The most natural and basic form of language use is dialogue, in which two or more interlocutors are engaged in both production and comprehension. But psycholinguistics focuses on the isolated comprehension or production of language, using paradigms such as reading or picture naming. In contrast, the researchers shall conduct a series of experiments that draws on the interactive-alignment model (Pickering & Garrod, 2004, Behavioral and Brain Sciences), in which interlocutors come to a common understanding by aligning their words, choice of grammatical forms, and so on. A confederate and an experimental participant take turns to describe pictures to each other and determine whether the descriptions match their pictures. The experiments vary the form and content of the confederate's description and investigate its effects on the participant's subsequent descriptions. It is known that participants are much more likely to use a particular grammatical form (eg, a passive) immediately after the confederate has used that form rather than an alternative form (eg, an active). Specific experiments manipulate whether the participant's description immediately follows the confederate's description or not; whether both descriptions involve some of the same words or not; and whether those words have the same meaning or not.

All participants were from the University of Edinburgh student community. Each experiment used a different set of participants (though a participant in one experiment could also be a participant in another experiment). There were 510 participants altogether. We output collected was either a participant's verbal description of a picture or a key press response for a picture. Both of these types of responses have been recoded into categories.