_participant["sd3_inf"]=summary_participant$rt-summary_participant$sd*3
kable(summary_participant,table.attr = "style = \"color: white;\"")
participant N rt sd se ci sd3_sup sd3_inf
36 60 4.039 3.2424 0.4186 0.8376 13.766 -5.6880
37 54 5.555 2.7179 0.3699 0.7418 13.708 -2.5990
38 54 3.620 1.4395 0.1959 0.3929 7.939 -0.6982
39 54 2.943 0.8065 0.1098 0.2201 5.363 0.5236
40 42 6.614 5.5200 0.8518 1.7202 23.174 -9.9455
41 54 9.696 7.1394 0.9715 1.9487 31.114 -11.7221
42 54 4.606 1.9955 0.2716 0.5447 10.593 -1.3804
43 54 2.821 1.0816 0.1472 0.2952 6.066 -0.4241
44 54 3.682 2.0963 0.2853 0.5722 9.971 -2.6073
45 60 2.900 2.4016 0.3100 0.6204 10.104 -4.3051
46 54 4.840 3.1033 0.4223 0.8470 14.150 -4.4697
47 54 2.450 1.5775 0.2147 0.4306 7.182 -2.2824
48 54 7.210 2.2847 0.3109 0.6236 14.064 0.3558
49 54 3.479 4.1156 0.5601 1.1233 15.826 -8.8679
50 60 3.753 2.4262 0.3132 0.6268 11.032 -3.5255
51 54 3.807 2.2750 0.3096 0.6210 10.632 -3.0183
53 54 3.541 2.1515 0.2928 0.5872 9.996 -2.9130
54 54 2.987 2.3884 0.3250 0.6519 10.152 -4.1782
55 54 2.911 1.0707 0.1457 0.2922 6.123 -0.3008
56 60 9.156 5.8324 0.7530 1.5067 26.653 -8.3414
57 54 4.161 1.6321 0.2221 0.4455 9.057 -0.7353
59 54 3.898 2.4110 0.3281 0.6581 11.131 -3.3351
60 54 3.398 0.9006 0.1225 0.2458 6.100 0.6966
61 54 4.877 5.1208 0.6969 1.3977 20.239 -10.4859
mn001 54 3.988 1.0858 0.1478 0.2964 7.245 0.7301
mn002 54 3.469 0.8974 0.1221 0.2449 6.161 0.7770
mn003 54 3.316 0.8633 0.1175 0.2356 5.906 0.7263
mn004 54 3.130 1.7554 0.2389 0.4791 8.396 -2.1359
mn005 54 4.008 2.4306 0.3308 0.6634 11.300 -3.2841
mn006 60 3.259 1.6742 0.2161 0.4325 8.282 -1.7634
mn007 60 4.629 5.1674 0.6671 1.3349 20.131 -10.8729
mn008 54 3.947 2.0169 0.2745 0.5505 9.998 -2.1039
mn009 54 4.250 2.3682 0.3223 0.6464 11.354 -2.8549
mn010 54 2.598 1.3983 0.1903 0.3817 6.793 -1.5970
mn011 54 2.699 1.4024 0.1908 0.3828 6.906 -1.5080
mn012 21 4.487 2.6577 0.5800 1.2098 12.460 -3.4862
mn013 54 5.068 4.6588 0.6340 1.2716 19.044 -8.9083
mn014 60 6.815 8.0413 1.0381 2.0773 30.939 -17.3085
mn015 54 3.617 1.0894 0.1483 0.2974 6.885 0.3487
mn017 60 4.732 3.6043 0.4653 0.9311 15.545 -6.0812
mn018 54 4.136 1.8123 0.2466 0.4947 9.573 -1.3009
mn019 60 3.775 3.8433 0.4962 0.9928 15.305 -7.7551
mn020 60 3.899 3.7828 0.4884 0.9772 15.247 -7.4497
mn022 54 2.007 1.2005 0.1634 0.3277 5.609 -1.5941
ucd1 60 5.247 5.6619 0.7309 1.4626 22.232 -11.7390
ucd10 54 3.202 1.1679 0.1589 0.3188 6.705 -0.3020
ucd12 54 3.085 1.4231 0.1937 0.3884 7.354 -1.1845
ucd13 54 2.341 1.1121 0.1513 0.3035 5.678 -0.9947
ucd15 54 2.556 1.4481 0.1971 0.3953 6.900 -1.7888
ucd16 54 4.500 3.8298 0.5212 1.0453 15.989 -6.9896
ucd17 60 3.380 2.1161 0.2732 0.5466 9.729 -2.9679
ucd18 54 4.892 4.2891 0.5837 1.1707 17.759 -7.9758
ucd19 54 3.399 1.9435 0.2645 0.5305 9.230 -2.4313
ucd2 54 2.178 0.8275 0.1126 0.2259 4.660 -0.3046
ucd22 54 3.304 2.0108 0.2736 0.5488 9.336 -2.7286
ucd25 54 2.774 1.0185 0.1386 0.2780 5.830 -0.2811
ucd26 54 3.470 1.4278 0.1943 0.3897 7.753 -0.8138
ucd27 54 4.485 1.7072 0.2323 0.4660 9.607 -0.6365
ucd28 54 1.880 1.4089 0.1917 0.3845 6.106 -2.3469
ucd3 54 3.554 1.8304 0.2491 0.4996 9.045 -1.9370
ucd31 54 5.141 2.8923 0.3936 0.7894 13.818 -3.5353
ucd32 54 3.117 1.9648 0.2674 0.5363 9.011 -2.7778
ucd33 54 4.529 4.2353 0.5764 1.1560 17.235 -8.1768
ucd34 54 4.181 1.9274 0.2623 0.5261 9.963 -1.6012
ucd35 54 2.541 2.0064 0.2730 0.5476 8.560 -3.4781
ucd36 54 2.881 1.2715 0.1730 0.3471 6.695 -0.9339
ucd37 54 3.060 1.8170 0.2473 0.4960 8.511 -2.3915
ucd38 54 4.152 3.2810 0.4465 0.8955 13.995 -5.6911
ucd39 54 3.782 1.3574 0.1847 0.3705 7.854 -0.2904
ucd4 54 3.712 2.4195 0.3292 0.6604 10.971 -3.5463
ucd40 54 2.736 1.1943 0.1625 0.3260 6.319 -0.8466
ucd41 54 2.763 1.4532 0.1978 0.3967 7.122 -1.5970
ucd42 54 2.461 1.5777 0.2147 0.4306 7.194 -2.2716
ucd43 54 3.079 1.4928 0.2031 0.4075 7.557 -1.3996
ucd44 54 2.966 1.2830 0.1746 0.3502 6.815 -0.8830
ucd45 54 3.112 2.2976 0.3127 0.6271 10.005 -3.7806
ucd46 54 1.912 1.8757 0.2552 0.5120 7.539 -3.7150
ucd47 54 3.895 2.7919 0.3799 0.7621 12.271 -4.4808
ucd48 54 2.192 1.1465 0.1560 0.3129 5.631 -1.2477
ucd49 54 2.274 1.2146 0.1653 0.3315 5.918 -1.3695
ucd5 54 2.219 0.9241 0.1257 0.2522 4.991 -0.5530
ucd50 54 3.092 1.2568 0.1710 0.3430 6.863 -0.6782
ucd52 54 3.048 2.3615 0.3214 0.6446 10.133 -4.0361
ucd53 54 2.597 1.4646 0.1993 0.3998 6.991 -1.7969
ucd54 54 3.062 2.0072 0.2731 0.5478 9.084 -2.9593
ucd55 54 2.624 1.6737 0.2278 0.4568 7.645 -2.3968
ucd56 54 3.460 2.3735 0.3230 0.6478 10.580 -3.6606
ucd6 54 4.690 4.0743 0.5544 1.1121 16.913 -7.5332
ucd7 54 3.140 1.5168 0.2064 0.4140 7.690 -1.4106
ucd9 54 3.543 1.0636 0.1447 0.2903 6.734 0.3523 
#
df_acc["rt_2"]=df_acc["rt"]
#Recodes missing trials from 99 to NA
df_acc$rt[is.na(df_acc$rt)] <- 99

#Replace values +-3SD for NAs
for(i in 1:nrow(df_acc)){
  dummy=match(df_acc$participant[i], summary_participant$participant)
  if(df_acc[i,"rt"] > summary_participant[dummy,"sd3_sup"]){
    df_acc[i,"rt"] = NA 
  } else {
    df_acc[i,"rt"] = df_acc[i,"rt"]
  }
}

#| df_acc[i,"rt"] < summary_participant[dummy,"sd2_inf"] #to exclude lower limit
#Excludes outliers from RT
df_acc_noout=df_acc
df_acc_noout=df_acc[-which(is.na(df_acc[,"rt"] & df_acc[,"choice"]!="")),]
kable(table(df_acc_noout$participant), table.attr = "style = \"color: white;\"")
Var1 Freq
36 58
37 53
38 52
39 54
40 41
41 52
42 54
43 51
44 53
45 59
46 52
47 52
48 53
49 52
50 59
51 53
53 52
54 53
55 53
56 58
57 52
59 53
60 53
61 53
mn001 52
mn002 54
mn003 53
mn004 53
mn005 52
mn006 59
mn007 59
mn008 53
mn009 53
mn010 52
mn011 52
mn012 20
mn013 52
mn014 58
mn015 53
mn017 59
mn018 53
mn019 59
mn020 58
mn022 52
ucd1 58
ucd10 53
ucd12 53
ucd13 53
ucd15 52
ucd16 53
ucd17 58
ucd18 54
ucd19 54
ucd2 53
ucd22 52
ucd25 53
ucd26 54
ucd27 53
ucd28 53
ucd3 52
ucd31 54
ucd32 54
ucd33 52
ucd34 53
ucd35 53
ucd36 53
ucd37 54
ucd38 53
ucd39 52
ucd4 53
ucd40 52
ucd41 53
ucd42 52
ucd43 53
ucd44 54
ucd45 53
ucd46 52
ucd47 53
ucd48 52
ucd49 53
ucd5 53
ucd50 54
ucd52 53
ucd53 53
ucd54 53
ucd55 53
ucd56 53
ucd6 52
ucd7 53
ucd9 53 
#trials by participant
too_slow_trials=subset(df_acc, is.na(df_acc$rt) & df_acc$choice!="")
correct_trials=subset(df_acc, df_acc$accuracy==1 & !is.na(df_acc$rt))
incorrect_trials=subset(df_acc, df_acc$accuracy==0 & !is.na(df_acc$rt) )
trials=as.data.frame(cbind(table(na_trials$participant),table(df_antcipatory$participant),table(too_slow_trials$participant),table(correct_trials$participant),table(incorrect_trials$participant)))
trials["total"]= rowSums(trials[c(1,3,4,5)])
trials["valid_trials"]=trials$V1+trials$V4+trials$V5
names(trials)=c("missed","anticipatory","too_slow","valid_correct","valid_incorrect",
                "total","valid_total")
kable(trials, table.attr = "style = \"color: white;\"")
missed anticipatory too_slow valid_correct valid_incorrect total valid_total
36 0 0 2 30 28 60 58
37 0 0 1 25 28 54 53
38 0 0 2 31 21 54 52
39 0 0 0 38 16 54 54
40 0 0 1 22 19 42 41
41 0 0 2 32 20 54 52
42 0 0 0 39 15 54 54
43 0 0 3 34 17 54 51
44 0 0 1 39 14 54 53
45 0 0 1 33 26 60 59
46 0 0 2 39 13 54 52
47 0 0 2 30 22 54 52
48 0 0 1 25 28 54 53
49 0 0 2 21 31 54 52
50 0 0 1 26 33 60 59
51 0 0 1 33 20 54 53
53 0 0 2 38 14 54 52
54 0 0 1 31 22 54 53
55 0 0 1 40 13 54 53
56 0 0 2 29 29 60 58
57 0 0 2 41 11 54 52
59 0 0 1 33 20 54 53
60 0 0 1 44 9 54 53
61 0 0 1 45 8 54 53
mn001 0 0 2 36 16 54 52
mn002 0 0 0 47 7 54 54
mn003 0 0 1 41 12 54 53
mn004 0 0 1 47 6 54 53
mn005 0 0 2 42 10 54 52
mn006 0 0 1 22 37 60 59
mn007 0 0 1 21 38 60 59
mn008 0 0 1 41 12 54 53
mn009 0 0 1 25 28 54 53
mn010 0 0 2 36 16 54 52
mn011 0 0 2 26 26 54 52
mn012 0 0 1 14 6 21 20
mn013 0 0 2 29 23 54 52
mn014 0 0 2 25 33 60 58
mn015 0 0 1 24 29 54 53
mn017 0 0 1 24 35 60 59
mn018 0 0 1 30 23 54 53
mn019 0 0 1 30 29 60 59
mn020 0 0 2 24 34 60 58
mn022 0 0 2 27 25 54 52
ucd1 0 0 2 33 25 60 58
ucd10 0 0 1 22 31 54 53
ucd12 0 0 1 30 23 54 53
ucd13 0 0 1 38 15 54 53
ucd15 0 0 2 35 17 54 52
ucd16 0 0 1 40 13 54 53
ucd17 0 0 2 31 27 60 58
ucd18 0 0 0 26 28 54 54
ucd19 0 0 0 27 27 54 54
ucd2 0 0 1 39 14 54 53
ucd22 0 0 2 40 12 54 52
ucd25 0 0 1 29 24 54 53
ucd26 0 0 0 29 25 54 54
ucd27 0 0 1 29 24 54 53
ucd28 0 0 1 39 14 54 53
ucd3 0 0 2 29 23 54 52
ucd31 0 0 0 34 20 54 54
ucd32 0 0 0 36 18 54 54
ucd33 0 0 2 37 15 54 52
ucd34 0 0 1 44 9 54 53
ucd35 0 0 1 24 29 54 53
ucd36 0 0 1 24 29 54 53
ucd37 0 0 0 32 22 54 54
ucd38 0 2 1 32 21 54 53
ucd39 0 0 2 37 15 54 52
ucd4 0 0 1 41 12 54 53
ucd40 0 0 2 33 19 54 52
ucd41 0 0 1 25 28 54 53
ucd42 0 0 2 28 24 54 52
ucd43 0 0 1 32 21 54 53
ucd44 0 0 0 27 27 54 54
ucd45 0 0 1 30 23 54 53
ucd46 0 1 2 32 20 54 52
ucd47 0 0 1 43 10 54 53
ucd48 0 0 2 33 19 54 52
ucd49 0 0 1 35 18 54 53
ucd5 0 0 1 25 28 54 53
ucd50 0 0 0 29 25 54 54
ucd52 0 0 1 30 23 54 53
ucd53 0 0 1 46 7 54 53
ucd54 0 0 1 40 13 54 53
ucd55 0 0 1 35 18 54 53
ucd56 0 0 1 39 14 54 53
ucd6 0 0 2 34 18 54 52
ucd7 0 0 1 37 16 54 53
ucd9 0 0 1 29 24 54 53 
View(trials)


#incomplete = c("120123BM", "13124EW","020624LB","020623LG") First participant is actually complete




a2 = separate(df_acc_noout, image, into = c("age2","trialtype"), sep = "_", remove = F)
## Warning: Expected 2 pieces. Additional pieces discarded in 4199 rows [7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, ...].
a2[which(a2$trialtype == "rect.png"), "trialtype"] <- NA
a3 =a2
participant_list = unique(a3$date)
for (x in 1:length(participant_list)) {
  #a3[which(a3$date == participant_list[x] & is.na(a3$trialtype)),"age"] = unique(a3[which(a3$date == participant_list[x] & !is.na(a3$trialtype)),"age"])
  a3[which(a3$date == participant_list[x] & is.na(a3$trialtype)),"trialtype"] = "practice"
}
length(unique(a3$participant))
## [1] 90
df_acc_noout = a3
df_acc_noout =df_acc

2.1 General Analyses

Practice trials

df_acc_noout_practice = subset(df_acc_noout, block == "practice")
df_acc_noout_practice$practice_block = ifelse(df_acc_noout_practice$X <10, "block1","block2")
table(df_acc_noout_practice$participant)
## 
##    36    37    38    39    40    41    42    43    44    45    46    47    48    49    50    51    53    54    55    56    57    59    60    61 mn001 mn002 mn003 mn004 mn005 mn006 mn007 mn008 mn009 mn010 mn011 mn012 mn013 mn014 mn015 mn017 mn018 mn019 mn020 mn022  ucd1 ucd10 ucd12 ucd13 ucd15 ucd16 ucd17 ucd18 ucd19  ucd2 ucd22 ucd25 ucd26 ucd27 ucd28  ucd3 ucd31 ucd32 ucd33 ucd34 ucd35 ucd36 ucd37 ucd38 ucd39  ucd4 ucd40 ucd41 ucd42 ucd43 ucd44 ucd45 ucd46 ucd47 ucd48 ucd49  ucd5 ucd50 ucd52 ucd53 ucd54 ucd55 ucd56  ucd6  ucd7  ucd9 
##    12     6     6     6     6     6     6     6     6    12     6     6     6     6    12     6     6     6     6    12     6     6     6     6     6     6     6     6     6    12    12     6     6     6     6    12     6    12     6    12     6    12    12     6    12     6     6     6     6     6    12     6     6     6     6     6     6     6     6     6     6     6     6     6     6     6     6     6     6     6     6     6     6     6     6     6     6     6     6     6     6     6     6     6     6     6     6     6     6     6
agg_ams_task_practice = aggregate(accuracy ~  participant*numRatio*age*practice_block, df_acc_noout_practice, mean)
descriptive_amstask_practice = summarySE(agg_ams_task_practice, "accuracy", c("age","numRatio","practice_block"))
kable(descriptive_amstask_practice, table.attr = "style = \"color: white;\"")
age numRatio practice_block N accuracy sd se ci
3 3.5 block1 32 0.7083 0.2504 0.0443 0.0903
3 3.5 block2 10 0.5333 0.2811 0.0889 0.2011
3 4.0 block1 32 0.6458 0.3161 0.0559 0.1140
3 4.0 block2 10 0.4667 0.3583 0.1133 0.2563
4 3.5 block1 41 0.8374 0.2251 0.0352 0.0711
4 3.5 block2 3 0.4444 0.1925 0.1111 0.4781
4 4.0 block1 41 0.8618 0.1969 0.0307 0.0621
4 4.0 block2 3 0.7778 0.1925 0.1111 0.4781
5 3.5 block1 17 0.9412 0.1762 0.0427 0.0906
5 4.0 block1 17 1.0000 0.0000 0.0000 0.0000 
agg_ams_task_practice$numRatio = as.factor(agg_ams_task_practice$numRatio)
agg_ams_task_practice$numRatio <- factor(agg_ams_task_practice$numRatio, levels=c('4', '3.5'))
agg_ams_task_practice$type = "Practice Trials"

agg_ams_task_practice$age = as.factor(agg_ams_task_practice$age)

ams_ratio_graph_practice = ggplot(agg_ams_task_practice, aes(x = interaction(practice_block), y = accuracy)) +
  #geom_paired_raincloud(aes(fill = as.factor(numRatio))) +
  # geom_bar(stat = "identity", data = descriptive_ams_task,
  #          fill = NA, color = "#f03b20", size = 1, width = 0.55) +
  ylab("Accuracy")+
  geom_hline(yintercept = 0.5, linetype ="dashed")+
  stat_summary(fun.data = data_summary, geom = "errorbar",
               position = position_dodge(width = 0.10), width = 0.10, colour = "black", size =.75)+
stat_summary(fun.y = "mean", geom = "point", size = 2.5, shape = 23, colour = "black", aes(fill = age))+
  geom_line(aes(group = interaction (participant)), color = "grey",
            alpha = 0.4,
            size = .5) +
  scale_fill_manual(values = c("#0186C7","#E96A01","#FED602"))+
  geom_point(aes(group = interaction (participant), color = participant),
             alpha = 1.5,
             size = .85, shape = 20) +
  facet_grid(age~numRatio)+
  theme_bw()+
  scale_y_continuous(trans = shift_trans(0))+
  theme(legend.position="none",
        axis.title.x=element_text(size=size_text),
        axis.text.x =  element_text(size=size_text),
        #axis.title.x =  element_blank(),
        panel.grid.major = element_blank(), panel.grid.minor = element_blank(),
        panel.background = element_rect(fill = "white", colour = "grey50"),
        strip.background =element_rect(fill="#f0f0f0"),
        strip.text = element_text(size = size_text),
        axis.text.y =  element_text(size=size_text),
        axis.title.y =  element_text(size=size_text),
        legend.text=element_text(size=size_text))+
  xlab("Ratio")
## Warning: Using `size` aesthetic for lines was deprecated in ggplot2 3.4.0.
## ℹ Please use `linewidth` instead.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was generated.
## Warning: The `fun.y` argument of `stat_summary()` is deprecated as of ggplot2 3.3.0.
## ℹ Please use the `fun` argument instead.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was generated.
#ams_ratio_graph_practice
##        
##         block1 block2
##   36         2      2
##   37         0      0
##   38         0      0
##   39         0      0
##   40         0      0
##   41         0      0
##   42         0      0
##   43         0      0
##   44         0      0
##   45         2      2
##   46         0      0
##   47         0      0
##   48         0      0
##   49         0      0
##   50         2      2
##   51         0      0
##   53         0      0
##   54         0      0
##   55         0      0
##   56         2      2
##   57         0      0
##   59         0      0
##   60         0      0
##   61         0      0
##   mn001      0      0
##   mn002      0      0
##   mn003      0      0
##   mn004      0      0
##   mn005      0      0
##   mn006      2      2
##   mn007      2      2
##   mn008      0      0
##   mn009      0      0
##   mn010      0      0
##   mn011      0      0
##   mn012      2      2
##   mn013      0      0
##   mn014      2      2
##   mn015      0      0
##   mn017      2      2
##   mn018      0      0
##   mn019      2      2
##   mn020      2      2
##   mn022      0      0
##   ucd1       2      2
##   ucd10      0      0
##   ucd12      0      0
##   ucd13      0      0
##   ucd15      0      0
##   ucd16      0      0
##   ucd17      2      2
##   ucd18      0      0
##   ucd19      0      0
##   ucd2       0      0
##   ucd22      0      0
##   ucd25      0      0
##   ucd26      0      0
##   ucd27      0      0
##   ucd28      0      0
##   ucd3       0      0
##   ucd31      0      0
##   ucd32      0      0
##   ucd33      0      0
##   ucd34      0      0
##   ucd35      0      0
##   ucd36      0      0
##   ucd37      0      0
##   ucd38      0      0
##   ucd39      0      0
##   ucd4       0      0
##   ucd40      0      0
##   ucd41      0      0
##   ucd42      0      0
##   ucd43      0      0
##   ucd44      0      0
##   ucd45      0      0
##   ucd46      0      0
##   ucd47      0      0
##   ucd48      0      0
##   ucd49      0      0
##   ucd5       0      0
##   ucd50      0      0
##   ucd52      0      0
##   ucd53      0      0
##   ucd54      0      0
##   ucd55      0      0
##   ucd56      0      0
##   ucd6       0      0
##   ucd7       0      0
##   ucd9       0      0
##   age practice_block  N accuracy      sd      se      ci
## 1   3         block1 10   0.4167 0.08784 0.02778 0.06284
## 2   3         block2 10   0.5000 0.28328 0.08958 0.20265
## 3   4         block1  3   0.4444 0.09623 0.05556 0.23904
## 4   4         block2  3   0.6111 0.09623 0.05556 0.23904
## 
##  Paired t-test
## 
## data:  accuracy by practice_block
## t = -1.5, df = 12, p-value = 0.2
## alternative hypothesis: true mean difference is not equal to 0
## 95 percent confidence interval:
##  -0.25385  0.04872
## sample estimates:
## mean difference 
##         -0.1026

Experimental trials

All trials

Analyses
df_acc_noout_experimental = subset(df_acc_noout, block != "practice")
df_acc_noout_experimental$age = as.factor(as.character(df_acc_noout_experimental$age))
distinct(df_acc_noout_experimental[c("participant","age")]) %>%
  {table(.$age)}
## 
##  3  4  5 
## 32 41 17
ratio_condition_model <- glmer(accuracy ~ 
                                          +  age  * numRatio * block+ 
                                          + (1|participant), data = df_acc_noout_experimental, family = binomial, control = glmerControl(optimizer = "bobyqa"))
summary(ratio_condition_model)
## Generalized linear mixed model fit by maximum likelihood (Laplace Approximation) ['glmerMod']
##  Family: binomial  ( logit )
## Formula: accuracy ~ +age * numRatio * block + +(1 | participant)
##    Data: df_acc_noout_experimental
## Control: glmerControl(optimizer = "bobyqa")
## 
##      AIC      BIC   logLik deviance df.resid 
##     5619     5740    -2790     5581     4250 
## 
## Scaled residuals: 
##    Min     1Q Median     3Q    Max 
## -2.907 -1.044  0.606  0.849  1.262 
## 
## Random effects:
##  Groups      Name        Variance Std.Dev.
##  participant (Intercept) 0.132    0.363   
## Number of obs: 4269, groups:  participant, 90
## 
## Fixed effects:
##                           Estimate Std. Error z value Pr(>|z|)   
## (Intercept)                0.00224    0.38305    0.01   0.9953   
## age4                      -0.24065    0.50091   -0.48   0.6309   
## age5                      -1.10374    0.79378   -1.39   0.1644   
## numRatio                   0.12240    0.16285    0.75   0.4523   
## blockblock2               -0.10470    0.53384   -0.20   0.8445   
## blockblock3               -0.16441    0.53398   -0.31   0.7582   
## age4:numRatio              0.27295    0.23642    1.15   0.2483   
## age5:numRatio              1.30303    0.50412    2.58   0.0097 **
## age4:blockblock2           0.12637    0.69753    0.18   0.8562   
## age5:blockblock2          -0.77204    1.14709   -0.67   0.5009   
## age4:blockblock3          -0.03023    0.69684   -0.04   0.9654   
## age5:blockblock3          -0.22688    1.10619   -0.21   0.8375   
## numRatio:blockblock2      -0.06268    0.23032   -0.27   0.7855   
## numRatio:blockblock3      -0.06931    0.23035   -0.30   0.7635   
## age4:numRatio:blockblock2  0.04316    0.33411    0.13   0.8972   
## age5:numRatio:blockblock2  0.71719    0.74789    0.96   0.3376   
## age4:numRatio:blockblock3  0.05426    0.33332    0.16   0.8707   
## age5:numRatio:blockblock3  0.25190    0.70896    0.36   0.7224   
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Correlation matrix not shown by default, as p = 18 > 12.
## Use print(x, correlation=TRUE)  or
##     vcov(x)        if you need it
em=emtrends(ratio_condition_model, pairwise ~ age|block, var="numRatio", mult.name = "age")
summary(em, infer=c(TRUE,TRUE), null=0, type = "response", adjust = "none")
## $emtrends
## block = block1:
##  age numRatio.trend    SE  df asymp.LCL asymp.UCL z.ratio p.value
##  3           0.1224 0.163 Inf   -0.1968     0.442   0.752  0.4523
##  4           0.3954 0.172 Inf    0.0593     0.731   2.306  0.0211
##  5           1.4254 0.477 Inf    0.4905     2.360   2.988  0.0028
## 
## block = block2:
##  age numRatio.trend    SE  df asymp.LCL asymp.UCL z.ratio p.value
##  3           0.0597 0.163 Inf   -0.2595     0.379   0.367  0.7139
##  4           0.3758 0.171 Inf    0.0405     0.711   2.197  0.0280
##  5           2.0799 0.528 Inf    1.0449     3.115   3.938  0.0001
## 
## block = block3:
##  age numRatio.trend    SE  df asymp.LCL asymp.UCL z.ratio p.value
##  3           0.0531 0.163 Inf   -0.2662     0.372   0.326  0.7445
##  4           0.3803 0.170 Inf    0.0481     0.713   2.244  0.0249
##  5           1.6080 0.471 Inf    0.6845     2.531   3.413  0.0006
## 
## Confidence level used: 0.95 
## 
## $contrasts
## block = block1:
##  contrast    estimate    SE  df asymp.LCL asymp.UCL z.ratio p.value
##  age3 - age4   -0.273 0.236 Inf    -0.736    0.1904  -1.155  0.2483
##  age3 - age5   -1.303 0.504 Inf    -2.291   -0.3150  -2.585  0.0097
##  age4 - age5   -1.030 0.507 Inf    -2.023   -0.0366  -2.032  0.0421
## 
## block = block2:
##  contrast    estimate    SE  df asymp.LCL asymp.UCL z.ratio p.value
##  age3 - age4   -0.316 0.236 Inf    -0.779    0.1468  -1.338  0.1808
##  age3 - age5   -2.020 0.553 Inf    -3.103   -0.9370  -3.655  0.0003
##  age4 - age5   -1.704 0.555 Inf    -2.792   -0.6161  -3.070  0.0021
## 
## block = block3:
##  contrast    estimate    SE  df asymp.LCL asymp.UCL z.ratio p.value
##  age3 - age4   -0.327 0.235 Inf    -0.788    0.1335  -1.392  0.1640
##  age3 - age5   -1.555 0.499 Inf    -2.532   -0.5777  -3.119  0.0018
##  age4 - age5   -1.228 0.501 Inf    -2.209   -0.2464  -2.452  0.0142
## 
## Confidence level used: 0.95
emeans=emmeans(ratio_condition_model, pairwise ~ age|block|numRatio, mult.name = "age", at=list(numRatio=c(1.5,2,2.5,3)))
summary(emeans, infer=c(TRUE,TRUE), null=0, type = "response", adjust = "none")
## $emmeans
## block = block1, numRatio = 1.5:
##  age  prob     SE  df asymp.LCL asymp.UCL null z.ratio p.value
##  3   0.546 0.0410 Inf     0.466     0.625  0.5   1.123  0.2613
##  4   0.588 0.0270 Inf     0.534     0.639  0.5   3.184  0.0015
##  5   0.738 0.0328 Inf     0.669     0.797  0.5   6.107  <.0001
## 
## block = block2, numRatio = 1.5:
##  age  prob     SE  df asymp.LCL asymp.UCL null z.ratio p.value
##  3   0.497 0.0414 Inf     0.416     0.577  0.5  -0.078  0.9380
##  4   0.586 0.0270 Inf     0.532     0.638  0.5   3.117  0.0018
##  5   0.758 0.0327 Inf     0.688     0.816  0.5   6.402  <.0001
## 
## block = block3, numRatio = 1.5:
##  age  prob     SE  df asymp.LCL asymp.UCL null z.ratio p.value
##  3   0.479 0.0414 Inf     0.400     0.560  0.5  -0.498  0.6185
##  4   0.534 0.0277 Inf     0.480     0.588  0.5   1.235  0.2169
##  5   0.715 0.0342 Inf     0.643     0.777  0.5   5.484  <.0001
## 
## block = block1, numRatio = 2.0:
##  age  prob     SE  df asymp.LCL asymp.UCL null z.ratio p.value
##  3   0.561 0.0292 Inf     0.504     0.618  0.5   2.081  0.0375
##  4   0.635 0.0245 Inf     0.585     0.681  0.5   5.226  <.0001
##  5   0.852 0.0409 Inf     0.753     0.916  0.5   5.400  <.0001
## 
## block = block2, numRatio = 2.0:
##  age  prob     SE  df asymp.LCL asymp.UCL null z.ratio p.value
##  3   0.504 0.0298 Inf     0.446     0.562  0.5   0.143  0.8865
##  4   0.631 0.0246 Inf     0.581     0.677  0.5   5.073  <.0001
##  5   0.899 0.0333 Inf     0.812     0.948  0.5   5.964  <.0001
## 
## block = block3, numRatio = 2.0:
##  age  prob     SE  df asymp.LCL asymp.UCL null z.ratio p.value
##  3   0.486 0.0297 Inf     0.428     0.544  0.5  -0.470  0.6383
##  4   0.581 0.0255 Inf     0.531     0.630  0.5   3.132  0.0017
##  5   0.849 0.0412 Inf     0.749     0.913  0.5   5.374  <.0001
## 
## block = block1, numRatio = 2.5:
##  age  prob     SE  df asymp.LCL asymp.UCL null z.ratio p.value
##  3   0.577 0.0290 Inf     0.519     0.632  0.5   2.597  0.0094
##  4   0.679 0.0342 Inf     0.609     0.742  0.5   4.778  <.0001
##  5   0.921 0.0393 Inf     0.802     0.971  0.5   4.534  <.0001
## 
## block = block2, numRatio = 2.5:
##  age  prob     SE  df asymp.LCL asymp.UCL null z.ratio p.value
##  3   0.512 0.0297 Inf     0.454     0.570  0.5   0.393  0.6940
##  4   0.673 0.0344 Inf     0.603     0.737  0.5   4.620  <.0001
##  5   0.962 0.0226 Inf     0.883     0.988  0.5   5.259  <.0001
## 
## block = block3, numRatio = 2.5:
##  age  prob     SE  df asymp.LCL asymp.UCL null z.ratio p.value
##  3   0.493 0.0298 Inf     0.435     0.551  0.5  -0.247  0.8047
##  4   0.627 0.0361 Inf     0.553     0.694  0.5   3.352  0.0008
##  5   0.926 0.0368 Inf     0.814     0.973  0.5   4.705  <.0001
## 
## block = block1, numRatio = 3.0:
##  age  prob     SE  df asymp.LCL asymp.UCL null z.ratio p.value
##  3   0.591 0.0400 Inf     0.511     0.667  0.5   2.234  0.0255
##  4   0.721 0.0463 Inf     0.622     0.802  0.5   4.124  <.0001
##  5   0.960 0.0298 Inf     0.840     0.991  0.5   4.104  <.0001
## 
## block = block2, numRatio = 3.0:
##  age  prob     SE  df asymp.LCL asymp.UCL null z.ratio p.value
##  3   0.519 0.0414 Inf     0.438     0.599  0.5   0.463  0.6434
##  4   0.713 0.0469 Inf     0.613     0.796  0.5   3.976  0.0001
##  5   0.986 0.0119 Inf     0.928     0.997  0.5   4.901  <.0001
## 
## block = block3, numRatio = 3.0:
##  age  prob     SE  df asymp.LCL asymp.UCL null z.ratio p.value
##  3   0.499 0.0414 Inf     0.419     0.580  0.5  -0.017  0.9861
##  4   0.670 0.0500 Inf     0.566     0.760  0.5   3.130  0.0017
##  5   0.966 0.0255 Inf     0.862     0.992  0.5   4.354  <.0001
## 
## Confidence level used: 0.95 
## Intervals are back-transformed from the logit scale 
## Tests are performed on the logit scale 
## 
## $contrasts
## block = block1, numRatio = 1.5:
##  contrast    odds.ratio     SE  df asymp.LCL asymp.UCL null z.ratio p.value
##  age3 / age4     0.8447 0.1684 Inf    0.5715    1.2486    1  -0.847  0.3972
##  age3 / age5     0.4271 0.1012 Inf    0.2684    0.6795    1  -3.590  0.0003
##  age4 / age5     0.5056 0.1026 Inf    0.3397    0.7526    1  -3.361  0.0008
## 
## block = block2, numRatio = 1.5:
##  contrast    odds.ratio     SE  df asymp.LCL asymp.UCL null z.ratio p.value
##  age3 / age4     0.6978 0.1392 Inf    0.4719    1.0317    1  -1.804  0.0713
##  age3 / age5     0.3152 0.0767 Inf    0.1956    0.5078    1  -4.745  <.0001
##  age4 / age5     0.4517 0.0949 Inf    0.2992    0.6820    1  -3.781  0.0002
## 
## block = block3, numRatio = 1.5:
##  contrast    odds.ratio     SE  df asymp.LCL asymp.UCL null z.ratio p.value
##  age3 / age4     0.8026 0.1602 Inf    0.5427    1.1868    1  -1.102  0.2705
##  age3 / age5     0.3672 0.0866 Inf    0.2314    0.5828    1  -4.250  <.0001
##  age4 / age5     0.4576 0.0921 Inf    0.3085    0.6787    1  -3.886  0.0001
## 
## block = block1, numRatio = 2:
##  contrast    odds.ratio     SE  df asymp.LCL asymp.UCL null z.ratio p.value
##  age3 / age4     0.7369 0.1171 Inf    0.5397    1.0062    1  -1.921  0.0548
##  age3 / age5     0.2226 0.0768 Inf    0.1132    0.4377    1  -4.355  <.0001
##  age4 / age5     0.3021 0.1029 Inf    0.1549    0.5890    1  -3.514  0.0004
## 
## block = block2, numRatio = 2:
##  contrast    odds.ratio     SE  df asymp.LCL asymp.UCL null z.ratio p.value
##  age3 / age4     0.5957 0.0947 Inf    0.4362    0.8136    1  -3.257  0.0011
##  age3 / age5     0.1148 0.0442 Inf    0.0540    0.2439    1  -5.628  <.0001
##  age4 / age5     0.1927 0.0733 Inf    0.0914    0.4063    1  -4.327  <.0001
## 
## block = block3, numRatio = 2:
##  contrast    odds.ratio     SE  df asymp.LCL asymp.UCL null z.ratio p.value
##  age3 / age4     0.6814 0.1080 Inf    0.4995    0.9297    1  -2.420  0.0155
##  age3 / age5     0.1688 0.0577 Inf    0.0863    0.3299    1  -5.202  <.0001
##  age4 / age5     0.2477 0.0835 Inf    0.1279    0.4796    1  -4.138  <.0001
## 
## block = block1, numRatio = 2.5:
##  contrast    odds.ratio     SE  df asymp.LCL asymp.UCL null z.ratio p.value
##  age3 / age4     0.6429 0.1265 Inf    0.4372    0.9454    1  -2.245  0.0248
##  age3 / age5     0.1160 0.0645 Inf    0.0390    0.3449    1  -3.875  0.0001
##  age4 / age5     0.1805 0.1020 Inf    0.0596    0.5463    1  -3.030  0.0024
## 
## block = block2, numRatio = 2.5:
##  contrast    odds.ratio     SE  df asymp.LCL asymp.UCL null z.ratio p.value
##  age3 / age4     0.5086 0.1000 Inf    0.3460    0.7477    1  -3.439  0.0006
##  age3 / age5     0.0418 0.0261 Inf    0.0123    0.1420    1  -5.088  <.0001
##  age4 / age5     0.0822 0.0520 Inf    0.0238    0.2837    1  -3.953  0.0001
## 
## block = block3, numRatio = 2.5:
##  contrast    odds.ratio     SE  df asymp.LCL asymp.UCL null z.ratio p.value
##  age3 / age4     0.5786 0.1128 Inf    0.3948    0.8479    1  -2.806  0.0050
##  age3 / age5     0.0776 0.0427 Inf    0.0264    0.2280    1  -4.647  <.0001
##  age4 / age5     0.1340 0.0749 Inf    0.0448    0.4008    1  -3.596  0.0003
## 
## block = block1, numRatio = 3:
##  contrast    odds.ratio     SE  df asymp.LCL asymp.UCL null z.ratio p.value
##  age3 / age4     0.5609 0.1588 Inf    0.3221    0.9768    1  -2.043  0.0411
##  age3 / age5     0.0605 0.0478 Inf    0.0128    0.2851    1  -3.546  0.0004
##  age4 / age5     0.1078 0.0870 Inf    0.0222    0.5243    1  -2.760  0.0058
## 
## block = block2, numRatio = 3:
##  contrast    odds.ratio     SE  df asymp.LCL asymp.UCL null z.ratio p.value
##  age3 / age4     0.4343 0.1228 Inf    0.2495    0.7558    1  -2.951  0.0032
##  age3 / age5     0.0152 0.0135 Inf    0.0027    0.0863    1  -4.728  <.0001
##  age4 / age5     0.0351 0.0315 Inf    0.0060    0.2042    1  -3.727  0.0002
## 
## block = block3, numRatio = 3:
##  contrast    odds.ratio     SE  df asymp.LCL asymp.UCL null z.ratio p.value
##  age3 / age4     0.4913 0.1377 Inf    0.2836    0.8509    1  -2.536  0.0112
##  age3 / age5     0.0356 0.0279 Inf    0.0077    0.1653    1  -4.259  <.0001
##  age4 / age5     0.0726 0.0579 Inf    0.0152    0.3465    1  -3.289  0.0010
## 
## Confidence level used: 0.95 
## Intervals are back-transformed from the log odds ratio scale 
## Tests are performed on the log odds ratio scale
plot(ggpredict(ratio_condition_model, terms=c("numRatio","age","block")))

Figure
age numRatio N accuracy sd se ci
3 1.50 32 0.5208 0.1680 0.0297 0.0606
3 2.00 32 0.5156 0.1562 0.0276 0.0563
3 2.50 32 0.5339 0.1568 0.0277 0.0565
3 3.00 32 0.5417 0.1720 0.0304 0.0620
4 1.25 41 0.5440 0.1455 0.0227 0.0459
4 1.50 41 0.5630 0.1204 0.0188 0.0380
4 2.00 41 0.6220 0.1526 0.0238 0.0482
4 2.50 41 0.6491 0.1942 0.0303 0.0613
5 1.17 17 0.6127 0.1348 0.0327 0.0693
5 1.25 17 0.6569 0.1611 0.0391 0.0828
5 1.50 17 0.7059 0.1617 0.0392 0.0831
5 2.00 17 0.8676 0.1956 0.0474 0.1006 
## `geom_line()`: Each group consists of only one observation.
## ℹ Do you need to adjust the group aesthetic?
## `geom_line()`: Each group consists of only one observation.
## ℹ Do you need to adjust the group aesthetic?

Age Effects

Figure
df_acc_noout_experimental_age = subset(df_acc_noout_experimental, numRatio== 1.5 | numRatio== 2)
df_acc_noout_experimental_age_overall = aggregate(accuracy ~ participant*age, df_acc_noout_experimental_age, mean)
df_acc_noout_experimental_age_overall$task = "AMS Task - Age Effects"
ams_age = ggplot(df_acc_noout_experimental_age_overall, aes(x = as.factor(age), y = accuracy, group = as.factor(age), color =as.factor(age))) +
  geom_bar(stat = "identity", data = summarySE(df_acc_noout_experimental_age, "accuracy", c("age")),
                                 aes(fill = as.factor(age) ), color = "black", size = .75, width = 0.55) +
   geom_dotplot(binaxis = "y", stackdir = "center", fill ="#737373", color = NA, dotsize =.5)+
  # geom_point(aes(group = interaction (participant)),
  #           alpha = 1,
  #           size = .75, colour = "#737373") +
  #geom_abline(slope = 1, intercept = 0, color = "#615e62", linetype = "dashed") + 
  stat_summary(fun.data = data_summary, geom = "errorbar", width=.01, size =0.5, color = "black")+
  #stat_summary(fun.y = "mean", geom = "point", size = 2, colour = "black", shape = 21, alpha =.85)+
  ylab("Accuracy")+
  geom_hline(yintercept = 0.5, linetype ="dashed")+  
  theme_bw()+
  scale_fill_manual(values = c("#0186C7","#E96A01","#FED602"))+
  scale_y_continuous(breaks=seq(0, 1, .1), limits=c(0,1.01),trans = shift_trans(0), expand = c(0,0))+
  theme(legend.position="none",
        axis.title.x=element_blank(),
        axis.text.x =  element_text(size=size_text),
        #axis.title.x =  element_text(size = size_text),
        panel.grid.major = element_blank(), panel.grid.minor = element_blank(),
        panel.background = element_rect(fill = "white", colour = "grey50"),
        strip.background =element_rect(fill="#f0f0f0"),
        strip.text = element_text(size = size_text),
        axis.text.y =  element_text(size=size_text),
        axis.title.y =  element_text(size=size_text),
        legend.text=element_text(size=size_text)) +
  facet_grid(task~.)
  #scale_y_continuous(breaks=seq(0, 1, .1), limits=c(0,1.01))
ams_age
## Bin width defaults to 1/30 of the range of the data. Pick better value with `binwidth`.