Running Behavioral Experiments Online: A Quick How-To

,

Building behavioral experiments online can feel daunting at first, but it’s becoming a crucial skill for researchers who want to reach a broad and diverse audience. Whether you’re moving online out of necessity—due to lab closures or difficulties recruiting participants in-person—or simply to take advantage of the vast pool of potential participants, the move from lab to online experiments requires thoughtful consideration. Here’s a brief guide on how to create, host, and recruit for successful behavioral studies online.

The Three Pillars of Online Behavioral Studies

Running a behavioral study online follows the same core steps as a traditional lab-based study: building, hosting, and recruiting. The main difference is that each step has to be integrated smoothly to create an experiment that runs seamlessly across platforms. Let’s break it down:

  1. Building Your Experiment
    Online experiments are typically built with specialized software like PsychoPy, OpenSesame, or Gorilla. These tools often provide a user-friendly drag-and-drop interface, but still allow scripting for more complex experimental designs. Because the web relies on JavaScript, many experiment builders now translate Python-based code into JavaScript to ensure everything runs smoothly in a browser.
  2. Hosting Your Study
    Once your experiment is built, it needs to be hosted online. This means uploading your experiment to a server where participants can access it. Hosting services like Pavlovia or JATOS provide easy integration for popular experiment builders, taking care of everything from data storage to managing participant access. You could host your own server, but centralized services are usually simpler and come with added features like user management and secure data storage.
  3. Recruiting Participants
    Online recruitment offers unprecedented access to diverse participant pools. Platforms like Prolific and Amazon Mechanical Turk make recruiting efficient and payment easy. Unlike lab studies that rely on local university participants, these platforms let you collect data from a wide range of demographics, including hard-to-reach populations—sometimes in just a few hours.

Why Go Online?

Online experimentation has several compelling benefits. First, it offers diverse sampling: you can reach participants from different cultural, social, and geographic backgrounds. It also scales effortlessly, meaning you can collect data from large samples without much extra work. This is ideal for projects needing broader population coverage or reaching niche demographics.

Tips for Success

Transitioning to online experiments isn’t without challenges. Here are some tips for a smooth transition:

  • Data Quality Matters: Dropout rates can be high online, and participants may not be as attentive. Ensure instructions are clear and simple, and consider adding comprehension checks to keep participants on track.
  • Shorter is Better: Keep online studies concise. Our survey found that most participants lose interest after about 15 minutes, so aim for shorter sessions or break your experiment into smaller parts.
  • Incentivize Wisely: Monetary rewards can help boost participant motivation, but it’s not just about the cash. Participants perform better when they know why their contribution matters, so make sure they understand the value of your research.

Online experiments offer incredible potential to scale behavioral research and reach broader audiences, but success hinges on smart choices regarding tools, hosting, and participant engagement. With the right approach, taking your research online can open up new possibilities and drive impactful results.

Read the full article: Sauter M, Draschkow D, Mack W. Building, Hosting and Recruiting: A Brief Introduction to Running Behavioral Experiments Online. Brain Sciences. 2020; 10(4):251. https://doi.org/10.3390/brainsci10040251