Overview
*********

.. article-info::
    :avatar: UCLA_Suit.jpg
    :avatar-link: https://www.decisionneurosciencelab.com/elijahgalvan
    :author: Elijah Galvan
    :date: September 1, 2023
    :read-time: 2 min read
    :class-container: sd-p-2 sd-outline-muted sd-rounded-1

Before Data Collection
==========

Compared to linear modeling approaches to data analysis, computational modeling requires much more planning and thoughtfulness before any data is collected. 
Typically, when conducting experiments whose data you intend to analyze using linear modeling (this includes t-tests, ANOVAs, simple regression, linear mixed effects modeling, and other such techniques), you must hone in on a design that allows you to gain some combination of experimental and statistical control over changes observed in the dependent variable; after this, you’re essentially free to conduct your experiment. 
This is only the start of the process for computational modeling: this first stage is certainly the most involved of the three outlined in this handbook and where most simple oversights that can be problematic during data analysis can be avoided or corrected.

Model Fitting
===========
After collecting data, we can fit and select a model. 
During this stage, we will fit free parameters for each subject, use the free parameters to compute model fit indices, and select a model to validate. 

Hypothesis Testing
============
Everything done prior to this has been in the service of ensuring that whatever conclusions we draw from our data are statistically valid and logically sound. 
Now is the fun and easy part really, we only have to now do really simple analyses in order to test whatever hypotheses we might have.