blog

Michael Hanke has published three articles about studyforrest – a neuroscience project involving the film Forrest Gump – on F1000Research. In this guest blog, he explains to us what the project is about and how it relates to the the multi-award winning film.  

What is studyforrest?

Studyforrest is an open science project that first started in 2013. Prior to its inception, there were no public brain imaging data that captured the breadth of human brain function in a single acquisition, rather than just one very specific aspect.  These data are needed in order to enable studies of inter-individual differences and similarities in how information is encoded in brain activity patterns under conditions that resemble real-life complexity. To meet this need, we wanted to create a resource for cognitive neuroscience research, and decided to use the motion picture Forrest Gump for a comparably rich and prolonged stimulation in the confined environment of an MRI scanner.

We hoped to prompt use and contributions of the resource by a interdisciplinary audience, by offering data with a versatility that are well beyond a typical neuroimaging study. With the help of Jörg Stadler and an increasing number of colleagues from Magdeburg University and other centers, we have continued to extend and enrich the dataset. The project’s data page offers a glimpse of where we are in this process. It is vital that we continue to deliver these resources in the most open and unrestricted way possible, to maximize the data’s utility for a wide and diverse audience.

Ultimate success of this project will be determined by our collective ability to integrate new findings and resources — contributed by independent groups and individuals — back into the original project in order to facilitate effective incremental research. There are great examples of such efforts in other fields; my favorite is the International Genetically Engineered Machine (iGEM) competition in synthetic biology. To this end, we are building software (datalad) to help deal with some of the technical aspects of such an effort, but even good tooling can only go so far. Ultimately, it will depend on the independent contributions of others.

We have already seen a number of diverse independent studies that were conducted based on the studyforrest dataset. As awareness of the project increases, we hope to see many more  dataset extensions published by independent contributors.

Forrest Gump as a research tool

The movie Forrest Gump is a great fit for the project’s central aim to encourage broad re-use of the data. The movie presents a story about life, love, and tragedy, while also covering decades of movie time and referencing familiar historical events. It conveys deep emotions, beyond the usual laughter and surprise in an average movie. At the same time, it depicts most basic functions of the human body, shows a great variety of landscapes, and features dozens of songs spanning decades and genres. Importantly, movie audio descriptions for-the-blind are available in several languages; those enable us to present the same story using different sensory input.

Why publish on F1000Research

The articles we have published in F1000Research are, so far, almost exclusively descriptions of data extensions of the studyforrest project. The Data Note is an attractive format for us to formally publish results of small projects. One instance is when I use the movie as an example subject for teaching scientific observation as a psychological method for characterizing quality and quantity of human behavior. This student-authored Data Note is the output of a single semester-long course, where they developed a system to describe portrayed emotions in the movie. We were able to collect the data and then write and publish the paper before the semester was over — a timeline that simply is not possible in a traditional publication scheme with pre-publication peer-review. Connecting the conception and execution of a project directly with the publication of the results greatly enhanced the learning experience.

Open data are seeds

I feel that not having data open is a giant waste. It leads to needless duplication, insufficient scrutiny, and missed opportunities. Many more people are capable of exploiting the potential of a dataset than are in the position to produce it. My hope is that, in a few years, the studyforrest project can be pointed out as a clear piece of evidence that open data are the seeds yielding the greatest harvests for everyone — for researchers, society, and the pursuit of knowledge itself.