Kimberly Stevens and Dr. Julianne Holt-Lunstad, Psychology Department
Midlife adults face a wide variety of physiological, emotional, and cognitive stressors that place them at risk for impaired physical health and longevity. Social integration—which includes high levels of social engagement as well as maintaining a diverse network of social roles—has been shown to protect against the negative effects of these stressors and decrease mortality rates (Holt-Lunstad, 2010). Research has suggested that “social buffering” provides protection from the physiological effects of stress (Cohen, 1985). In fact, Sheldon Cohen hypothesized a “social buffering” pathway, which suggests that social integration effects psychological and physiological exposure to and reaction to stress (Cohen & Wills, 1985). In essence, those who are socially integrated are hypothesized to show a better emotional tone and reduced stress reactivity. To better understand the possible neurobiological correlates of social buffering, we examined if measures of social integration predicted neural indications of emotional processing and regulatory reactions. The prefrontal cortex, specifically the anterior cingulate (ACC) is involved in regulation and inhibition (Oschner & Gross, 2007). Downstream paralimbic areas, specifically the amygdala, are associated with stress and emotional reactivity (Murphy, Nimmo-Smith & Lawrence, 2003). Therefore, we predicted that adults with higher levels of social integration will show less activation in the amygdala and greater activation in the ACC when presented with emotional regulation and stress inducing tasks.
We recruited 40 participants (20 male, 20 female) from the local population, between the ages of 40 and 68. They were recruited using flyers and facebook advertisement. Participants completed an extensive survey before their appointment, which included the Social Network Index (SNI), a measurement of social integration and social network size. Upon arrival at the MRI facility, participants underwent an hour of behavioral testing. Participants were administered the Rey-Valt (a test of auditory memory), the Stroop Color-Word task, and the Trails test (both measurements of executive functioning). This was followed by a blood draw to measure blood glucose and cholesterol levels. Participants were then trained on two functional neuroimaging tasks they were asked to complete while in the scanner. These tasks included the Emotional Regulation task developed by Pete Gianaros and colleagues (2014), designed to assess how participants manage emotional reactions to upsetting stimuli, such as pictures of natural disasters, and the Multi-Source Interference Task (MSIT) is a visual numbers game, designed to stress the participant and provide a neural measure of inhibition and executive functioning. Participants were then provided with earplugs, basic MRI safety training, and fitted with a blood pressure cuff. Neuroimaging data was collected using the 3 Tesla MRI scanner at the MRI Research Facility. The scan lasted approximately 30 minutes, beginning with a structural scan, followed by two functional scans, during which participants will be expected to complete the tasks. During the first two scans, participant blood pressure and heart rate were measured a total of 12 times as an additional measure of stress. Following the completion of the scan, participants were removed from the scanner, compensated, and sent a picture of their brain via email.
Due to the status of the project, the data of 31 participants was included in these preliminary analyses. All participants (18 female, 13 male) were between the ages of 40-68 (M = 49.77, SD = 7.47). Social Integration scores were not predictive of BOLD activity in the ACC or Amygdala during emotional regulatory task. In an exploratory, whole brain analysis, social network size was predictive of BOLD activation in the right superior parietal lobe under the decrease negative (beta=-.496 p=.005) and look neutral conditions ([beta=-.404; p=.024).
A small difference in BOLD activation in the right superior parietal lobe when the look “decease negative” condition as compared to the “look neutral” condition is indicative of less reactivity when presented with disturbing images. As the right superior parietal lobe is associated with visual working memory, it is possible that adults with higher levels of social integration have less need for vigilance to threats in their environments. This is consistent with the social support/ buffering hypothesis, indicating a social protection against the physiological effects of stress that would accompany constant vigilance. This suggests that those who are more socially integrated have less of a neural reaction to emotionally provocative stimuli, and therefore may provide potential neurobiological explanations for why some adults show better age-related trajectories that affect health than others. However, this does not explain the lack of correlational activation in the ACC and amygdala, which was one of the most interesting and unexpected findings, and inconsistent with our original hypothesis.
These findings are promising to better elucidate the neurobiological correlates of the connection between social integration and health. However, this data is only a preliminary analysis representative of 75% of the data collected as part of this project. Therefore, further analysis is needed and we will continue further data analysis, including other correlates of age, blood pressure and executive functioning.