Language enables people to talk about entities in the world. However, the same entity can be addressed with different linguistic expressions — and the choice is never arbitrary. At the Dahan Lab, such choice is studied through the investigation of a psychological concept called common ground. More specifically, we investigate how people decide which referring expression to use when discussing a specific entity for multiple repetitions with their addressee.
Participants recruited from the greater Philadelphia area are invited to play a matching game, where they are presented a set of 16 cards in various fixed configurations. Through verbal communication, the director (A) instructs the matcher (B) to arrange B’s cards the same way A’s is presented. The roles of director and matcher are evenly switched throughout trials. Their linguistic choices are recorded and analyzed using a coding program called Praat. Not only are the head nouns and descriptors coded for each mention of each card, the articles used are also coded based on its definiteness — definite articles may convey the presupposition that the entity discussed can be uniquely identified by the addressee because of mutually shared knowledge. “These are the red pumps” is definite while “This is a pair of red pumps” is indefinite.
After the matching game, participants are asked to fill out several surveys that provide insight on their personality traits, cognitive abilities (ADHD tendencies), and demographic variables such as age, sex, and education level. Taking such parameters into consideration, we examine how people adopt different strategies and which of these variables may predict people’s behaviors using a programming language called R.
In order to understand and conduct research at the Dahan Lab, I’ve internalized all the steps of empirical psychological research, from understanding the basis of “common ground,” to designing a study, to contacting participants and running experiments, to data coding and analysis. I recognized the importance in converting language use into analyzable data, and how such data can be arranged in order to reflect patterns between certain demographic traits and linguistic characteristics. I learned to insert codes and create textgrids using a speech analysis software called Praat and taught myself the basics of R — including packages such as tidyverse and functions such as pipe, mutate, select, and filter. The ability to draw inferences from data that I personally collected and reviewed provided me a more wholistic view of conducting psychological experiments. The lab as a whole also served as an excellent connect for my interest in science and my passion for languages.