- Note The Following Scenario Will Be Revisited For Several Questions Suppose We Are Interested In Whether People Simu 1 (42.25 KiB) Viewed 235 times
- Note The Following Scenario Will Be Revisited For Several Questions Suppose We Are Interested In Whether People Simu 2 (57.62 KiB) Viewed 235 times
- Note The Following Scenario Will Be Revisited For Several Questions Suppose We Are Interested In Whether People Simu 3 (29.43 KiB) Viewed 235 times
- Note The Following Scenario Will Be Revisited For Several Questions Suppose We Are Interested In Whether People Simu 4 (46.74 KiB) Viewed 235 times
Suppose we are interested in whether people simulate the implied shape of an object. Participants read sentences like "The lobster was in the sea" (e.g., a living creature) vs. "The lobster was on the plate" (e.g., a dish for dinner). Later, they are shown pictures of different objects, and are asked whether these pictures appeared in the sentences they read before. Some of these pictures match the orientation from before, while others don't. We measure subjects' d- prime (d'), a kind of measure of recall accuracy. If people do simulate implied shape, what should we predict about this measure? D-prime will be lower (i.e., worse recall accuracy) when the picture shape matches the implied shape from the sentences. People will read sentences faster when the picture shape matches the implied shape from the sentences. OD-prime will not change as a function of the match/mismatch manipulation. People will read sentences more slowly when the picture shape matches the implied shape from the sentences. D-prime will be higher (i.e., better recall accuracy) when the picture shape matches the implied shape from the sentences.
Consider the same experiment investigating the simulation of implied shape. If we did find evidence consistent with simulation, what could we conclude from this experimental design? Some degree of simulation happens, but the study demonstrates that it is purely epiphenomenal. There's no evidence that simulation happens in this study. Some degree of simulation happens, but it's unclear whether it is necessary for comprehension. Some degree of simulation happens, and it's necessary for comprehension.
Now we run another study. Now, we want to know whether the degree of simulation is larger when the entity described is the subject or object of a sentence. Thus, some sentences have the entity as the subject ("The lobster was in the sea", "the lobster was on the plate"), and others have the entity as the object ("He caught the lobster", "He ate the lobster"). Which of the following would constitute an interaction between Grammatical Role (subject vs. object) and Match Condition (match vs. mismatch)? People exhibit worse recall for mismatch items, but there is no difference in recall accuracy as a function of Grammatical Role. People exhibit worse recall for mismatch items, but only when the entity is the subject of a sentence. People exhibit worse recall for mismatch items in general; they also exhibit worse recall when the entity is the object of a sentence in general. There is no effect of Grammatical Role or of Match Condition.