Cognitive science, affective computing, intelligent agents, emotional intelligence, appraisal theory and other buzzwords

In order to understand complex topics, humans break them into smaller parts. That process, named analysis, is a common method of scientific and philosophical investigations since before Aristotle.  The exploration of the human mind, as an extremely complex topic, is not an exception.  The sciences, psychology in the first place, traditionally investigate three aspects of mind: cognition (or reasoning), affects (including emotions), and conation (or will). Kant stated it in the „Critique of Judgment“, that those three aspects are irreducible, but such a distinction still doesn’t mean that reasoning and affects lie in separate worlds. In fact, they’re always intertwined.

Cognition was the focus of scientific research on mind during the second half of 20th century. New, multidisciplinary „cognitive science“ has been focused on the set of mental abilities and processes related to knowledge, although its broader scope includes any kind of mental operation or structure that can be studied in precise terms. Cognitive science is closely related with another modern discipline, artificial intelligence, the science and engineering of making intelligent agents, machines and software that can match and even overcome cognitive abilities of a human.

Scientists started to pay more attention to affects at the turn of the centuries. Affective computing, the study and development of systems and devices that can recognize, interpret, process, and simulate human affects, was born. It is an interdisciplinary field spanning computer science, psychology, and cognitive science. The primary motivation for the research in affective computing is the ability to simulate empathy. The robots and virtual personal assistants of the future are expected to interpret the affective state of humans and adapt its behavior to them. The essential role of emotion in both human cognition and perception, as demonstrated by recent neurological studies, indicates that affective computers should not only provide better performance in assisting humans, but also might enhance intelligent agents’ abilities to make decisions, learn, select actions, manage goals, control memory, and act in critical situations.

The meanings of words like “feelings” or “emotions” seem so natural, clear, and direct to us that we cannot see how to start thinking about them. The number of scientific definitions of emotions has grown to the point where counting seems quite hopeless. Contemporary psychologists make distinction between terms „emotions“, „feelings“ and „affects“. Emotions are considered to be a type of affective phenomena, others being moods and attitudes.  The distinction between those phenomena is based on intensity, duration, consequences in terms of stability, impact on behavior choices and so on. Feeling is the emotional experience – just one, subjective component of an emotion.

Basic Emotions theory by Paul Ekman assumes the existence of six basic emotions: anger, fear, disgust, surprise, joy and sadness. In the 1990s, Ekman proposed an expanded list of  emotions, which besides basic emotions incudes amusement, contempt, contentment, embarrassment, excitement, guilt, pride in achievement, relief, satisfaction, sensory pleasure, and shame.

Artificial intelligence has borrowed the term “rational agents” from economics to describe autonomous programs that are capable of goal directed behavior. Most AI researchers take rationality as the essence of the intelligence. The identification of affective with irrational was one of the reasons for science to let affective aspect of mind aside for a long time.  Today, it is widely accepted that most of the affects have rational background and that emotions arise from an evaluation of how environmental events relate to an agent’s plans and goals (appraisal theory). Still, there are some affective phenomena, e.g. „gut feelings“ that are generally regarded as not modulated by conscious thought, and as a reflection of intuition rather than rationality. Those affective phenomena , just like intuition, are not in the scope of science, because they can’t be in every case empirically verified or rationally justified.

The theories about multiple intelligences, or different intelligence modalities, have been around since the ’80s. While it’s still not clear whether „emotional intelligence“ is a valid concept, there’s no doubt that emotions management is an important part of human intelligence.  The intelligent agents are expected to monitor own and other’s emotions, to discriminate between different emotions and label them appropriately, and to use emotional information to guide thinking and behavior.