[R-lang] Re: Mixed model for eyetracking data anlaysis

[Alex Fine]

Hi Jeonghwa,

Jeonghwa Shin wrote:
> Dear ling-r-lang users,
>
> I'm writing to get some advice on the use of logit mixed model for
> eyetracking data anlaysis. My experiment has three IVs with two levels for
> each - language (English vs.Korean),stress pattern (trochaic vs.  iambic),
> phonation type (aspirated vs.lax), and  one continuous IV, time. And the
> DV is binary, either 0 or 1. What I want in the test is where in the time
> course of word recognition, the trochaic and iambic words are different in
> their activation of the target words, and how they are interacting with
> the phonation types in the two language groups.
>
> For this, I first tried logit mixed effect model as below.
>
> lmer(gaze~stress*lg*phonation*time+(1|subj)+(1|item), data
> ,family=="binomial")
>
> The issue that I had with this model is that it doesn't show interactions
> between specific levels of factors. For example, I couldn't test whether
> English speakers' behavior for aspiraed trochaic words (default level) is
> different from the one for aspirated iambic ones.
>
> So I have made a dummy combinatorial variable column, "int", which
> combines the levels of lg, stress pattern, and phontion type (e.g.,
> "eiasp" as a combination of English spk's respose for aspirated iambic
> words) and ran the model as below:
>
> lmer(gaze~int*time+(1|subj)+(1|item), data, family=="binomial")
>
> My question is whether having such a dummy combinatorial variable is a
> legitimate for the mixed effect model. If it's not legitimate, I'd like to
> know what's the way to examine the interactions between specified levels
> of different factors of interest.
>   
Don't do this dummy variable thing you've described.  It'll be easier to 
explain interpreting the coefficients on the interactions if you (1) 
paste in the output of the first model you mention above and (2) paste 
in the "contrasts" for each of the categorical variables.  that is, in 
R, do contrasts(variablename) and paste in the output.  you have to know 
how the categorical variables are coded to interpret the coefficients 
(put another way, you can test specific hypotheses by carefully 
specifying the way the categorical variables are coded).  Once you do 
that, this question will be much easier to answer.
> My another question is how can we test where in the timecourse the two
> levels of interest are sigificantly different from each other (in terms of
> slope change). For this , I have segmented time into every 100ms window
> and treated the window as a factor. It looks the outcome supports slope
> change in plot (in logit) and compares two levels of interest in each time
> window. But again, I'm not sure whether this is a right way to examine the
> time effect. If not, what model or approach do I have to make?
>   
At least on the face of it, it seems reasonable to do what you've done, 
i.e. treat "time" as a predictor and see if it interacts with other 
predictors of interest.  But I think there are some papers that touch on 
this exact problem in eye-tracking if i'm not mistaken, so you probably 
want to figure out if there's a standard.  I'm thinking of papers like...

Barr D.J., Gann T.M., Pierce, R.S. (in press) Anticipatory baseline 
effects and information integration in visual world studies Acta 
Psychologica

...and all the relevant stuff cited there.  someone else on this list 
can give a better answer probably.

hth,
alex

> My questions might have been arisen by my misunderstanding of the model,
> so it would be greatly appreaciated if you would be able to give me your
> valuable advice.
>
> Thank you!
>
> Best regards,
> Jeonghwa Shin
>
>
>
>
>