What Must a Theory of Emotion Explain?

What is fear?

Is it the feeling? A change in the body? The urge to escape? A facial movement? A pattern of neural activity? An interpretation that something is dangerous?

Emotion researchers study all of these phenomena. Different theories assign them different roles.

That is precisely why the scientific problem cannot begin by choosing one of them as the answer.

Before asking which theory of emotion is right, we first need to ask what a sufficiently broad theory is required to explain.

An Emotion Label Is a Starting Point, Not a Theory

Words such as fear, anger, sadness and joy help organise emotional life.

They allow people to communicate about recurring experiences, researchers to define experimental conditions, and scientists to investigate patterns that may recur across situations and people.

But naming an episode does not explain how it arose.

Calling something "anger" does not tell us what processes produced the experience, what happened in the body or brain, why particular actions became likely, why the episode became intense or remained mild, or why another person in a similar situation might have responded differently.

A label can therefore identify a phenomenon for investigation without identifying its mechanism.

This matters because research finds both regularity and variation within familiar emotion categories.

Episodes called fear or anger are not randomly unrelated. Across several experimental traditions, researchers find recurring patterns in experience, physiology, behaviour and neural activity.

But those patterns are not perfectly invariant.

Not every episode assigned the same emotion label produces an identical physiological response, facial movement, neural pattern or behaviour.

Average differences can exist while individual instances vary.

The opposite conclusion would also go too far. Variation within a category does not by itself show that the category is scientifically meaningless.

The problem is to explain the structure of both.

An emotion label tells us where investigation can begin.

It does not tell us how the phenomenon works.

Emotional Episodes Span More Than One Level

Across emotion science, researchers measure and theorise about subjective experience, bodily change, neural activity, attention, interpretation, expression, action and context.

A broad theory must explain how whichever of these processes prove relevant are related within emotional episodes.

As established in Mind, Brain and Body Form One System, these processes can be distinguished analytically without being assigned to separate systems.

The brain is not having one event while the body has another and the mind observes them from somewhere else. Neural, bodily, cognitive and experiential processes occur within the same interacting organism.

But this does not mean that every emotional episode contains one fixed package of components.

Nor does it mean that everything moves together in perfect synchrony.

In controlled studies, subjective experience, expressive behaviour and physiological responses can show meaningful coordination while still providing partly different information. The strength of that coordination can also vary with the emotion, situation and intensity being studied.

This creates an important explanatory problem:

Why do different aspects of an emotional episode sometimes cohere and sometimes come apart?

Simply measuring several processes does not explain their organisation.

Experience is especially important here.

Emotion is encountered partly through changes in the character of experience. A sufficiently broad theory that explained bodily and neural processes while leaving the character and variation of emotional experience unexplained would therefore leave part of its target unresolved.

But subjective experience cannot simply replace the rest of the problem either. What someone feels does not by itself explain bodily organisation, neural activity, action or context.

The same caution applies to bodily sensing.

As explored in Interoception, information about internal bodily states is associated with emotional experience and may contribute to it. But this does not establish that emotion is simply a readout of the body.

The role of bodily change and bodily sensing is itself something an adequate theory must explain.

Finding a Correlate Does Not Finish the Explanation

Emotion research has identified many informative correlates.

Physiological responses change during emotional episodes. Facial and bodily behaviour can carry emotion-relevant information. Neural activity differs across experimental emotional conditions. People report changes in feeling and motivation.

These findings constrain theories.

But they do not all answer the same scientific question.

As established in Correlation, Prediction, Causation and Mechanism, a correlate is not automatically a cause, and identifying a cause is not the same as specifying a complete mechanism.

Consider physiology.

Meta-analytic and review evidence suggests that different emotional conditions can show systematic physiological differences on average. At the same time, responses vary substantially within categories and overlap between them.

Physiological responses therefore show some systematic emotion-related patterning without supplying one invariant bodily fingerprint for every instance.

A change in heart rate can be informative without being fear itself. Similar changes can occur in other emotions, physical exertion or non-emotional states. Finding the change does not establish what role it played in producing or maintaining the episode.

Neural evidence creates a related challenge.

Some multivariate and meta-analytic studies can distinguish experimental emotion categories from distributed patterns of brain activity above chance. Neural data therefore contain category-relevant information.

But this does not imply that each emotion resides in one dedicated brain region.

Emotion-related neural activity is distributed and overlapping, and many of the same neural systems participate in processes extending beyond any single emotion category.

A brain pattern can therefore be scientifically informative without constituting a complete theory of the emotional episode.

Facial behaviour provides another example.

People can extract emotion-relevant information from faces, including above chance across cultural boundaries in some experimental paradigms. Yet similar facial movements can occur in different situations, and instances assigned the same emotion label need not produce one invariant facial configuration.

Context matters to what a facial movement means and how another person interprets it.

The broader principle is simple:

A marker can be informative without being the thing it marks.

An adequate theory must explain what role these different correlates play and how they relate to the rest of the episode.

The Same Emotion Can Take Different Forms

Consider two episodes both described as anger.

One person becomes loud and confrontational. Another becomes quiet.

One experiences intense physiological activation. Another reports a colder, more controlled experience.

One episode disappears quickly. Another persists.

The circumstances surrounding them may also be very different.

Research across physiology, neural activity and expressive behaviour shows substantial variation within nominal emotion categories.

That variation creates a real explanatory problem.

But it does not settle the theoretical interpretation.

Variation could reflect context, intensity, individual differences, different subtypes, regulation, measurement, different causal histories or limitations in the categories researchers are using. Different theories make different claims about which of these possibilities matter.

This piece does not need to decide among them.

The principle established in The Same Outcome Can Have Different Causes is enough to show why causal homogeneity cannot simply be assumed.

Apparently similar emotional outcomes may sometimes arise through different combinations of contributing processes.

But "sometimes" matters.

The existence of variation does not prove that every instance has a different mechanism. Regularities may be substantial and scientifically important.

An adequate theory must therefore explain which features remain regular within an emotion category, which vary, under what conditions, and why.

Both the regularity and the variation belong to the explanatory problem.

Different Emotions Can Share the Same Features

The reverse problem is equally important.

Different emotional categories can share many features.

Fear and anger can both involve elevated arousal, heightened attention and preparation for action. Different emotional states can recruit overlapping neural systems. Similar physiological responses may occur across several categories.

So what produces differentiation despite those shared processes?

Again, two conclusions would be premature.

The first is that every emotion must possess one biological or behavioural feature found nowhere else.

The second is that overlap means emotion categories contain no meaningful structure.

Research supports a more complicated picture: measurable differentiation can coexist with substantial overlap.

A sufficiently broad theory therefore needs some account of both.

Why are different emotional episodes similar in some respects?

Why do they differ in others?

How do shared bodily, neural or behavioural processes relate to differences in experience, meaning and action?

No single measurement level can answer those questions by itself.

The explanandum is multi-level.

The Same Situation Can Lead to Different Emotional Outcomes

Now consider receiving criticism in front of other people.

One person becomes angry.

Another feels ashamed or hurt.

Someone else regards the criticism as fair and experiences relief.

Another person barely reacts.

The outward event may look similar. The emotional outcome does not.

As established in Human Behaviour Has Multiple Causes, human outcomes can emerge from interacting conditions rather than one isolated influence.

Potentially relevant factors in an emotional episode include the person's goals, prior learning, relationship with the other person, current bodily state, expectations, interpretation of the event, available actions and the consequences attached to those actions.

The environment matters too. Actual threats, resources, opportunities and social consequences can change what a situation permits and what is at stake.

Different theories assign these influences different roles.

The important point here is that a simple mapping—

event → fixed emotion

—cannot be assumed.

The principle established in The Same Cause Can Produce Different Outcomes helps clarify the problem: the same kind of event or causal influence can contribute to different outcomes under different conditions.

Research on appraisal and context provides substantial evidence that emotional responding varies with relations between the person and the situation.

That evidence establishes an explanatory constraint.

It does not establish appraisal theory—or any other theory—as the complete account of emotion.

The question remains:

Why does a similar situation become anger for one person, shame for another, relief for another and little more than information for someone else?

Different emotion theories can answer that question differently.

Any sufficiently broad account has to confront it.

Emotion Develops Through Time

Emotional episodes are not merely snapshots.

They develop.

Intensity may rise or fall. A state may persist, fluctuate or transform. Emotional responding can continue after the event that initiated it has ended, or change rapidly when new information becomes available.

Research on emotion dynamics examines precisely these trajectories. Emotional states differ in persistence and change, and people differ in how strongly a current emotional state predicts what they experience later.

This means that identifying what is happening at one moment cannot explain the entire episode.

A broad account may also need to explain why an emotional response emerged when it did, what altered its intensity, what sustained it, what changed its direction and what contributed to its resolution.

Consider fear following an unexpected sound.

For one person, the response may dissipate rapidly once the source becomes clear.

For another, heightened arousal may persist.

Another episode may shift toward relief, embarrassment or anger as the situation changes.

No universal sequence needs to be assumed.

The fact that trajectories differ is itself part of what requires explanation.

An emotional episode is an event in time, not merely a state captured at one instant.

Variation Across People, Relationships and Cultures Is Part of the Phenomenon

Variation extends beyond individual episodes.

People differ.

Relationships change the meaning and consequences of events.

Cultural environments differ in language, norms, practices, expectations and patterns of social interaction.

These differences can shape emotional experience, expression and interpretation.

Cross-cultural evidence again resists simple conclusions.

In some experimental paradigms, people recognise emotion-relevant facial information across cultural boundaries at above-chance levels. Research also finds systematic cultural differences, including advantages for interpreting signals from more familiar cultural environments.

Other work suggests that cultures can differ in the situations associated with particular emotions, patterns of appraisal and attention, norms governing expression, and ways emotional experiences are categorised or communicated.

The amount of similarity can therefore depend on which part of the emotional process is being measured.

Culture should not simply be treated as decoration added to an otherwise fully specified universal emotion system.

But cultural variation does not establish that emotion is purely a social invention either.

The same discipline applies to relationships.

Anger toward a stranger, a child, an employer or a close partner may share important features while differing in expression, action, consequences and experience because the interpersonal situation itself differs.

Individual differences add another layer. Learning history, goals, bodily state, developmental history and other characteristics may influence emotional responding, although which factors matter and how they matter remains an empirical question.

An adequate theory therefore has to confront a difficult combination:

structured regularity together with structured variation.

It must explain what generalises, what changes systematically, and under which conditions.

What Must an Adequate Theory Account For?

Emotion science has already established substantial empirical constraints.

Emotional experience matters, but it does not exhaust the phenomenon.

Bodily and interoceptive processes matter, but bodily arousal alone is not a complete explanation.

Neural data contain emotion-relevant information, but neural correlates are not definitions.

Expression and behaviour carry information without functioning as transparent readouts of internal state.

Different aspects of emotional episodes sometimes cohere and sometimes dissociate.

Nominal emotion categories contain both regularities and variation.

Different categories can share physiological, neural and behavioural processes.

Similar situations can contribute to different emotional outcomes.

Emotional episodes develop through time.

And emotional patterns show both commonality and systematic variation across people, relationships and cultures.

These are not necessarily an exhaustive inventory of everything emotion science will ever need to explain. They are robust constraints and recurring explanatory problems that any account claiming broad adequacy must confront.

Specialised theories can legitimately explain only part of this landscape.

A model of autonomic regulation does not fail because it does not explain cultural emotion concepts. A theory of facial communication does not need to become a complete theory of neural implementation.

The mistake would be to treat success on one part of the problem as sufficient evidence for a comprehensive theory of emotion.

This distinction also changes how theoretical disagreement should be interpreted.

Before two theories are treated as direct competitors, three questions matter.

Are they explaining the same phenomenon?

One theory may focus on experiential differentiation while another explains physiological organisation.

Are they operating at the same level?

A cognitive account and a neural implementation may be related without being alternatives.

Do they actually make incompatible claims?

Different terminology can hide agreement, while similar terminology can conceal genuine disagreement.

This is why the explanatory target has to be established before the theories are compared.

Emotion is not scientifically difficult because nothing is known.

Researchers have identified substantial regularities in experience, physiology, neural activity, behaviour and social interaction, alongside equally important variation across instances and contexts.

The challenge is explaining how those findings belong together.

That is what any theory claiming to explain emotion broadly must confront.

Only once that problem is clear does the next question become meaningful:

What are the major families of emotion theory, which parts of this explanatory landscape does each attempt to explain, and where do their claims genuinely differ?

That is the task of Families of Emotion Theory.