Table 1.
Estimate | Std. Error | z value | Pr(>∣z∣) | 2.5% CI | 97.5% CI | |
---|---|---|---|---|---|---|
Intercept | −0.690 | 0.173 | −3.979 | <0.0001 | −1.030 | −0.350 |
Age (in years) | 0.251 | 0.020 | 12.805 | <0.0001 | 0.212 | 0.289 |
Est. drawing frequency | −0.062 | 0.173 | −0.356 | 0.721 | −0.402 | 0.278 |
Avg. tracing rating | 0.267 | 0.020 | 13.529 | <0.0001 | 0.229 | 0.306 |
Time spent drawing | 0.039 | 0.021 | 1.868 | 0.062 | −0.002 | 0.079 |
Ink used | −0.031 | 0.020 | −1.546 | 0.122 | −0.070 | 0.008 |
Number of strokes | 0.008 | 0.018 | 0.477 | 0.634 | −0.026 | 0.043 |
Age × Drawing frequency | 0.017 | 0.017 | 1.042 | 0.298 | −0.015 | 0.050 |
Results of a generalized linear mixed-effect model predicting the recognizability of each drawing (i.e., binary classification scores) from VGG-19 FC6 classifications, including random intercepts for each category and participant. All predictors were z-scored prior to analysis such that coefficients are standardized and comparable. All significance tests are Wald significance tests based on the coefficient values; these tests are two-tailed. No adjustments were made for multiple comparisons. See Methods for further model specifications.