MEG Picture Naming Task

The details of the experiment can be found in an article titled: MEG Neural Decoding Pipeline: The issues residing within the data and methods to improve your decoding accuracy.

The subjects were placed inside a superconducting self-shielded magnetoencephalography (MEG) system created by Sumitomo Heavy Industries Ltd. (SHI), in an upright standing position. To reduce leg fatigue, a leg-support device (ArchelisFX, Archelis Inc.) was used during the experiment. A delayed picture naming task was used in this experiment. The participants were asked to name the items in the images. The visual stimuli (i.e. images) were presented on the screen in a randomized order followed by a specific cue for response initiation. Six pictures (sushi, dumplings, cookies, comb, knife, and pencil) were selected from two categories (food and tools). 

The linguistic expressions corresponding to the images were matched with their length (number of mora) and frequency for each category. The stimuli were visually presented on a nonmagnetic projection screen inside the MEG machine chamber. Each trial consists of four steps. Step 1: stimuli presentation. Step 2: the ‘*’ cue presentation. Step 3: the ‘**’ cue presentation. Step 4: the ‘***’ cue presentation. The instructions for step 1 were to silently observe the image. For step 2, to imagine an action involving the item from step 1 (eating the food or using the tool). For step 3, to name the item silently. For step 4, to overtly pronounce the name of the item. Each session was 10 minutes long. Each subject performed 4 sessions during the experiment, making a total number of repetitions per image to be 50. After each session a short break was introduced, and subjects were allowed to exit the machine. They were allowed to rest until they were ready to proceed. If any subject wished to exit the machine in the middle of a session, the session was terminated, and the data of that session was discarded. The subjects were allowed to redo the session and complete the experiment if they wished to continue, otherwise, the experiment data was discarded completely. In total, participants spent no longer than an hour performing the experiment.