Defining Emotions | Core Affect & Emotion
October 31st, 2006 by Kevin Capota
This model from Russell tries to explain the eliciting of emotions and is based around core affect. It’s the first model covered in my thesis and I think it has many interesting definitions and points of view that are useful to support my further research. It’s a complex model as it approaches emotions as prototypical emotional episodes. Furthermore it’s purely a psychological model, thus does not include biological factors, evolutionary history, nor the role of society. It took me a while to understand the theory as it’s not a subject I have a lot of experience with. I hope the interpretation went without too many mistakes…
Core affect & two dimensions
Affective states as emotions and moods are grounded to simple structures as feeling good or bad, energized or enervated. “These states – called core affect – influence reflexes, perception, cognition, and behaviour and are influenced by many causes internal and external” (Russell). The two dimensions have been found in many researches and there is a wide consensus on this subject.
There is also criticism about the two dimensions. Core affect covers not only emotions but also includes a range of states not covered by emotions, such as sleepy or placid. Discrete emotions are not fully specified by the dimensions of core affect and the two dimensions are ambiguous. Definition of core affect
Core affect is primitive, universal and simple, it can exist without being labelled, interpreted or attributed to any cause. A person always has core affect. Russell describes the analogy with felt body temperature. Felt body temperature exists before one even uses the words hot or cold, you can note it anytime you want. No cognitive or reflective processes have to occur to feel the temperature.
Function and consequences
he function of core affect is the guiding of cognitive processing by continuously monitoring one’s current state. A change in core affect evokes a search for the cause of this change. This happens according to the principle of mood congruency, the more pleasant core affect is, the more positive evaluative judgements are.
It also involves complex decision making, people generally seek behavioural options that maximize pleasure and minimize displeasure. This means decisions involve predictions of future core affect. Activation influences cognitive performance; optimal level occurs at intermediate level with a higher activation for simpler tasks and a lower activation for difficult tasks (flow). Positive core affect leads to more heuristic and divergent thinking, whereas negative core affect leads to more detailed and critical thinking. Affective quality
Stimuli as objects, events, and places (real or fictional) enter consciousness affectively interpreted. Russell defines affective quality as a property of a stimulus: it is the object’s capacity to change core affect. “Perception of affective quality is a perceptual process that estimates this property.”
Attributed affect
The change in core affect is linked to a perceived cause (person, thing, event, place, or state of affairs) by attributed affect. The cause, or Object is that what makes the person believe is making him feel the current core affect.
Sometimes the cause of the change is obvious, when both the core affect and Object are salient. The attribution process is quick, simple and automatic. It’s like pushing the light switch and the light turns on (without understanding how).
In cases where it requires a search to either the core affect or Object because only one of them is focal, the attribution process will be slower.
The function of attributed affect is to guide attention to and behaviour directed at an Object.
The other (information) function of attributing core affect to an Object is to discover the affective quality of that object. While interacting with an Object for the first time you can notice the change in core affect and attribute affective properties to that Object. A memory of this core affect will then be stored in the schema for that Object. This way attributed allows anticipation of how core affect will change in future events when the just attributed Object is also present.
Emotional Episode
A prototypical emotional episode is an actual occurrent event that fits the prototype for a given category of emotion, such as that labelled fear, very well. An emotional episode is an actual occurrent event that comes close enough to count as an instance of the category even if not an exemplary one.
The mental model (or prototype) of fear exists in the mind of an observer (as the intension of the mental category). A prototypical emotional episode of fear exists in the external world (as the extension of the mental category). It is temporal and has causal connections with an Object. An actual emotional episode is long-lasting and subject to change.
Russell’s framework does not see the eliciting of an emotion as being triggered by the antecedent event. Triggering is replaced with an attribution process in which the Object can be sought and analyzed. In a prototypical emotional episode, antecedent and Object largely coincide. The framework presented in the picture below shows the different components shaping a specific emotion. The components influence each other but they cannot be ordered in a sequence “Emotional life consists of the continuous fluctuations in core affect, in pervasive perception of affective qualities, and in the frequent attribution of core affect to a single object, all interacting with perceptual cognitive, and behaviour processes.”.

The components of an emotional episode
A prototypical emotional episode includes a change in core affect. One has to feel great displeasure and activation to be afraid.
The antecedent is perceived to its affective quality
The attribution process is an important factor as described above.
Appraisals go beyond simple pleasantness and arousal rate.
The existence of physiological and expressive signatures (actions) for specific emotions remains a viable but unconfirmed hypotheses. Core affect already has a general mobilization in its vertical axis of activation. The valence dimension is also related to action, pleasant Objects invite to approach where unpleasant Objects withdraw cause withdrawal. Still, core affect is only a general guide to behaviour.
People experience discrete emotions, they do not just feel pleasant and unpleasant or aroused to a certain degree. Russell interprets these states as emotional meta-experience. It is the categorization of one’s state, to perceive oneself as afraid, angry, jealous or an other discrete emotion. This helps cognitive processing and, later, in memory by categorizing one’s life events in order to learn from them.
Emotion regulation is the attempt of self control during an emotional episode of the different influencing factors, for various goals including presentation both to self and others. Emotional meta-experience also mediates in this as an evaluation factor to oneself with respect to the rules important to that person.
Concluding
This is the model proposed by Russell. I’m not sure about fully using it as it encounters the subject in a way that is far too complex for my goals. However, the model offers some useful definitions and processes related to the eliciting of emotions. I’ll surely relate back to this model in a latter stage.
Hi Kevin,
You blog contains very useful information which I found mostly related to the term “Kansei” in Japanese. I am also keen in reading literatures on Theories and Models that you’ve summarized and/or elaborated in your writing. I am happy if you could share those literatures (Russell, Van Gorp, Sylvia, etc) with me.
Best regards
Zaidi
Nagaoka JAPAN
20061227 2329hrs
Hi Zaidi,
Thanks for your comment and indeed many theory is related to Kansei as it is one of the best-known methods of ‘design for emotion’. I shall see what I can do for you when I’m back at the office (january 2nd) relating to the theories.
Kind regards,
Kevin
Thanks Kevin,
Have a wonderful time and see you in the brand new year 2007(^_^)