Decide which type of priming to test (2 hours)

↵ Back to module homepage

Your task in this module is to design and carry out a priming experiment with data from at least 10 volunteers. Your experiment may use any type of priming, in any language, with any kind of people participating. It does not need to be a novel topic that's never been done before; it's totally fine if you do an experiment that is on the same topic that other experiments have already done. But I would encourage you make the research question something you are interested in knowing about.

To be a priming experiment, it must include at least an unrelated condition and a related condition; if you are feeling ambitious, you can also include more (think about the Zhou & Marslen-Wilson paper we discussed, which had four conditions: related meaning, related pronunciation, and the corresponding "unrelated" conditions for each of those). However, for each condition you add, you're going to need to do a bit more work, so don't get carried away.

The first step is for you to decide what kind of priming you would like to do. For reference, you can think about the types of priming you learned about at the end of the "Priming" module.

Once you have decided, write a brief description of your experiment plan. Your description must include the following information:

• What type(s) of priming relationship you are interested in testing;
• What language your stimuli will be in;
• What kind of participants you plan to use (e.g., adult native speakers of Cantonese, children learning Punjabi as a second language, etc.);
• A list of conditions you will include in the experiment;
• An example set of stimuli (i.e., a target word, along with all the possible prime words it might appear with in different conditions);
• What type of priming presentation paradigm you will use (I recommend masked priming or paired priming; list priming or long-lag priming will be more difficult to arrange);
• What task your participants will do (I recommend lexical decision, but you might also choose semantic judgment; using a naming task will require extra work and equipment, so I don't recommend that);
• What you predict the results will be like. The easiest way to show your predictions is to draw a graph of what you expect the results will look like. For example, below I have included a graph of the results I would expect in a simple meaning priming experiment: 
  

Your plan does not need to be long. Here is an example of what a plan might look like for the experiment we did in class (you can copy the format, but you should NOT copy this exact experiment idea): Experiment Plan

When you have a satisfactory experiment plan, send it to me (Eric), and I'll let you know what to do next.

By Eric Pelzl, based on content created by Stephen Politzer-Ahles. Last modified on 2025-Sep-21. CC-BY-4.0.