What are emotions?

What a silly question! Everybody knows what emotions are - or do they?
As it turns out, emotion science has a history of arguing about what it is that it deals with. However, the situation is not so bad - over time a consensus is emerging that involves some key aspects of what emotions, as studied by scientific research, are. In my opinion much of the discussion and debate is of historical interest and not of practical significance. However, important emotion researchers still differ with regard to what they feel is part of emotion and what not, what the functions of emotions are, and other aspects (see, Izard, 2010; Widen & Russell, 2010). At a recent conference, my colleague Andrea Scarantino suggested that the situation is a bit like with cooking recipe where everybody agrees what is in it, but not the exact quantities and how things are combined.

In everyday language, emotions are often seen as synonymous for feelings. However, today, most scientists will also consider other things part of emotions, such as changes in the activity of the body, including the brain. Some of these bodily changes can be seen, such as a blush, or a smile, and others cannot, such as a change in blood pressure. Often researchers will also emphasize that emotions cause changes in action tendencies, for example getting ready to move away, or eliciting explicit actions, such as hugging someone. These different aspects of emotions are also often referred to as
components of emotional reactions. Some researchers in the 20th century believed that emotions are packages or programs that would lead to very stereotypical changes, in the face, in other parts of the body, and in how we feel. However, now that much empirical research has been collected, we understand that the components are only loosely connected. When researchers try to measure emotions, they will assess more than one component.

woman in black long sleeve shirt and black pants sitting on red couchwoman in black long sleeve shirt and black pants sitting on red couch
Duchenne de Boulogne in action
Duchenne de Boulogne in action
turned-on Ferris Wheel miniature
turned-on Ferris Wheel miniature

In scientific terms, emotion is a construct. In using constructs, we give a single name to a bunch of related processes because it facilitates talking about them. Yet, when we look very closely to distinguish one construct from another, it becomes obvious that it is not easy to clearly define the boundaries of such things. This is not unique to emotion, but consider terms such as intelligence, democracy, market, family, or impressionism. Not only does it take more than a paragraph to define such a complex construct, our understanding is in always in flux as new findings appear and scientific exchanges lead to a better understanding of the phenomena we look at.

Emotion, motivation, and cognition are three psychological terms that relate in interesting ways to one another. It is not so easy to separate them in the brain, based on what we have learned in the last decades of neuroscience research. It becomes important to define such constructs when one actually wants to do something with them - for example, building artificial systems are often associated with the terms cognition and intelligence (do not get me started on that one), but rarely with emotion and motivation. And yet, there is an increasing interest in implementing emotions in machines. So what does that mean? The capacity to classify nonverbal behavior in humans? Express emotions in motions and via faces and sounds? Empathy? Should machines feel something? So clearly, it is critical to be straight regarding what somebody refers to when they say that a particular system, say a robot, should have emotions. Or somebody else who says that robots could never have emotions.

In Kappas (2013) I have stated: "Emotions are evolved systems of intra- and interpersonal processes that are regulatory in nature, dealing mostly with issues of personal or social concern." - this sentence is densely packed and refers to many different aspects of the causes and consequences of emotions. If you are curious, I invite you to read the paper here.

Some take-home messages:

  • feelings are not emotions

  • emotions have biological and cultural determinants

  • emotions have evolutionary benefits for the species

  • language and culture influence how we perceive our feelings

  • many emotional processes are social in how they come about, how they manifest, and in their effects

Izard, C. E. (2010). The many meanings/aspects of emotion: Emotion definitions, functions, activation, and regulation. Emotion Review, 2, 363–370.
Kappas, A. (2013). Social regulation of emotion: Messy layers. Frontiers in Psychology, 4, 51. doi: 10.3389/fpsyg.2013.00051.
Widen, S.C. & Russell, J.A. (2010a). Descriptive and Prescriptive Definitions of Emotion. Emotion Review, 2, 377-378.

ChatGPT, why are psychological constructs so difficult to define?

Psychological constructs are often difficult to define because they are abstract concepts that cannot be directly observed or measured. Unlike physical objects or phenomena, which can be easily defined and quantified, psychological constructs are based on subjective experiences, emotions, thoughts, and behaviors that are difficult to measure in a precise and objective manner.

Moreover, psychological constructs can be complex and multifaceted, encompassing a range of different dimensions and factors that interact with each other in complex ways. For example, constructs such as intelligence, personality, motivation, and emotion involve a variety of different cognitive, affective, and behavioral processes that can be difficult to disentangle and define in a clear and concise way.

Furthermore, the meaning of psychological constructs can vary depending on the cultural, social, and historical context in which they are studied, adding another layer of complexity to the process of defining and measuring them.

Overall, the abstract and multifaceted nature of psychological constructs, combined with their subjective and context-dependent nature, makes them inherently difficult to define and measure with precision and objectivity.