|
Aaron Fisher
University of California, Berkeley, USA
|
The Promise and Possibility of Discrete Data for Emotion-Related Research
Traditionally, the field of psychology has measured emotion variables continuously and employed correlational methods derived from the general linear model to analyze them (e.g. regression, multilevel models, structural equation models). Of course, these methods have yielded decades of insights into the form and function of human emotion. However, like the proverbial fish who does not know that it is in water, the ubiquity of our contemporary methods sometimes blinds us to their limitations, and to the possibilities that lie outside of them. For instance, we rarely take full advantage of time, specifically the element of timing.
Emotion is a dynamic, time-varying phenomenon and much of our current cutting edge emotion science is conducted with repeated measures and time series designs. However, in estimating the statistical parameters associated with these methods—the slopes, autoregressions, and variability measures—we are required to aggregate across time points, losing the capacity to examine each point in time independently. Lost is information on timing, frequency, duration, and sequence. These measurements require discrete data representations.
Invoking the notion of discrete data, especially for ostensibly continuous constructs like emotion, inevitably provokes a negative response from researchers. We are told that we “lose information” when we discretize continuous data. And, certainly, this can be the case, especially when we try to retrofit discrete data into linear models that are optimized for continuous inputs.
In the present talk I will describe two parallel areas of research in my lab that demonstrate the promise and possibilities of discrete data representations in emotion research. First, I will discuss methods for effectively and validly discretizing multivariate emotion data into discrete compound emotion states. Next, I will present recently developed information theoretic algorithms that we have created for building compound state-context sets, which can then be used to make accurate predictions about discrete behaviors and events. In all, I hope to demonstrate the promise and possibility of discrete data structures and to illuminate the new and exciting areas of description, prediction, and intervention that these approaches open to us.
|
|
Mariska Kret
Leiden University, NL
|
Emotion Processing in Homo and Pan
Evolution prepared group-living species, (non)human primates included, to quickly recognize and adequately respond to conspecifics’ emotional expressions. Different theories propose that mimicry of emotional expressions facilitates these swift adaptive reactions. When species unconsciously mimic their companions' expressions of emotion, they come to feel reflections of their emotions that influence emotional and empathic behavior. The majority of emotion research has focused on full-blown facial expressions of emotion in humans. However, facial muscles can sometimes be controlled; humans know when to smile, and when not to. In this talk, I therefore argue for a broader exploration of emotion signals from sources beyond the face or face muscles that are more difficult to control. More specifically, I will argue that implicit sources including the whole body and subtle autonomic responses including pupil-dilation are picked up by observers and influence subsequent behavior. Across different primate species, seeing a conspecific being emotional and expressing that in one way or another, immediately and automatically attracts attention, yields mimicry and triggers action tendencies in observers. Taking a comparative approach I investigate similarities and differences in the perception of emotions between humans, chimpanzees (Pan troglodytes) and bonobos (Pan Paniscus).
|
|
Kristen Lindquist
University of North Carolina at Chapel Hill, USA
|
Constructing Emotion
Questions about the nature of emotion are some of the most enduring in psychology and neuroscience. We have been studying emotion scientifically for over a century, but answers to questions about the nature of these important states have remained elusive. Traditionally, attempts to weigh in on the mechanisms of emotion have used a single level of analysis and focus almost exclusively on cognitive, neurophysiological, or cultural mechanisms. In this talk, I discuss work that spans all three. I will begin by showing experimental evidence that emotions are mental states characterized by cognitive features such as valence, arousal, and situated semantic meanings. Next, I’ll demonstrate that these features are the product of interactions amongst distributed brain networks that predictively regulate visceromotor outputs by making best guesses about adaptive actions. Finally, I’ll close by showing that such predictions are learned via experience within particularly cultural contexts. Together, this work forms the basis of a new constructionist model in which emotions are both deeply embodied and encultured states.
|
|
Batja Mesquita
KU Leuven, BE
|
Between us: How cultures create emotions.
I will take you on the journey that my own research has taken me to show that emotions are not universal responses coming from within, but that we craft our emotions to attune them to the values, norms, and practices of the cultural worlds outside our individual selves. Emotions are learned and co-constructed in the course of our many everyday social interactions. What makes us human is not that emotions are identical for all but that they connect us to others, in our direct environment and in our broader cultural context. I argue that appreciating that emotions are fundamentally linked to our outside worlds – and that we have a role in creating and changing them – is a bridge to more effective relationships in all of the many multicultural spaces in our communities.
|