search

jwdink / eyetrackingR

This package is designed to make dealing with eye-tracking data easier. It addresses tasks along the pipeline from raw data to analysis and visualization.
http://eyetrackingr.com
Other
82 stars 20 forks source link

Future enhancements #30

Open jwdink opened 8 years ago

jwdink commented 8 years ago

To avoid cluttering up issues forum, will put brainstorming for all future enhancements here.

jwdink commented 8 years ago

Better tools for analyzing "first looks" (e.g., first saccade to AOI, starting from fixation).

brockf commented 8 years ago

Add Bayesian time series analysis using brms() and growth curves.

jwdink commented 8 years ago

Doesn't this break our workflow? We don't have any functions that internally use lmer, would we be adding a function with brms? Or did you just mean have a vignette with brms? @brockf

brockf commented 8 years ago

True: Let's just add a vignette with BRMS using time_sequence data. The only brms-specific benefit is that we can get interpretable times of divergence (based on some threshold of distribution overlap), but this can be discussed in a vignette.

samhforbes commented 8 years ago

I actually just came on here to suggest exactly this (doing this at the moment). Divergence estimates using Bayesian analysis would give a more accessible output, and using brms would mean there is no need to change the syntax from your previous vignettes.

jwdink commented 8 years ago

@brockf , @sforbes88 Is the idea here to fit a bayesian growth-curve then use the posterior to estimate divergence?

samhforbes commented 8 years ago

Yeah, that's what I was thinking, because it's fairly easy to get prediction intervals from brms. Either way, great package guys.