impossibilities_pilot_analysis

1 preprocessing

2 plotting

2.1 binary

plot_probability_stacked(df_long_bi_prob)

combined_plot_with_title(filter(df_prob, batch == "binary"), filter(df_sur, batch == "binary"), plot_probability_means, plot_surprise_means, "participants (N = 16), binary batch")

Warning: Removed 336 rows containing non-finite outside the scale range
(`stat_summary()`).
Removed 336 rows containing non-finite outside the scale range
(`stat_summary()`).

Warning: Removed 312 rows containing non-finite outside the scale range
(`stat_summary()`).
Removed 312 rows containing non-finite outside the scale range
(`stat_summary()`).

2.2 continuous batch

combined_plot_with_title(filter(df_prob, batch == "continuous"), filter(df_sur, batch == "continuous"), plot_probability_means, plot_surprise_means, "participants (N = 16), binary batch")

Warning: Removed 328 rows containing non-finite outside the scale range
(`stat_summary()`).
Removed 328 rows containing non-finite outside the scale range
(`stat_summary()`).

Warning: Removed 312 rows containing non-finite outside the scale range
(`stat_summary()`).
Removed 312 rows containing non-finite outside the scale range
(`stat_summary()`).

2.3 combined

combined_plot_with_title(df_prob, df_sur, plot_probability_means_nofacet, plot_surprise_means_nofacet, "all participants (N = 32), combined batch")

Warning: Removed 664 rows containing non-finite outside the scale range
(`stat_summary()`).
Removed 664 rows containing non-finite outside the scale range
(`stat_summary()`).

Warning: Removed 624 rows containing non-finite outside the scale range
(`stat_summary()`).
Removed 624 rows containing non-finite outside the scale range
(`stat_summary()`).

combined_plot_with_title(df_prob, df_sur, plot_probability_means, plot_surprise_means, "all participants (N = 32), combined batch")

Warning: Removed 664 rows containing non-finite outside the scale range
(`stat_summary()`).

Warning: Removed 664 rows containing non-finite outside the scale range
(`stat_summary()`).

Warning: Removed 624 rows containing non-finite outside the scale range
(`stat_summary()`).
Removed 624 rows containing non-finite outside the scale range
(`stat_summary()`).

2.4 compare binary vs continuous question asking method

combined_plot_with_title(filter(df_prob, batch == "binary"), filter(df_prob, batch == "continuous"), plot_probability_means, plot_probability_means, "Top: binary (yes, no), Bottom: continuous (0-100%)")

Warning: Removed 336 rows containing non-finite outside the scale range
(`stat_summary()`).
Removed 336 rows containing non-finite outside the scale range
(`stat_summary()`).

Warning: Removed 328 rows containing non-finite outside the scale range
(`stat_summary()`).
Removed 328 rows containing non-finite outside the scale range
(`stat_summary()`).

3 model

# all subs
## probability/impossibility
m_prob_all <- lmerTest::lmer(response ~ impossible_degree + violation_type + (1|subject), df_prob)
m_prob_all_int <- lmerTest::lmer(response ~ impossible_degree * violation_type + (1|subject), df_prob)
anova(m_prob_all, m_prob_all_int)

refitting model(s) with ML (instead of REML)

Data: df_prob
Models:
m_prob_all: response ~ impossible_degree + violation_type + (1 | subject)
m_prob_all_int: response ~ impossible_degree * violation_type + (1 | subject)
               npar    AIC    BIC  logLik deviance  Chisq Df Pr(>Chisq)
m_prob_all       13 6521.0 6579.1 -3247.5   6495.0                     
m_prob_all_int   22 6527.8 6626.3 -3241.9   6483.8 11.128  9      0.267

anova(m_prob_all)

Type III Analysis of Variance Table with Satterthwaite's method
                  Sum Sq Mean Sq NumDF  DenDF F value    Pr(>F)    
impossible_degree  14478   14478     1 606.04  11.932 0.0005904 ***
violation_type    151778   16864     9 606.04  13.898 < 2.2e-16 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

plot(effects::allEffects(m_prob_all))

## surprise
m_sur_all <- lmerTest::lmer(response ~ impossible_degree + violation_type + (1|subject), df_sur)
m_sur_all_int <- lmerTest::lmer(response ~ impossible_degree * violation_type + (1|subject), df_sur)
anova(m_sur_all, m_sur_all_int)

refitting model(s) with ML (instead of REML)

Data: df_sur
Models:
m_sur_all: response ~ impossible_degree + violation_type + (1 | subject)
m_sur_all_int: response ~ impossible_degree * violation_type + (1 | subject)
              npar  AIC    BIC  logLik deviance  Chisq Df Pr(>Chisq)  
m_sur_all       13 5860 5917.7 -2917.0     5834                       
m_sur_all_int   22 5863 5960.6 -2909.5     5819 15.027  9     0.0902 .
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

anova(m_sur_all)

Type III Analysis of Variance Table with Satterthwaite's method
                  Sum Sq Mean Sq NumDF  DenDF F value    Pr(>F)    
impossible_degree  13746 13745.7     1 582.14  22.353 2.849e-06 ***
violation_type     88181  9797.9     9 582.10  15.933 < 2.2e-16 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

plot(effects::allEffects(m_sur_all))