Image generated with Nightcafe. Prompt: "A highly stylized abstract painting depicting body movement at a gym, with many silhouettes, high intensity and dynamic movement."
See your stimuli move a little (or a lot!) using the RAISE methodology.
This is a Web version of the Companion Appendix for RAISE: A New Method to Develop Experimental Stimuli for Advertising Research with Image Generative Artificial Intelligence, developed by César Zamudio, Jamie Grigsby, and Meg Michelsen, and published in the Journal of Advertising. Questions? Feel free to contact us!
The RAISE method allows to manipulate broad body movement. Consider a scenario in which a researcher wants to manipulate movement in three conditions (no movement, low movement, and high movement), and will use a gym member as the visual element on which movement will be induced. Application of RAISE proceeds in a similar manner as our first example (gender manipulation of waiters). Note that movement is the visual element to be manipulated, and not the gym member herself. Since the first condition is of no movement, Stimulus A can be prompted as follows.
To induce movement, first, the image generative AI text prompt for Stimuli B and C must reflect changes in movement. Aside from the updated text prompts, to induce movement, image strength must be lowered considerably such that Stimuli B and C are sufficiently different from the initial, still condition, to depict movement in different intensities, but similar enough to retain similarity and control. Stimuli B was constructed as follows,
with a very similar strategy for Stimulus C, except that image strength is lower, to allow generation of a less similar image, with more action:
Below is a handy summary card outlining the process you just learned:
Aside from the updated text prompts, to induce movement, image strength must be lowered considerably such that Stimuli B and C are sufficiently different from the initial, still condition, to depict movement in different intensities, but similar enough to retain similarity and control. However, because further movement requires a higher degree of precision, a potential drawback of this approach is that low image strength results in minor loss of control (e.g., sleeve color). The Technical Details section discusses how to refine these issues.
Ready to learn more?
Feel free to browse our gallery of examples with full tutorials, and the technical details section to learn the finer points of generating visual experimental stimuli using the RAISE methodology.