by Michael D. Anestis, M.S.
The vast majority of my research career has been spent examining the role of emotions in problematic behaviors. In previous PBB posts, I have touched on a number of emotion-based variables that interest me - particularly negative urgency and distress tolerance - but these variables are not the only ones worth consideration when researching emotions. That being said, in recent years, another emotion-based variable has caught my attention: affective lability. Defined as the degree to which an individual tends to experience frequent shifts in emotional valence (positive versus negative) and intensity, affective lability has been linked to a number of problematic outcomes including aggression, substance abuse, excessive reassurance seeking, suicide in older adults, and symptoms of borderline personality disorder (Anestis et al., 2009; Coccaro, 1991; Ebner-Priemer et al., 2007; Simons & Carey, 2006; Turvey et al., 2002). Given its connection to all of these harmful variables, affective lability is quite obviously a dangerous tendency and an ideal target for therapeutic interventions.
I first became interested in affective lability when I was granted access to an amazing database comprised of 134 women diagnosed with bulimia nervosa (BN). My co-authors and I found that the degree to which these women self-reported a general tendency to experience rapidly shifting emotions predicted the degree to which they engaged in a number of problematic behaviors ranging from excessive reassurance seeking to the various behaviors assessed by the Impulsive Behavior Scale (e.g., non-suicidal self-injury, aggression). This was true even when we accounted for the impact of age, depression, state and trait anxiety, and general impulsivity. So, in other words, in a sample of women who were habitually engaging in a problematic set of behaviors (binge eating and purging), the degree to which they experienced frequently shifting emotions was highly related to whether or not they also engaged in a host of other problematic outcomes.
We were very happy with this study, which was published in the International Journal of Eating Disorders in 2009, but I was interested in investigating two facets of affective lability left untested in our work. First, I was interested in the degree to which affective lability was related to disordered eating behaviors themselves. Second, I was interested in determining the best way to assess affective lability.
In the 2009 project, my co-authors and I used the most common method for assessing affective lability: we asked participants to fill out a questionnaire that asked them to what degree they generally experience various components of that concept. In other words, we asked people to think back over their lives and to assess the degree to which their emotions tend to shift in valence and intensity. This approach is inexpensive and quick, but it also rests on the assumption that people can accurately recall the degree to which their emotions tend to fluctuate and that such fluctuations occur at a relatively stable rate over an extended period of time. Other researchers have taken a different approach known as ecological momentary assessment (EMA). EMA involves sending participants home with a device - usually a palm pilot - that they use to fill out a series of questionnaires repeatedly in their natural environment. Several times each day, the participant is beeped, at which point they fill out a series of questionnaires, including one that assesses their levels of various emotions. Additionally, any time participants engage in a particular behavior, they are told to fill out the questionnaires again. In this way, researchers can get real time measurements of variables that usually require them to ask people to remember retrospectively in a laboratory setting. This is particularly important for affective lability, as researchers can actually examine the degree to which emotions fluctuate by looking at momentary ratings over time.
All this being said, shortly after the publication of the 2009 study, I was granted access to another amazing dataset. In this project, I worked with Eddie Selby, Ross Crosby, Steve Wonderlich, Scott Engel, and Thomas Joiner (Anestis et al., in press) on a study that was just recently accepted for publication in Behaviour Research and Therapy. Once again, the sample consisted of women diagnosed with BN, but this time we had access to both questionnaire-based affective lability and EMA measures of affective lability. This data thus allowed us to determine if one particular method was more useful in predicting specific behavioral outcomes. In this case, we were looking at two particular outcomes: general eating disorder symptoms and the number of binge eating episodes experienced during the course of the experiment.
Our sample consisted of 131 women ranging in age from 18 to 55, 97% of whom were Caucasian (an obvious weakness of our study). Each participant was given $200 for participating and an additional $50 if they completed at least 85% of the scheduled EMA assessments. Participants were beeped six times each day for two weeks and were told to fill out the questionnaires immediately following the use of any of a number of behaviors. Each time a participant filled out the questionnaires, they indicated the degree to which they were currently experiencing a number of specific emotions. To calculate affective lability, we used the mean square successive difference (MSSD), which is a measure of the degree to which an individual's score on a particular item differs on average from his or her preceding score on that same item. In this case, we looked at the MSSD specifically for negative emotions, which means we were interested specifically in the degree to which participants experienced rapidly shifting levels of negative emotions.
After looking at the data, we reported a number of interesting findings. As expected, the two measures of affective lability were significantly correlated with one another (r = .38), although the correlation was small enough to indicate that the two measures capture at least slightly different ideas. Additionally, both measures of affective lability significantly predicted general eating disorder symptoms and neither relationship was stronger than the other. This indicates that the degree to which an individual is experiencing rapidly shifting emotions at the moment is no better at predicting overall eating pathology than is a questionnaire asking an individual to what degree this fits a general pattern in their life. In some ways, this is very good news, as it means that the more efficient and economical measure is just as useful as the more expensive and difficult to implement one. More interestingly, however, EMA affective lability was a significantly better predictor of the number of binge eating episodes that participants had during the actual study than was the retrospective questionnaire measure of affective lability. In other words, if a clinician or researcher is interested in predicting whether or not an individual is at imminent risk for binge eating behavior, they will obtain much more useful information by assessing affective lability in real time rather than asking the client to recall general tendencies throughout their lives.
These findings are by no means earth shattering, but they provide us with some useful information. The bottom line is, certain methods of assessment are more expensive and difficult than others. At times, those more difficult methods are justified, as they provide much more accurate predictions in those circumstances. This was the case here with respect to binge eating behavior during the two weeks of the study. At other times, the more intricate and involved methods are no better than a simple questionnaire at predicting a particular outcome and, in those cases, it makes more sense to go with the practical, inexpensive approach. This was the case here with overall eating disorder symptoms.
In addition to clarifying the utility of particular measurement techniques, this study helped solidify our understanding of the role of emotions in binge eating behavior. That being said, there were also a number of important limitations. As mentioned above, there was essentially no racial diversity in our sample and every individual in the sample was female and diagnosed with BN, which limits our understanding of the degree to which these findings generalize to other populations. Additionally, the effect sizes for our findings were fairly small, meaning that although the relationships appear to be real, they account for only a small portion the variability in binge eating.
My question for you is this: given that our findings indicate that EMA methods are significantly better in the prediction of binge eating over a discrete period of time, should clinicians feel compelled to use this approach in their clinics? EMA is expensive and difficult to implement, so there are some serious costs and benefits considerations at play. At what point does an expensive and difficult measurement method become an ethical requisite?
************
If you would like to learn more about this or other topics discussed on PBB, we hope you will consult our online store for scientifically-based psychological resources.
Mike Anestis is a doctoral candidate in the clinical psychology department at Florida State University and an incoming resident at the University of Mississippi Medical Center.
Subscribe in a reader
