Can lmer be used for a longitudinal study? I have a longitudinal dataset where kids at 4,5, and 6 years of age came to lab and were trained with different pictures. We collected reaction times (DV) before and after the training (pre_post). Does the following model look okay?
model<-lmer(RT~pre_post*pictures*Age+ (1+Age|participant),data=data,REML=F)

Or is there other ways to handle longitudinal datasets in lmer?
 A: Yep, this is a classic textbook use case linear mixed models for repeated measures/longitudinal datasets.
The random-effects portion of your parameterization allows for varying intercepts and age slopes which makes sense.
The variable pre_post variable isn`t very clear to me. It is common to have a pre-test/procedure measurement, which makes sense, but the "post" part seems like a data leakage.
I wouldn't fit all interactions between main effects as well. I would only keep the ones important for the research to measure and test for the ones that appear to exist in the exploratory/inspective/descriptive part of the analysis.
A: Another way to parameterize it would be to fit a longitudinal ANCOVA with post-test RT as the outcome, and pre-test as a time-varying predictor. If your main interest is in tracking post-test RT as a function of age, this gives you a direct measure of the age effect at post test, while controlling for the pretest. This seems more straightforward and should be a more reliable way to produce estimates.
See this for more: http://hbiostat.org/bbr/md/change.html#whats-wrong-with-change-in-general
This approach will require you to change the structure of your dataset, unfortunately. The way you have it now, your DV is (both pre and post) RT, and then you've got the pre_post factor to tell the difference between pre and post. To run the revised model, you'd need your DV to be POST RTs only, and add a column for PRE RT. So in pseudo code:
newmod <- lmer(PostRT ~ PRE_RT + pictures*Age + (1+Age|participant), data=data, REML=F)
However, if, as in your comment, you would like to test the interaction (moderation, not mediation) between age and the RT difference, you'd need to keep it the way you have it.
Finally, do you think that the picture effect is fixed or random? Is your view that the pictures (stimuli) are a random sample from a population of possible pictures? If so, you might change your version so pictures represent a random effect:
model<-lmer(RT ~ pre_post*Age + (1+Age|participant) + (1|pictures), data=data, REML=F)
Also, for reference this is a great article from psycholinguistics that discusses general approaches to your situation: https://linkinghub.elsevier.com/retrieve/pii/S0749596X07001398
