1. Relationships, emotion, and eating: A dynamic systems investigation of weight gain (NIH funded)
2. Surviving breast cancer: The dynamics of inflammation, emotion, and family (ACS funded)
3. Computational temporal interpersonal emotion systems ("Comp-TIES")(NSF Funded)
Funded by the National Institute of Health
Despite extensive research and intervention efforts the percentage of overweight Americans has continued to climb, putting a large segment of the population at risk for chronic disease. One reason for this failure may be that traditional research is not well suited for tackling the dynamic complexity of factors that result in a person becoming overweight. The present study takes up this challenge and adopts a dynamic systems approach to test a model that predicts the combined effects of relationship dynamics, emotions, and autonomic physiology on eating, activity, and short term weight gain in newly formed families. During the early years of marriage couples establish shared eating and activity habits that will subsequently be passed on to their children. Interrupting the obesity epidemic requires research focused on how couples develop shared obesity relevant behaviors. These mechanisms are amenable to behavioral intervention, but a key challenge for developing such interventions is our lack of understanding about how these factors interact.
The basic premise of our model is that couple-level behavioral and emotional patterns focused around eating and activity provide proximal forces perpetuating eating and activity habits, with accompanying effects on body weight. We focus on two patterns that research on smoking and alcohol addiction suggests are important for maintaining unhealthy behaviors. The first, System-Symptom Fit (SSF), refers to an unhealthy behavior, such as excessive eating, helping to preserve relationship well-being by increasing positive emotion or couple closeness (Rohrbaugh, Shoham, Butler, Hasler, & Berman, 2009; Shoham, Butler, Rohrbaugh, & Trost, 2007). The second pattern, Demand-Withdraw (DW), occurs when one partner demands change in the other partner’s health behaviors, which causes the “nagged” partner to withdraw from the interaction and to resist change (Eldridge, Sevier, Jones, Atkins, & Christensen, 2007). We propose that the presence of SSF or DW in newly formed couples will predict increases in weight and deteriorating health habits 6 months later. To test our hypotheses we will recruit 100 heterosexual couples who are within the first year of cohabiting. All participants will be in the upper-end of the healthy weight range at the beginning of the study. As such, they will be at risk of becoming overweight. Multi-method assessments of SSF and DW at baseline will be used to predict changes in eating/activity habits and weight 6 months later.
This project has the potential for broad impact due to integrating across traditionally separate scientific and clinical communities, and by articulating a dynamic systems approach to the problem of obesity. As such, it could provide a theoretical basis for researchers and clinicians from diverse backgrounds. In addition, results from the study would inform research and practice regarding optimal, targeted ways to integrate family systems, emotion regulation, and biofeedback approaches to understanding and treating obesity.
Funded by the American Cancer Society
Psychosocial interventions such as group therapy sometimes help women survive longer after a breast cancer diagnosis, but sometimes they don’t. Similarly, lower emotional distress and greater family support are sometimes associated with better outcomes for breast cancer patients, but sometimes they aren’t. One reason for this lack of reliable findings may be that the associations amongst these factors are not simple, linear, or static. Instead biological factors such as inflammation likely interact with emotional distress in complex ways over time. As an example, for some patients distress and inflammation may become coupled in an escalating vicious circle that results in a devastating state of biological and psychological break down. As an added complexity, patients with better emotion regulation skills, or who have families that are better able to help them regulate their emotions effectively, may be less susceptible to this potentially deadly pattern.
Although many scientists agree that we need to understand these complex processes as they unfold over time, research has been hindered by a lack of statistical tools capable of representing nonlinear dynamic patterns of association amongst multiple variables. Fortunately, several powerful statistical methods developed in fields such as engineering and physics have recently been adapted for social science applications. We will apply these methods (coupled latent difference score and coupled linear oscillator models) to existing data from two large population based longitudinal studies of breast cancer patients in the years immediately following diagnosis. We will address two questions: 1) Does higher emotional acceptance (a form of individual emotion regulation) prevent a pattern of coupled, escalating emotional distress and inflammation from developing? and 2) Does higher emotional coregulation (a form of family-level emotion regulation) reduce the connection between emotional distress and lower chances of survival?
Answers to these questions would help us to predict cancer outcomes for individual patients and to develop integrated biological-psychological interventions that could be tailored for specific patient profiles. For example, if the answer to our first question is “yes” it would suggest that when distress and inflammation become coupled it may be necessary to intervene with both medication and intensive training in emotion regulation. In contrast, a patient who does not show such coupling may respond well without the need for psychosocial intervention. Similarly, if the answer to our second question is “yes” it would suggest that clinicians need to look closely at family-level emotion regulation, not just at patient distress. If a patient is embedded in a family that is not capable of providing emotional coregulation then any purely individual treatments that do not impact the family context are unlikely to be effective.
Funded by the National Science Foundation
In collaboration with Dr. Kobus Barnard, School of Information: Science, Technology, and Arts, University of Arizona
See our website at http://www.compties.org/
The overarching goal of this project is to develop and disseminate statistical models of dynamic interpersonal emotional processes. Emotion is often framed as an intra-personal system comprised of subcomponents such as experience, expressive behaviors, and autonomic physiology that interact over time to give rise to emotional episodes. One critical limitation is that emotions often occur in the context of social interactions or ongoing relationships. When this happens an inter-personal emotion system is formed, in which the subcomponents of emotion interact not only within the individual but across the partners as well. Predicting and intervening in these temporal interpersonal emotion systems (TIES) requires statistical models that represent complex interdependencies across emotional subcomponents over time both within individuals and between social partners.
TIES are central to many aspects of human functioning including child development, close relationships, physical and mental health, social coordination, group performance, and intractable conflict. The field is fragmented, however, due to the lack of a theoretical framework and limited knowledge of existing statistical approaches. In addition, readily available approaches are themselves severely limited compared to those being developed in fields such as computational science. Thus, the objective of our resarch is to develop and disseminate more powerful and accurate probabilistic TIES models. Our first specific endeavour will be to develop a dynamic latent state model to investigate emotional coregulation, which refers to the process by which parent and child form a dyadic system that dynamically maintains an optimal joint emotional state. Coregulation may continue to provide a central form of emotion regulation across the lifespan, with implications for optimal functioning and well-being, but research has been hindered by the lack of statistical models specific enough to distinguish coregulation from other related processes.