[fabian]

Hi Irene,

Yes – it is possible to calculate motion-energy values from recordings of children-parent interactions.
However, as you mention yourself:
“if the constant moving around of the participants, or the shared use of the objects, contact between participants, ect. would pose a problem with the ROI area selection.”
These instances will most probably pose problems, because participants (here children and parents) cannot cross into each other’s regions. The ROIs have to belong to one and only one person. Sharing objects will thus also be a problem (because of potential ROI-crossing).
In other words: MEA may be applied, but you would have to make sure that instances of ROI-crossing are left out of the analysis (e.g. by trimming the video, or by deleting these sequences later in the rawdata).

Alternatively, you might consider a different approach based on something else than frame-differencing. (e.g. a machine-learning based approach such as OpenPose)

Best, Fabian

[fabian]

I would do it on the total sample.
Do you expect large differences between separate groups?

A good way to reach an empirically-based solution is to carefully look at the lagplot (and/or the heatmaps).
These graphical displays should give you a good insight into the decision on separate or group-level pre-processing.
Cheers, Fabian

[fabian]

lagsec = 5s and winsec = 10s
–> this may work, but you have to be aware that in such short windows, a lag of up to 5seconds is high (you are losing data at the beginning and end of your short window).
I suggest to lower the lag to 3s or even 2s
Just make sure to carefully inspect your results.

[fabian]

Dear Hervé,
Thanks for your message!
1) The best settings depend on your video-material.
For parent-infant interaction, I would suggest to not go higher than the established settings for psychotherapy (lag 5s; winsec 60s; incsec 60s):

lagsec=5 (you can check the setting of lagsec in the lagplot of rMEA: where real interactions and pseudo-interactions cross each other, this should be the maximum lag)

winsec=30 (how long are your sessions?) 30 or lower is probably better, as the turns of parent vs. infant are most probably much shorter than in psychotherapy

incsec=10 I would use overlapping windows, especially if you use winsec of 30 or lower.

2) Yes – we analyzed marital therapy. A simple way is to just use all possible combinations.

Cheers, Fabian

[fabian]

A simple solution is to read in (readMEA) two versions of your raw data:
One with ROI X (in Subj 1 and 2) and one with ROI Y (in Subj 1 and 2). Then you append the two datasets and assign them to two different groups.

From then on, you can run the analysis in the same “large” MEA-List.

For the bootstrap-data, I suggest to base the shuffled data on only one set of data (e.g. ROI X, or the combination of X and Y). Otherwise, you could end up with permutations of one subject’s ROI X with the same subject’s ROI Y. (This would most probably lead to a peak at lag-zero).

Best, Fabian

[fabian]

General answer:
Sometimes, the automated email may get stuck in your spam folder …
Please check there or send a direct inquiry to ramseyer – at – psync.ch

[fabian]

Hi,

the percentages represent the relative amount above the threshold.
How much of the time was there movement in your region?
Or: Percentage of movement for each subject.

The other variables provide the information of your analysis:
s1_% = your subject 1 movement-data: s1-percentage of movement
other_% = your other subject? movement-data: “other-percentage of movement
all_lags = overall synchrony of the interaction
s1_lead = synchrony-score for the side where s1 was the leader
other_lead = synchrony-score for the side where other was the leader
lag_zero = synchrony at lag0, i.e. without considering delayed (lagged) associations

Please refer to the rMEA documentation or see the accompanying info on OSF.

Best, Fabian

[fabian]

Hi gr220,

The email you should have gotten probably landed in your spam folder.
Sorry for that.
I have sent you another mail with the links.

Best, Fabian

–> contact me at ramseyer@psync.ch

[fabian]

Hi, I am not sure what you mean by “extract”?
If you want to access your MEA-data in R, you simply have to save the results of your MEA-analysis as a .txt-file. Then you point your R-package/analysis to this file/folder and from there, you’ll be able to run your statistical analyses.
This is really simple in rMEA: use readMEA and specify the parameters of your choice.
In CRQA, this will be probably similar, please refer to the information available for the CRQA-package.

Best, Fabian

[fabian]

rMEA works with the current version of R (3.5.x), and it should run on older versions too. Please refer to the rMEA documentation for further details.
A paper describing rMEA is coming soon. We will announce it here, once it is ready.

[fabian]

Hi Edith,
Yes, you need the program R.
It is freely available and runs on all popular platforms. I suggest using R-Studio, which has a very nice user interface.

Best, Fabian

[fabian]

The fix is now available: Version 4.03b

use the form once again to get automatic access to the download.

[fabian]

Some things take longer than anticipated …
The script I have been talking about is now available on CRAN:

https://CRAN.R-project.org/package=rMEA

Cheers, Fabian

[fabian]

Handling of ROIs is now very easily completed in the newly available R-package (rMEA).

[fabian]

MEA 4.0 is now provided as an install package.
This may help with previously found problems concerning Apple’s security settings.

[Author]

Posts