Appetitive Behavior

Abstract

Appetitive behavior is a form of animal behavior: the variable, flexible, goal-directed phase of a motivated act—the searching, approaching, and working that brings an animal into the situation where a fixed consummatory act, such as eating or copulation, can occur. The distinction between this searching phase and the stereotyped end act was drawn by Wallace Craig in 1918 and became a cornerstone of ethology and the neuroscience of motivation. This article sets out what separates appetitive from consummatory behavior, how the MeSH classification divides it, and the modern account in which the mesolimbic dopamine system energizes appetitive “wanting” without generating the “liking” of the consummatory phase. A worked example computes an effort-based choice, and three interactive demonstrations let the reader step an act from search to consummation, dissociate wanting from liking, and shift choice under changing dopamine.

Keywords: appetitive behavior, incentive salience, motivation, mesolimbic dopamine, consummatory behavior

A hungry rat does not eat; first it searches. It rears, sniffs, crosses an open field, presses a lever, and only then, having reached food, does it bite and chew. The searching is as much a part of feeding as the chewing, yet the two differ in almost every respect: the search is variable, sensitive to circumstance, and shaped by learning, while the bite is brief, rigid, and nearly the same on every occasion. Appetitive behavior is the name for that first, flexible phase—the behavior that appetite drives when its object is not yet to hand.

This article treats appetitive behavior as a problem in the organization of motivated action. It begins with Wallace Craig's original appetitive–consummatory distinction and the hierarchical model of instinct it anchored, turns to the modern dissociation of wanting from liking and the mesolimbic circuitry that implements it, and closes with the effort-based view of dopamine and the open questions now driving the field. Throughout, the pursuit of a goal is read as a computation, not a reflex.

What Appetitive Behavior Is

Appetitive behavior is the introductory, variable phase of an instinctive or motivated sequence: the behavior through which an animal seeks out and approaches the stimulus that will release a terminal act. Wallace Craig, in the paper that named the concept, defined an appetite as a state of agitation and restless activity that continues until a particular stimulus—the appeted stimulus—is received, whereupon the agitation ceases in a consummatory act and a state of relative rest follows (Craig, 1918). An aversion is the mirror image: a state of disturbance that persists until a disliked stimulus is removed. The appetitive phase is thus defined functionally, by what it is for, not by any particular movement, and the same hunger can express itself as stalking in a cat, lever-pressing in a rat, or foraging flight in a bird.

Two features distinguish appetitive from consummatory behavior. The first is variability: appetitive behavior is plastic, adjusting moment to moment to the layout of the environment and improving with learning, whereas the consummatory act is comparatively stereotyped and species-typical (Tinbergen, 1963). The second is that the appetitive phase is where most of the interesting decisions are made—where to go, how hard to work, which of several goals to pursue—so it is the phase that carries the motivational and cognitive load of a behavior. Craig's distinction survived the eclipse of classical instinct theory precisely because this functional division maps cleanly onto mechanisms the brain really does separate (Berridge, 2004).

Types of Appetitive Behavior

In the Medical Subject Headings vocabulary, appetitive behavior is indexed under its parent descriptor animal behavior and divided into three narrower descriptors, shown in Table 1. These subtypes are indexing categories, not a theory of how motivated behavior is built: MeSH is a classification for retrieving literature, so the list reflects how studies are catalogued rather than a claim that nature is carved at exactly these joints. The categories are also not mutually exclusive—grooming can be predatory preparation, and host-seeking is a specialized form of the search that defines appetitive behavior generally—so a single episode of behavior may belong to more than one.

Table 1. Direct subtypes of appetitive behavior in the MeSH classification (tree F01.145.113.111).

The Appetitive–Consummatory Distinction

Niko Tinbergen gave Craig's distinction a formal structure. In his hierarchical model of instinct, motivated behavior is organized as a branching cascade: a high-level drive, once activated, releases appetitive behavior that narrows through successive levels—from general restlessness, to movement toward an appropriate habitat, to search for a specific object—until it arrives at the fixed action pattern that consummates the sequence (Tinbergen, 1963). The appetitive behavior at each level is the searching that leads to the next, more specific level; the consummatory act is the terminal, stereotyped movement at the bottom of the branch. Figure 1 sketches this descent.

Sexual behavior became the classic testbed for the distinction because its two phases are so cleanly separable. Barry Everitt showed in male rats that the appetitive pursuit of a mate and the copulatory acts themselves depend on partly different neural systems, so that a lesion or treatment can abolish one while sparing the other (Everitt, 1990). Benjamin Sachs argued that sexual arousal in males is best defined contextually, as a central state that appetitive measures index indirectly, cautioning against reading motivation straight off any single response (Sachs, 2007). Gregory Ball and Jacques Balthazart later asked how far the appetitive–consummatory division should be pressed in the neuroendocrine control of sexual behavior, noting that the categories are useful but can mislead when treated as sharply bounded rather than as ends of a continuum (Ball & Balthazart, 2008). The demonstration below steps a single episode from the variable appetitive search through to the fixed consummatory act.

Wanting and Liking

The most consequential modern development is the dissociation of the appetitive pull of a reward from the pleasure of consuming it. Kent Berridge and Terry Robinson argued that reward is not one process but at least three—learning, liking (the hedonic impact of a reward), and wanting (the motivation to obtain it)—and that these are separable in the brain (Berridge & Robinson, 1998). Wanting, in their account, is incentive salience: a motivational property that the brain attributes to a reward and its cues, making them attractive, sought, and able to trigger appetitive behavior. It is the mechanistic heart of the appetitive phase. Liking is the hedonic reaction that belongs to the consummatory phase, measurable in the stereotyped facial responses that sweet tastes elicit across species.

The striking claim is that mesolimbic dopamine implements wanting but not liking. Manipulations that raise or lower dopamine change how hard an animal will work for a reward and how strongly cues attract it, while leaving the hedonic reaction to the reward itself essentially unchanged (Berridge & Robinson, 2016). Pleasure, by contrast, depends on small, fragile hedonic hotspots—opioid- and endocannabinoid-sensitive sites in the nucleus accumbens shell and ventral pallidum where Berridge and Morten Kringelbach located the generation of liking (Berridge & Kringelbach, 2015). Ileana Morales and Berridge extended the same wanting–liking framework to food reward, showing how the two can be driven apart in ways that matter for overeating (Morales & Berridge, 2020). The demonstration lets the reader move dopamine up and down and watch wanting swing while liking holds steady.

The Brain's Seeking System

The appetitive phase has a recognizable neural signature. James Olds and Peter Milner's discovery that rats will work tirelessly to deliver electrical stimulation to certain brain sites gave the first physiological handle on appetitive approach, revealing circuits whose activation is itself a goal (Olds & Milner, 1954). Jaak Panksepp named the broader circuit the SEEKING system: an ancestral, dopamine-energized network that drives exploration, foraging, and the eager pursuit of all manner of rewards, and whose activity feels, in his account, like interest and anticipation rather than consummatory pleasure (Alcaro & Panksepp, 2011). Satoshi Ikemoto and Panksepp had earlier placed nucleus accumbens dopamine at the center of this appetitive drive, interpreting it as a general engine of reward-seeking rather than a pleasure signal (Ikemoto & Panksepp, 1999). These appetitive, affective readings of dopamine stand alongside an influential alternative from computational neuroscience, in which phasic dopamine encodes a reward-prediction error—a teaching signal reporting the gap between received and expected reward that drives reinforcement learning (Schultz, Dayan, & Montague, 1997). The wanting and effort accounts do not refute this view so much as locate dopamine's role on the motivational rather than the purely predictive side, and how a single signal serves both functions remains an open question.

John Salamone sharpened the point with effort-based choice. In his experiments, rats with reduced accumbens dopamine still prefer and consume the better reward, but they will not pay a high effort cost to get it, switching to an easier, lesser option—evidence that dopamine supports the activational, appetitive side of motivation, the vigor and persistence of work, rather than hedonic evaluation (Salamone & Correa, 2012). Which stimuli come to command appetitive effort is itself learned: Shelly Flagel and colleagues showed that only some rats attribute incentive salience to a reward cue, approaching and working for the cue itself—sign-tracking—and that this attribution depends on dopamine, dissociating the learning of a prediction from the assignment of motivational value to a cue (Flagel et al., 2011). Daniel Castro and Michael Bruchas mapped the fine-grained accumbens-shell microcircuitry in which appetitive and consummatory signals are organized, showing that even a single structure contains anatomically distinct zones for wanting-like and aversive responses (Castro & Bruchas, 2019). The demonstration implements Salamone's choice: as dopamine falls, the cost of effort rises and the animal abandons the high-effort, high-reward option.

Worked Example: Effort-Based Choice

Salamone's effort-based choice can be made quantitative with a simple net-value rule. Suppose an animal chooses between a high option—4 reward pellets behind 16 lever presses—and a low option—2 pellets behind just 2 presses. Let the net value of an option be its reward minus a cost that is the number of presses times a per-press cost k, where k rises as mesolimbic dopamine falls (low dopamine makes every press feel more effortful):

net value = reward − k × presses

With dopamine intact, take k = 0.10. The high option is worth 4 − 0.10 × 16 = 4 − 1.6 = 2.4; the low option is worth 2 − 0.10 × 2 = 2 − 0.2 = 1.8. The animal chooses the high-effort, high-reward option (2.4 > 1.8).

Now deplete dopamine, doubling the per-press cost to k = 0.20. The high option falls to 4 − 0.20 × 16 = 4 − 3.2 = 0.8, while the low option falls only to 2 − 0.20 × 2 = 2 − 0.4 = 1.6. The ranking reverses: the animal now prefers the low-effort option (1.6 > 0.8), even though the reward it likes most has not changed.

The switch happens at the k where the two options are equal: 4 − 16k = 2 − 2k, giving 2 = 14k, so k = 1/7 ≈ 0.143. Below that cost the high option wins; above it, the low option does. This is the behavioral signature Salamone reports—dopamine depletion reallocates behavior toward low-cost rewards without blunting the preference for the better reward itself (Salamone & Correa, 2012). The EffortChoiceDemo above is built on this rule: its dopamine slider sets k, and the option it selects is whichever has the higher net value.

Discussion

A century of work has converged on a single organizing idea: the pursuit of a goal and the consumption of it are different processes, with different dynamics and different brain substrates, and appetitive behavior is the name for the first of them. Craig's functional distinction, Tinbergen's hierarchy, Berridge and Robinson's wanting–liking dissociation, and Salamone's effort-based account are not competing theories so much as successive refinements of the same insight, each pushing it to a finer mechanistic level (Berridge, 2004). What began as a description of how an instinct unfolds has become a claim about how the brain allocates motivation.

The dissociation matters because it predicts failures that a one-process view of reward cannot. If wanting and liking are separable, then a reward can be wanted without being liked—the core abnormality in the incentive-sensitization account of addiction, in which drug cues acquire pathological incentive salience and command appetitive behavior long after the drug has ceased to be pleasurable (Berridge & Robinson, 2016). The same logic illuminates disorders of effort and apathy, where an intact capacity for pleasure coexists with an unwillingness to work for it, and disorders of eating, where the wanting of food outruns its liking (Morales & Berridge, 2020). Reading appetite as a computation over value and cost, rather than a simple drive, is what makes these patterns intelligible.

Current Directions

The most active questions concern how finely the appetitive machinery can be resolved. Optogenetic and chemogenetic tools have turned the accumbens shell from a single reward structure into a mosaic: Castro and Bruchas showed that neighboring microzones, defined by their neuropeptide and circuit identity, generate opposite appetitive and aversive motivations, so that where in the shell a signal arrives can flip behavior from approach to avoidance (Castro & Bruchas, 2019). Mapping this geography, and understanding how a normal brain coordinates it, is a central current problem.

A second front extends wanting beyond the striatum. Shelley Warlow and Berridge showed that circuitry in the central amygdala can amplify the incentive salience of a specific reward cue, narrowing an animal's wanting onto one target with near-addictive intensity (Warlow & Berridge, 2021). This locates part of the appetitive engine in a structure once assigned chiefly to fear, and it raises the question of how distributed the wanting system really is. Running through both fronts is a conceptual reckoning with the limits of the appetitive–consummatory dichotomy itself—an organizing distinction that remains indispensable even as the data increasingly show the two phases to be the ends of a graded, interwoven continuum rather than a clean switch (Berridge, 2018).

Common Misconceptions

Appetitive behavior is just feeding behavior.

The word appetitive names the searching, approaching phase of any motivated act, not appetite for food specifically. The same appetitive organization applies to sexual, defensive, parental, and exploratory behavior; feeding is only the most familiar example (Craig, 1918).

Dopamine is the brain's pleasure chemical.

Dopamine implements wanting, not liking. Raising or lowering mesolimbic dopamine changes how hard an animal will work for a reward and how strongly cues attract it, while leaving the hedonic reaction to the reward itself largely unchanged; pleasure depends on separate opioid-sensitive hotspots (Berridge & Robinson, 2016; Berridge & Kringelbach, 2015).

Wanting a reward means expecting to enjoy it.

Wanting and liking are separable, so a reward can be powerfully wanted without being liked. This is the core abnormality in the incentive-sensitization account of addiction, where cues command appetitive behavior long after the reward has stopped being pleasurable (Berridge & Robinson, 1998).

The appetitive and consummatory phases are a clean switch.

The distinction is indispensable but not sharp. Studies of sexual and feeding behavior show the two phases grade into each other and share overlapping circuitry, so they are better treated as ends of a continuum than as a binary (Ball & Balthazart, 2008).

Glossary

Appeted stimulus.

In Craig's terminology, the particular stimulus whose reception ends an appetite and releases the consummatory act; the goal object of appetitive behavior.

Appetitive behavior.

The variable, flexible searching and approaching phase of a motivated act, through which an animal reaches the situation in which a consummatory act can occur.

Aversion.

The mirror image of an appetite: a state of disturbance that persists until a disliked stimulus is escaped or removed, driving behavior away from it.

Consummatory act.

The relatively stereotyped, species-typical terminal behavior—eating, copulation, the kill—that ends an appetitive sequence and quiets the drive.

Effort-based choice.

A decision between a larger reward requiring more work and a smaller reward requiring less; sensitive to mesolimbic dopamine, which supports the willingness to pay effort.

Fixed action pattern.

A stereotyped, largely innate motor sequence that, once triggered by a releasing stimulus, runs to completion; the ethological form of a consummatory act.

Hedonic hotspot.

A small, opioid- and endocannabinoid-sensitive site in the nucleus accumbens shell or ventral pallidum where the liking reaction to a reward is generated.

Incentive salience.

The motivational property the brain attributes to a reward and its cues, making them attractive and sought; the mechanism of wanting and the engine of appetitive behavior.

Instinct.

An organized, largely inherited behavioral program comprising an appetitive search phase and a consummatory act, as formalized in Tinbergen's hierarchical model.

Intracranial self-stimulation.

The behavior, discovered by Olds and Milner, in which an animal works to deliver electrical stimulation to reward-related brain sites, demonstrating circuits that drive appetitive approach.

Liking.

The hedonic impact of a reward—the pleasure of consuming it—belonging to the consummatory phase and dissociable from wanting.

Mesolimbic dopamine system.

The dopamine projection from the ventral tegmental area to the nucleus accumbens that energizes wanting and effortful appetitive behavior.

Nucleus accumbens.

A ventral striatal structure central to reward and motivation, containing both hedonic hotspots for liking and microcircuits that organize appetitive and aversive drive.

Reward-prediction error.

The difference between received and expected reward; an influential computational account holds that phasic mesolimbic dopamine encodes this teaching signal to drive reinforcement learning.

Reward.

A stimulus or outcome that an animal will work to obtain; in modern accounts a compound of at least three dissociable processes—learning, liking, and wanting.

SEEKING system.

Panksepp's name for the ancestral, dopamine-energized circuit that drives exploration, foraging, and the eager anticipatory pursuit underlying appetitive behavior.

Sign-tracking.

Approaching and working for a reward-predicting cue itself rather than the reward; an index of incentive salience attributed to the cue, dependent on dopamine.

Wanting.

The appetitive motivation to obtain a reward, implemented as incentive salience and dissociable from the liking of the reward once obtained.

Key Researchers

Kent C. Berridge

(University of Michigan). American affective neuroscientist whose dissociation of wanting (incentive salience) from liking (hedonic impact) reshaped the understanding of appetitive behavior and reward, showing that mesolimbic dopamine drives the appetitive pursuit phase rather than pleasure itself. ORCID

Wallace Craig

(1876–1954). American zoologist and comparative psychologist whose 1918 paper introduced the appetitive–consummatory distinction, defining appetitive behavior as the variable searching phase that precedes the stereotyped consummatory act. Wikipedia

Barry J. Everitt

(University of Cambridge). British behavioral neuroscientist whose analysis of appetitive versus copulatory responses in male rats dissected the brain circuitry separating the pursuit of a reward from its consummation. ORCID

Morten L. Kringelbach

(University of Oxford and Aarhus University). Danish-British neuroscientist whose work on pleasure systems and the brain's hedonic hotspots maps the consummatory, liking phase that completes an appetitive sequence. ORCID

Jaak Panksepp

(1943–2017). Estonian-American affective neuroscientist who named the SEEKING system, the ancestral appetitive-motivational circuit energizing exploration and reward pursuit across mammals. Wikipedia

Terry E. Robinson

(University of Michigan). American behavioral neuroscientist who, with Berridge, developed the incentive-sensitization theory and the sign-tracking paradigm, showing how appetitive cues can acquire exaggerated motivational force in addiction. ORCID

John D. Salamone

(University of Connecticut). American behavioral neuroscientist whose effort-based choice work recast mesolimbic dopamine as a regulator of the exertion and persistence of appetitive behavior rather than a pleasure signal. ORCID

Nikolaas Tinbergen

(1907–1988). Dutch-British ethologist and Nobel laureate whose hierarchical model of instinct placed appetitive behavior within a rigorous framework, formalizing its descent from a general drive toward a fixed consummatory act. Wikipedia

Frequently Asked Questions

What is appetitive behavior?

It is the variable, flexible searching and approaching phase of a motivated act—the seeking, pursuing, and working that brings an animal into the situation where a fixed consummatory act such as eating or copulation can occur. Wallace Craig introduced the term in 1918 to name the agitation that continues until a sought stimulus is reached (Craig, 1918).

How does appetitive behavior differ from consummatory behavior?

Appetitive behavior is variable, sensitive to circumstance, and shaped by learning, whereas the consummatory act is relatively stereotyped and species-typical. The appetitive phase carries the motivational and cognitive decisions—where to go, how hard to work—while the consummatory act is the fixed movement that ends the sequence (Tinbergen, 1963).

What is the difference between wanting and liking?

Wanting is the appetitive motivation to obtain a reward, implemented as incentive salience; liking is the hedonic pleasure of consuming it. They are separable in the brain, so a reward can be strongly wanted without being much liked (Berridge & Robinson, 1998).

Does dopamine cause pleasure?

No. Mesolimbic dopamine implements wanting rather than liking: changing dopamine alters how hard an animal works for a reward and how strongly cues attract it, while the hedonic reaction to the reward itself is generated by separate opioid-sensitive hotspots and is largely unchanged (Berridge & Robinson, 2016; Berridge & Kringelbach, 2015).

What is the SEEKING system?

It is Jaak Panksepp's name for an ancestral, dopamine-energized brain network that drives exploration, foraging, and the eager anticipatory pursuit of rewards—in effect the neural substrate of appetitive behavior across mammals (Alcaro & Panksepp, 2011).

How is appetitive behavior studied in the laboratory?

Through tasks that separate the work of obtaining a reward from its consumption, such as lever-pressing, effort-based choice, and sign-tracking, often combined with manipulations of dopamine. Olds and Milner's intracranial self-stimulation was an early demonstration that animals will work for activation of appetitive circuits (Olds & Milner, 1954).

What does appetitive behavior have to do with addiction?

The incentive-sensitization theory holds that drug cues acquire pathological incentive salience, so they command intense appetitive behavior—craving and drug-seeking—even after the drug has stopped being pleasurable, a direct consequence of wanting and liking being separable (Berridge & Robinson, 2016).

Are the appetitive and consummatory phases completely distinct?

Not completely. The distinction is indispensable for organizing motivated behavior, but studies of sexual and feeding behavior show the phases grade into one another and share overlapping circuitry, so they are best treated as ends of a continuum rather than a binary switch (Ball & Balthazart, 2008).

References