Study and script purpose

To examine visual patterns and engage in data exploration prior to statistical analysis.

  1. Do you get an overall effect of talker dialect (AAE vs GAE) across both not shared and shared social interests (think of this as a general dialect mismatch effect)
  2. Do you get an effect of dialect differences separately for not shared and shared?

##Load packages

Read and clean

Grab the accurate trials only (“respCode:target”)

see the accuracy Rmdfile for accuracy analysis)

#Early Exclusions There are participants that we need to exclude for the following reasons: DLD concerns DELV part II flagged for disorder(L03), fussing out (L06) and aged out of study (L10, turned 11 day of first visit), too young (less than age 6; l07, l11)

Bin the data

Split into trial halves

Answer followup questions

One challenge we needed to address was understanding how school-age children distribute attention between target and distractor images in the gaze data. To assess real-time learning, we considered whether children looked more at the target image.

Grand mean averaged across social interests and dialects heard

Social interest will be used to refer to the social conditions (shared or not shared) Dialects heard will be used to refer to the dialects (AAE and GAE). It’s a bit imprecise to call one unfamiliar or familiar so I always prefer a straightforward route. Interpretation linked to familiar/unfamiliar or matching/non-matching can be included later as a way to understand or explain the findings.

Does their attention overall change over the course of the experiment?

Yes, kids seem to improve over time, yeilding more targ vs. distractor looks prior to word offset. However at word offset, they no longer attend as much to the target image. # How do children’s attention to targ and distractor vary as a function of each condition (separately and interactively)?

Dialect Heard Overall

Does dialect effect change over time?

Visualized two different ways

# Includes a plot sequestered to the not-shared condition I think of this like a “pure” dialect effect not tainted by sharing interests

Social Effect

Does the social effect change over time?

In the not-shared condition, children look to the distractor more during the first half and look at the target after the word finishes. This seems to shift in the second half of the experiment in which the reverse is shown. Learning improves over the course of the task. Is this typical among children?

In the shared condition, children look to the target more in the first half and then this diminished over time. Is this a boost?

Showing a different visualization of this information above

Plotting targ and distractor looks for shared and not shared on the same plot. I’m also shifting to new colors (blue/green) has been used to visual targ and distractor. Now linetype is going to showcase AOI (targ vs. distractor).

THe colors will showcase conditions (dialect or social interests given the plot)

Social interests plotted differently

## `geom_smooth()` using formula = 'y ~ s(x, bs = "cs")'

How does this ineraction change over time?

Overall Interaction Dialect and Social (shown 2 ways)

  1. first plot shows targ vs distractor faceted by social conditions – vWPREs default settings
## Grand average calculated using Event means.
## Scale for colour is already present.
## Adding another scale for colour, which will replace the existing scale.

A different visualization on the same plot

## try to do the overlaid plots manually because VWPre was not cooperating

Interaction faceted by dialect heard

#Interaction faceted by social conditions #And further separated because you can’t really see the targ vs. distractor