Error Bars Do NOT Match Multiple Linear Regression Model

Question

I am trying to fit a linear mixed model using the lme4 package. Every model I try says that 2a and 1a are not significant.

I am curious as to why the error bars do not match what I am seeing in the lmer output. Specifically, why would 2a and 1a have no significant difference when there is no overlap between the error bars.

I am not allowed to embed a picture yet so here is graph