Table 2.

Coefficients and t-statistics for age and learning mixed effects model.

Predictor	β	t
Age (linear)	−0.318	−94.6***
Age (quadratic)	−0.003	−11.7***
Education (1–7, some high school – PhD)	0.957	32.7***
Gender (Male = 1, Female = −1)	0.558	12.8***
C1: Fluid (1,1) vs. Crystallized (−1,−1)	−3.887	−83.5***
C2: Memory Match (1) vs. Memory Matrix (−1)	−0.934	−12.3***
C3: Raindrops (1) vs. Word Bubbles (−1)	−1.807	−35.6***
Session [t(1) = 0, t(25) = 1]	15.8	265.1***
C1 × Age (linear)	−0.079	−32.9***
C2 × Age (linear)	0.017	4.3***
C3 × Age (linear)	−0.063	−26***
C1 × Age (quadratic)	0.003	20.3***
C2 × Age (quadratic)	0.001	4.7***
C3 × Age (quadratic)	−0.0005	−2.7**
C1 × Education	−0.494	−28.9***
C2 × Education	−0.171	−6.3***
C3 × Education	0.235	12.5***
C1 × Gender	0.297	11.4***
C2 × Gender	−0.172	−4.1***
C3 × Gender	1.577	56.4***
C1 × Session	1.546	25.9***
C2 × Session	5.468	54.2***
C3 × Session	−0.106	−1.7
Session × Age (linear)	−0.099	−34.2***
Session × Age (quadratic)	0.001	4.8
C1 × Age (linear) × Session	−0.057	−19.5***
C2 × Age (linear) × Session	−0.006	−1.3
C3 × Age (linear) × Session	0.004	1.2
C1 × Age (quadratic) × Session	0.000	−1
C2 × Age (quadratic) × Session	−0.003	−8.0***
C3 × Age (quadratic) × Session	−0.001	−5.4***

The model was fit using the lmer function, part of the lme4 package in R. Significance values are based on highest posterior density intervals derived from 10000 Markov Chain Monte Carlo samples, using the pvals.fnc function in R's languageR package.

^*p < 0.01,

^**p < 0.001,

^***p < 0.0001.