#72589 | AsPredicted

'EmoWord'
(AsPredicted #72589)


Author(s)
This pre-registration is currently anonymous to enable blind peer-review.
It has 2 authors.
Pre-registered on
08/11/2021 03:47 PM (PT)

1) Have any data been collected for this study already?
No, no data have been collected for this study yet.

2) What's the main question being asked or hypothesis being tested in this study?
In this experiment, we look at 2- to 4-year-old children's ability to map emotion words to emotional facial expressions and body language. Participants will see two pictures in which two children display two different emotions (through both their faces and body language) and hear an audio that involves an emotion word (e.g., "Who is happy? Can you show me who is happy?"). The participants will be asked to point at the picture that matches the audio they hear. Four emotion words and their corresponding emotional facial expressions and body language are tested: happy, sad, angry, and scared. We predict that children as young as 2 or 3 years old are able to understand these emotion words and connect them to their corresponding emotional facial expressions and body language.

3) Describe the key dependent variable(s) specifying how they will be measured.
Whether a participant points to the target picture (score: 1) or the distractor (score: 0) on each trial

4) How many and which conditions will participants be assigned to?
There are 4 emotion category conditions with 6 images in each of the emotion categories. In total we will have 24 trials. Each child will partake in all 24 trials. On each trial, we will randomly select one image from one category and another image from another category, and randomly select an audio that matches one of the two images, yet we also add the constraint that each image and each audio will be selected evenly across the 24 trials.

5) Specify exactly which analyses you will conduct to examine the main question/hypothesis.
5.1 We will use a mixed-effects model to analyze our data: glmer (score ~ trial_type * age + (trial_type * emotion | subject))
(1) score: score (0 or 1) on each trial.
(2) trial_type: within-valence or cross-valence trial - sum coded (-.5 for within valence, +.5 for cross-valence).
(3) age: each child's age, centered for interpretability.
(4) emotion: target emotion on each trial.
(5) subject: subject id.
5.2 If the model fails to converge, we will simplify the random effects according to Langcog Lab standard operating procedure (first prune random interaction slopes, then random slopes, then intercepts).
5.3 We will interpret the intercept and the corresponding pvalue as evidence regarding whether the grand mean is above chance. We will interpret the estimates for trial_type and age and their pvalues as evidence regarding whether these factors have significant effects.

6) Describe exactly how outliers will be defined and handled, and your precise rule(s) for excluding observations.
We will exclude subjects if we observe the following: if a child has seen another kid complete the study, if the child is a non-English speaker, if there are technical issues related to online testing (e.g., slow internet speed), or if the overall accuracy is 3 standard deviations of the mean. We will also exclude data points on particular trials if there is parental or sibling interference. If there is parental or sibling interference on over ⅓ of the trials, we exclude data from the subject as a whole.

7) How many observations will be collected or what will determine sample size?
No need to justify decision, but be precise about exactly how the number will be determined.

We will collect N = 48 children (16 2-year-olds, 16 3-year-olds, and 16 4-year-olds).

8) Anything else you would like to pre-register?
(e.g., secondary analyses, variables collected for exploratory purposes, unusual analyses planned?)

At the end of the study we will be including a survey about children's exposure to other people wearing masks in the past year. We are curious to see if this has any effect on the children's performance on our task.