Adaptation and communicative design.

Two theoretical frameworks that examine the nature of adaptability and mutual influence in interaction, interpersonal deception theory and interaction adaptation theory, were used to derive hypotheses concerning patterns of interaction that occur across time in truthful and deceptive conversations. Two studies were conducted in which senders were either truthful or deceptive in their interactions with a partner who increased or decreased involvement during the latter half of the conversation. Results revealed that deceivers felt more anxious and were more concerned about self‐presentation than truthtellers prior to the interaction and displayed less initial involvement than truthtellers. Patterns of interaction were also moderated by deception. Deceivers increased involvement over time but also reciprocated increases or decreases in receiver involvement. However, deceivers were less responsive than truthtellers to changes in receiver behavior. Finally, partner involvement served as feedback to senders regarding their own performance.     

The impact of violations on uncertainty and the consequences for attractiveness

This investigation aimed at extending past research on expectation violations by arguing that violations vary both in valance and in their influence on uncertainty and that the combination of valence and uncertainty states conjointly influences judgments of a violator's social attractiveness. In explaining the predicted variance in uncertainty following violations, a distinction is proposed between congruent violations (i.e., behaviors that are more intense instantiations of a previously displayed message) and incongruent violations (i.e., behaviors that are opposite in meaning from previously displayed messages). Five models for explaining violation effects are contrasted. Results (a) confirm that violations differ in their impact on uncertainty, (b) generally support the validity of the proposed distinction between congruent and incongruent violations, and show that the inclusion of uncertainty and valence in models of violation outcomes accounts for greater variance in social attraction than either one separately.