In this study, two advanced computational text analysis tools were used to catalogue lexical and cohesive features of deceptive language and language accordance (i.e. agreement/disagreement on topic of conversation) in a corpus of dyadic conversations. The study specifically focused on how the variable of accordance conditions the process of deception in terms of lexical and cohesive features. The results indicated that there is no interaction between deception and accordance in deceptive conversations in terms of cohesive or lexical sophistication indices. The results also did not show main effect for indices of cohesion and lexical sophistication for deceptive versus non-deceptive conversations. However, main effects were observed for indices of cohesion and lexical sophistication in distinguishing conversations characterized by agreement or disagreement. The linguistic differences related to the cohesive and lexical sophistication aspects of agreement versus disagreement conversations are discussed.

Cognitive psychology and artificial intelligence have provided valuable insights into the scope and limitations of human thought and behavior. As technology becomes more of a fixture in our daily routines, advances in artificial intelligence increasingly impact how we think and interact with others. This track is motivated by these two fronts of research: the basic theoretical integration of cognition and artificial intelligence; and its application to real-world domains. As such, the track will cover a wide range of issues. We welcomed submissions in any area where cognition and computers are mutually explored, but especially encouraged work in how humans and computers communicate or how artificial intelligence facilitates communication.