studyforrest

Fresh preprint out! Enning combines #NLP algorithm and #naturalistic #movie watching. >53,000 time points. Core insight: The default network dominates neural responses to evolving movie stories biorxiv.org/content/10.110… @Mila_Quebec @McGillBME @mcgillu @yaledailynews @dartmouth

We used data from 14 participants who watched Forrest Gump while their gaze was tracked (studyforrest.org; thanks @eknahm and colleagues). We determined phases of relatively high/low heart rate and used them as indicators of high/low arousal.

We used the amazing @studyforrest dataset and an in-lab Raiders dataset to assess our model. These two datasets were collected with different movies, different participants, and different scanners and scan parameters, allowing us to evaluate model performance in diverse settings.

Psst... We'll be at the @OHBM in Glasgow next month, and fired up a laser cutter to bring hand-crafted goodies - we just need more puns! So here's the deal: The DataLad-related pun with most likes underneath this posts wins a custom DataLad pen with their own message. Let's go!🥳

@datalad @OHBM Run, DataLad, run!

Announcing the new datalad.org website! Take a tour with our data minions and find out how DataLad can help you manage your data.

FAIRly big: A framework for computationally reproducible processing of large-scale data biorxiv.org/cgi/content/sh… #bioRxiv

But does the framework generalize? To demonstrate its potential for transparent sharing we adapted it to open data (@studyforrest) and open source software (#fmriprep). Thanks to @ReproNim and @yarikoptic for readily providing it in a container collection!

7 years after the first data publication of the project there is now an independent extension with full-length MEG data! A studyforrest extension, MEG recordings while watching the audio-visual movie “Forrest Gump” biorxiv.org/content/10.110…

This is crazy cool! I did not have a close look at the data yet, but the possibilities are huge! Finally data to explore the temporal domain, and maybe even play with fusion approaches. I am so glad that the authors did this, and took the open path. Thank you very much!

Initially published 7 years ago, the @studyforrest dataset is the 2nd most active dataset on @OpenNeuroOrg today (check it out on openneuro.org). Thanks y'all for the continued interest. Every day, the time spent on it in the past becomes more meaningful!

Just published in #JOSS_theOJ: 'DataLad: distributed system for joint management of code, data, and their relationship' doi.org/10.21105/joss.…

@aina_puce @eknahm @yarikoptic @g_kiar @OHBM I feel like everyone should leave an emoji-appreciation on the PR that did the magic, as a little bit of feedback on how much this has improved all of our experiences: github.com/sparkletown/sp…

Fantastic use of the dataset!

And I hope that it will add a "Tell me how to contribute to your data!"

If you have computed other features for the stimulus, please consider sharing them, or share the code to generate them. Thanks! 5/5

The latter are like md5sums for the pixel content of a movie frame (also available), but their bitwise distances are interpretable as "magnitude of visual difference". This information can be used to quantify the visual "load" of the stimulus (e.g. to detect cuts). 4/5

Thanks to @PsyInfOlaf low-level perceptual confound annotations are now available for the Forrest Gump movie stimulus and its audio description. 1/5 #studyforresthackathon github.com/psychoinformat…