Someone says your name from behind you.
Before you turn, the body has already changed.
Your shoulders lift. The heart seems louder. The room narrows for a fraction of a second. Then you recognize the voice.
A friend.
The body changes again.
What happened in those two seconds?
Was the emotion the bodily response?
The feeling of alarm?
The recognition of a possible threat?
The action tendency to orient?
The meaning created when the voice became familiar?
Emotion science has spent more than a century arguing about questions like these because the phenomenon is obvious in ordinary life and difficult to define scientifically.
Definition
For this Knowledge Library, emotion refers to a temporally unfolding episode in which changes in bodily state, perception, attention, appraisal or meaning, action readiness, expression and conscious experience may become coordinated around something significant to the organism.
That definition is deliberately plural and cautious.
Not every emotional episode contains every component in the same way.
Researchers disagree about which components are necessary, which are causes, which are consequences and whether emotion categories such as anger or fear have universal biological signatures.
A definition that hides that disagreement would be falsely precise.
Emotion is not the same as feeling
In ordinary language, emotion and feeling are often interchangeable.
In scientific writing, they can be separated.
Feeling usually refers to the consciously experienced aspect of an emotional or affective state.
A person can report feeling anxious, relieved, irritated or joyful.
But emotion research may also examine physiological changes, facial or bodily action, attention, memory and behavior that are not exhausted by the verbal report.
Some theories place conscious feeling near the center of emotion.
Others treat it as one component of a broader process.
The distinction is useful as long as it is not mistaken for a universally agreed ontology.
Emotion is not the same as affect
Affect is often used for more basic dimensions of valence and arousal, such as pleasant versus unpleasant and activated versus deactivated experience.
Affect can be present without a fully categorized emotion.
A person may feel vaguely bad and activated before identifying anger, fear, embarrassment or illness.
Different research traditions use the term affect differently, so authors should define it rather than assume shared meaning.
The next canonical entry in this hub treats affect separately for this reason.
Emotion is not the same as mood
A mood is generally more diffuse, longer-lasting and less tightly linked to a specific eliciting object or event than an emotion episode.
Anger at a particular insult can be relatively brief and object-directed.
An irritable mood can color perception for hours without one clear target.
The boundary is not perfect.
Emotions can persist, moods can have causes, and people use the words loosely.
Still, separating episode from background state helps keep claims precise.
Emotion regulation is not emotion
Emotion regulation concerns processes that influence which emotions arise, when they arise, and how they are experienced, expressed or acted upon.
It is therefore a different construct.
A person may regulate by changing a situation, shifting attention, reappraising meaning, accepting the experience, suppressing expression or altering behavior.
The effectiveness of a strategy depends on context, goals, timing and individual differences.
Johan Bjureberg's 2026 review from Karolinska Institutet examines the role of emotion regulation in clinical interventions and is especially relevant to the Institute's Nordic research context.
But clinical emotion-regulation evidence should not be used to define emotion itself.
There is no single theory of emotion
Several major theoretical families continue to shape the field.
Basic emotion approaches
Basic emotion theories propose that at least some emotion categories reflect evolved systems with characteristic functions and partially patterned expressions, physiology or neural organization.
The exact claims differ across theorists.
A strong version expects relatively distinct emotion kinds.
More flexible versions allow substantial variation while retaining evolved functional organization.
Appraisal approaches
Appraisal theories emphasize how an event is evaluated in relation to goals, novelty, control, agency, certainty and other dimensions.
Different patterns of appraisal are proposed to help generate different emotional responses.
This family of theories makes meaning and situational evaluation central.
Constructionist approaches
Constructionist theories argue that emotions are not fixed packages detected inside the body.
The Theory of Constructed Emotion, associated with Lisa Feldman Barrett and colleagues, proposes that emotion categories are dynamically constructed through processes involving allostasis, prediction, sensation and learned conceptual knowledge.
Barrett and colleagues' 2025 paper explicitly defends this theory against attempts to reduce it to a theory of feeling alone.
Dynamical and integrative approaches
Other researchers emphasize emotion as a dynamic, context-sensitive system in which bodily, cognitive, social and environmental processes interact over time.
Some recent work seeks bridges across theoretical traditions without pretending that deep disagreements have disappeared.
Van Heijst, Ploeger and Kret's 2025 contribution is one example of an attempt to foster connection in emotion-theory debates, followed by forceful criticism from constructionist theorists.
The debate is active.
Interoception is part of the story
Interoception concerns sensing and representing signals from within the body.
Heartbeats, breathing, gut activity, temperature and other bodily processes contribute to the information available during emotional episodes.
Greenwood and Garfinkel's 2025 Annual Review of Psychology article synthesizes evidence on interoceptive mechanisms and emotional processing.
The relationship is not one-to-one.
A racing heart does not uniquely identify fear.
The same physiological change can occur during exercise, excitement, anger, illness or anticipation.
Bodily signals acquire meaning in context.
This is one reason single-biomarker accounts of emotion are inadequate.
Emotion unfolds in time
An emotion is not a static label attached to a face.
It develops.
Something is noticed or anticipated. Bodily state shifts. Attention changes. Meaning is assigned or revised. An impulse becomes stronger or weaker. The social environment responds. The person may regulate, suppress, express or reappraise.
The temporal pattern matters.
A measure taken ten seconds after an event can capture something different from a retrospective report given the next day.
Emotion research therefore needs time-sensitive methods.
Culture changes emotional life without making biology irrelevant
Cross-cultural research complicates any simple choice between universal biology and cultural construction.
Human beings share bodies, developmental needs and many broad motivational problems.
They also learn culturally specific categories, display rules, values, narratives and regulatory norms.
A 2025 systematic review and meta-analysis by Chen and colleagues examined emotion regulation and mental health across cultures, highlighting the importance of cultural context in how strategies function and relate to outcomes.
The existence of cultural variation does not imply that physiology is unimportant.
The existence of shared biology does not imply that emotion categories are expressed, valued or regulated identically everywhere.
Both levels matter.
Facial expression is informative but not a transparent readout
People often infer emotion from faces.
Faces carry information.
But a facial movement is not a universal barcode for an inner state.
Context, culture, task demands and perceiver expectations affect recognition and interpretation.
The same expression can be read differently depending on the surrounding situation.
This has practical consequences for technologies that claim to infer emotion automatically from facial video.
A system may classify facial configurations without validly identifying a person's internal emotion.
The inferential leap needs evidence.
Physiology is similarly many-to-many
Heart rate, skin conductance, respiration, hormones and neural measures can all change during emotional episodes.
None is a universal single-channel detector of anger, fear, sadness or joy.
Different emotions can share physiological features.
The same nominal emotion can show different physiological patterns across people and situations.
This variability is not merely noise to be wished away.
It may be part of the phenomenon theories are trying to explain.
Self-report is indispensable and limited
If scientists want to know what an experience feels like, self-report is difficult to replace.
But language shapes reporting.
People differ in emotional vocabulary, introspective access, willingness to disclose and cultural norms.
Retrospective reports are also affected by memory.
A strong emotion study therefore often benefits from multiple methods: experience sampling, behavior, physiology, context, observation and carefully designed self-report.
Multi-method evidence does not guarantee truth, but it reduces dependence on one imperfect window.
Measurement follows theory
There is no theory-free emotion measure.
If a researcher codes six facial expressions, the design already assumes something about categories.
If they measure valence and arousal, they assume a dimensional representation.
If they collect appraisals, they foreground meaning evaluations.
If they model time-varying physiology and context, they foreground dynamics.
Measurement choices therefore encode theoretical commitments.
Researchers should make those commitments visible.
Emotion has functions, but “every emotion is useful” is too simple
Emotions can organize attention, learning, action and social communication.
Fear can promote avoidance or preparation. Anger can mobilize confrontation. Shame can reshape social behavior. Joy can broaden exploration and affiliation.
But functional accounts should not become moral slogans.
An emotion can be understandable and still contribute to harmful behavior.
A response that was adaptive in one context can become costly in another.
Clinical distress often involves persistent patterns that cannot be reduced to the idea that an emotion merely needs to be “listened to.”
Scientific respect for emotion does not require romanticizing every emotional impulse.
The person and the environment form a system
An emotion is shaped by what the person brings and by what the environment affords.
Sleep, stress, learning history, relationships, hormones, bodily state, language, social norms and immediate context can all matter.
Institutions matter too.
A workplace that punishes uncertainty may produce different emotional dynamics from one that permits questions. A school that humiliates error changes the meaning of evaluation. A platform that amplifies outrage changes which emotional signals are socially rewarded.
Emotion science therefore belongs not only to individual psychology but to the design of environments.
Appraisal research shows why the same event can produce different emotions
Two people can receive the same email and react differently.
One sees a threat to status. Another sees useful feedback. A third sees an unfair accusation. The sensory input overlaps, but the meaning of the event in relation to goals, control and responsibility differs.
Appraisal theories formalize this observation.
They examine dimensions such as novelty, pleasantness, goal relevance, agency, certainty, controllability and compatibility with norms. Different appraisal patterns are proposed to organize different emotional responses.
The evidence does not imply that people consciously calculate these dimensions before feeling. Appraisal can be rapid and partly automatic.
Its importance is conceptual: emotional response depends not only on what happened but on what the event means for this organism in this situation.
Development changes the emotional repertoire
Infants do not begin life with an adult vocabulary for jealousy, nostalgia, humiliation or moral outrage.
Emotional life develops through maturation, caregiving, language, learning and culture.
Children learn not only labels but distinctions: the difference between fear and excitement, disappointment and anger, guilt and shame. Those categories can change what they notice, report and regulate.
Developmental change also affects control. A strategy that requires abstract reappraisal may be unavailable to a young child even when simpler attentional or interpersonal regulation is possible.
This developmental perspective is important for education. Teaching emotional literacy is not merely handing children a larger list of words. It involves connecting bodily cues, situations, meanings, action tendencies and socially appropriate response.
Emotion concepts can help and constrain
Words give people tools for organizing experience.
More differentiated emotion concepts may help a person distinguish “I feel bad” from grief, envy, fear, shame or exhaustion, each of which may imply different needs and actions.
But labels can also overfit experience.
Once a category is available, people may interpret ambiguous sensations through it. Cultural narratives can make some emotions easier to recognize and others harder to admit.
Emotion concepts are therefore part of emotional intelligence without being transparent mirrors of fixed inner states.
They are tools for making meaning.
Social regulation precedes and accompanies self-regulation
Emotion is often described as if it occurs inside an isolated individual.
Human emotional life is deeply interpersonal.
Infants rely on caregivers to regulate arousal. Adults use conversation, touch, proximity, reassurance, conflict, humor and shared ritual to change one another's emotional states. Relationships can buffer stress or amplify it.
This matters because “self-regulation” can become an overly individualistic ideal.
A worker in a threatening environment cannot solve every emotional problem through breathing or reappraisal. A child cannot be expected to regulate effectively in a classroom organized around humiliation. A patient may need relational safety as much as a technique.
Regulation capacity is partly personal and partly environmental.
Emotion changes cognition without simply defeating it
Popular language often places emotion and reason in opposition.
The research picture is more complex.
Emotion can bias attention, memory and judgment, but it can also signal relevance, mobilize effort and support learning. Decision-making without affective valuation would not automatically become more rational because choices require some representation of significance and consequence.
The useful distinction is not emotion versus cognition.
It is whether the emotional process is helping the person track the situation accurately and act in line with longer-term goals.
Strong emotion can narrow attention when broad search is needed. It can also focus attention on a real threat that detached analysis would underestimate.
Context determines whether the influence is useful.
Clinical categories should not be projected onto ordinary emotional variation
Sadness is not identical to depressive disorder.
Fear is not identical to an anxiety disorder.
Anger is not a diagnosis.
Clinical conditions involve patterns of duration, severity, impairment and associated symptoms that cannot be inferred from the presence of an emotion alone.
This boundary protects ordinary emotional life from pathologization and protects clinical language from dilution.
It also matters for workplace and educational technologies that claim to detect “stress,” “depression” or “anxiety” from brief behavioral signals. Screening, diagnosis and momentary affect classification are different tasks with different evidentiary requirements.
The ethics of emotion inference are part of the science
As cameras, wearables and AI systems make behavioral and physiological data easier to collect, organizations increasingly claim to infer emotional states.
Even if predictive accuracy improves, ethical questions remain.
Did the person consent? Is the inference necessary? What happens when the model is wrong? Can a worker challenge a label? Does the system perform equally across cultures and disabilities? Is a probabilistic classification being treated as fact?
Scientific uncertainty and governance are connected here.
A weak model used casually in research creates one kind of problem. The same model used to discipline employees, assess students or triage patients creates another.
Evidence standards should rise with consequence.
What the evidence does not show
Current evidence does not show that:
- every emotion has one unique facial expression;
- a single physiological measure can reliably identify a person's emotion across contexts;
- all scientists agree on a single definition or theory of emotion;
- emotional experience is either purely biological or purely cultural;
- conscious feeling exhausts the emotional process;
- emotion and emotion regulation are the same construct;
- an emotion being understandable means acting on it is wise;
- one regulation strategy is universally adaptive;
- AI systems can infer private emotional states accurately simply from a face or voice;
- cultural differences make cross-cultural emotion science impossible.
These limits are central to responsible interpretation.
Institute synthesis: emotion as information without infallibility
Within Inner Technology, the Institute treats emotion as information without infallibility.
This is a synthesis, not an established scientific definition.
Emotional episodes can reveal what the organism is detecting, valuing, predicting or preparing for.
But the information can be incomplete, historically shaped, contextually mismatched or based on mistaken interpretation.
The mature capacity is neither suppression nor obedience.
It is the ability to notice emotion, understand what evidence it may contain, regulate when necessary and preserve enough reflective distance to decide what to do.
That framing keeps emotion inside human intelligence without turning feeling into unquestionable truth.
From individual emotion to civilization
Societies also organize emotional life.
Media systems amplify some emotions more than others. Institutions define which emotions can be shown. Political and commercial actors learn which emotional cues move attention. Schools teach, implicitly or explicitly, what children should do with fear, frustration and shame.
An emotion science that remains confined to the individual risks missing these environments.
At the same time, social explanation cannot erase the body.
Civilization is built through nervous systems that become tired, vigilant, attached, hopeful, threatened and moved.
The voice behind you is familiar now.
What began as a bodily jolt has become recognition, relief and perhaps affection.
No single measure contains the whole event.
That is not a failure of science.
It is a reason for precision about what each theory and method can actually see.
Research references
- Barrett LF, Atzil S, Bliss-Moreau E, et al. The Theory of Constructed Emotion: More Than a Feeling. Perspectives on Psychological Science. 2025;20(3):392-420. https://doi.org/10.1177/17456916251319045
- van Heijst K, Ploeger A, Kret ME. Beyond Right and Wrong: Fostering Connection in Emotion Theory Debates. Perspectives on Psychological Science. 2025. https://doi.org/10.1177/17456916251319042
- Greenwood BM, Garfinkel SN. Interoceptive Mechanisms and Emotional Processing. Annual Review of Psychology. 2025;76:59-86. https://doi.org/10.1146/annurev-psych-020924-125202
- Chen MS, et al. Emotion regulation and mental health across cultures: a systematic review and meta-analysis. Nature Human Behaviour. 2025. https://www.nature.com/articles/s41562-025-02168-8
- Bjureberg J. The Role of Emotion Regulation in Clinical Interventions. Annual Review of Clinical Psychology. 2026;22:105-133. https://doi.org/10.1146/annurev-clinpsy-061324-072853
- Barrett LF. How Emotions Are Made: The Secret Life of the Brain. Houghton Mifflin Harcourt; 2017.
- Gross JJ. Emotion regulation: current status and future prospects. Psychological Inquiry. 2015;26(1):1-26. https://doi.org/10.1080/1047840X.2014.940781