New research to identify what makes conversations supportive

People going through hard times often find assistance by talking about their problems with others, and the support people receive has documented health benefits. However, the quality of support can vary and what happens during conversations can determine whether or not the support is actually beneficial.

In a new three-year project funded by the National Science Foundation, researchers from Penn State, the University of Mississippi and the University of Minnesota aim to clarify how conversations about everyday stressors convey support and how they lead to different emotional outcomes.

According to Denise Solomon, co-principal investigator and professor of communication arts and sciences at Penn State, previous research has focused either on stand-alone, brief support messages or on perceptions of supportive conversations as a whole.

“What is missing is an understanding of how messages unfold over the course of a conversation to regulate the emotions of a person in distress," said Solomon. "Our project will focus on studying the conversation linkages between one person’s supportive messages and the other person’s cognitive and emotional responses in an effort to map those dynamic patterns.”

The researchers will analyze data from four previous studies involving 461 video-taped conversations in which one person discloses a stressor to another person, who provides verbal support and assists in the coping process. The research will develop and test a theoretical model that examines how variations in these interactions lead to differences in the distressed person’s thoughts and feelings as expressed through language during the conversation and also reported after the interaction.

To identify speech acts that positively and negatively affect emotion regulation, the researchers will focus on verbal response modes (VRM) to capture eight mutually exclusive categories of speech acts that are defined by the form and intent of a speaker’s individual utterances. To reveal the distressed person’s thoughts and emotions in response to particular VRMs, their speaking turns will be analyzed using a linguistic inquiry and word count (LIWC) program.

“The goals of the analysis will be to identify clusters of speech acts within turns and to test associations between the support provider’s speaking turns and the distressed person’s emotion regulation as these unfold over the course of the conversations,” said Nilam Ram, professor of human development and family studies, and psychology. “We will be breaking down video recordings of conversations, utterance by utterance, and turn by turn, and developing a tool kit of analytical strategies to show how emotional and cognitive processing can be affected dynamically during a conversation.”

The researchers predict that individuals who receive effective support messages will leave the conversation with a new understanding of their problem and in an improved emotional state.

“Importantly, though, it’s not enough for a partner to simply provide good support. We expect the most positive outcomes to occur when a distressed person receives high-quality support messages and is responsive to those messages as they occur during the interaction,” said Solomon. “Conversely, we expect that being unresponsive to high-quality support won’t lead to emotional improvement, and receiving and responding negatively to low-quality support will increase the negative consequences of receiving poor support.”

In the future, Solomon hopes that their findings have the potential to assist those who are support providers and will lead to research explaining why some people or relationships show more or less responsiveness during supportive conversations.

“The novelty in this research is mapping responsiveness within interactions onto important conversational outcomes, which opens the door to new questions about why those patterns differ between people and between relationships. We also envision that the tool kit we develop can be used to illuminate the dynamics of other types of consequential conversations, such as in conflict negotiations or attempts to influence a partner’s health behavior.”

Ram, who is also director of the Quantitative Development Systems Methodology Core, part of the Social Science Research Institute, will be conducting the analysis of the coding. Other researchers on the project include co-principal investigator Graham Bodie, visiting professor of integrated marketing communications at the University of Mississippi; Susanne Jones, associate professor of communication studies at the University of Minnesota; and Peter Molenaar, distinguished professor of human development and family studies at Penn State.