RESEARCH DOSSIER
Processing Fluency and Aesthetic Pleasure
Abstract. Processing fluency describes the experienced ease or difficulty of mental processing. Research in empirical aesthetics proposes that fluently perceived or understood objects often elicit more positive affect and greater aesthetic liking. This monograph reconstructs the historical, conceptual, experimental, and methodological basis of that proposal, examining mere exposure, priming, visual clarity, repetition, symmetry, prototypicality, semantic understanding, temporal change in comprehension, and affective attribution. It compares hedonic-marking theory with relative-fluency, fluency-amplification, pleasure–interest, appraisal, and predictive-processing approaches, integrating laboratory studies, meta-analyses, registered reports, and current findings through 2026. Its central synthesis treats aesthetic response as an interaction among processing dynamics, perceived change, observer expectations, evaluative goals, stimulus meaning, and the form of judgment requested. Processing fluency is a demonstrable contributor to particular liking judgments; its importance and direction depend on context, and its experiential, behavioral, and physiological indices require separate measurement. This work is a critical narrative scientific monograph based on existing published research, not a newly conducted systematic review or experiment.
Keywords: processing fluency; aesthetic pleasure; perceptual fluency; conceptual fluency; hedonic marking; mere exposure; aesthetic liking; prototypicality; symmetry; predictive processing; pleasure–interest model; appraisal; metacognition; fluency amplification; experimental aesthetics; art expertise.
Scientific Problem and Scope
A perceptual encounter unfolds through operations whose difficulty varies from moment to moment. A familiar shape emerges immediately from its background, a word appears recognizable before its meaning is explicitly recalled, and a seemingly chaotic image becomes orderly when its latent organization is discovered. The psychological question is whether the efficiency of these operations participates in what the observer experiences as aesthetically pleasurable. In the influential synthesis by Reber et al. (2004), processing fluency becomes a candidate explanatory link between apparently heterogeneous sources of liking: repetition, contrast, symmetry, and categorical typicality. The proposition is substantive because manipulations designed to facilitate processing can be separated experimentally from the objective qualities of the stimulus.
The present object of inquiry is the contribution of ease of perceptual and conceptual processing to affective and aesthetic evaluation. It includes the dynamics of processing itself, the observer's monitoring or interpretation of those dynamics, and the judgmental context in which a feeling becomes an evaluation of the object. It also concerns the trajectory of appreciation over time: confusion may resolve into pleasure, routine familiarity may become boring, and a surprising success in understanding may become intrinsically rewarding. These dimensions help explain why a single formula connecting faster recognition to beauty has limited scope. The more comprehensive question is how specific cognitive transitions acquire evaluative significance.
This inquiry concerns aesthetic pleasure as one outcome among several, rather than a complete theory of beauty, artistic quality, visual art, or all positive emotions. Aesthetically pleasing stimuli can be ordinary artifacts, geometric figures, words, designs, or artworks. Judgments of beauty, liking, interest, creativity, and artistic significance overlap empirically but require separate operationalization. The broad empirical-aesthetics review by Palmer et al. (2013) makes this distinction central: preferences for colors and spatial structures provide tractable experimental material, while response differences among observers and contexts prohibit treating one rating scale as the entire phenomenon.
A scientific explanation needs to specify a stimulus variable, a measurable processing event, an experience or interpretation, and an evaluation. A manipulation may improve stimulus discriminability without altering pleasure, or improve measured recognition speed while producing boredom. Conversely, a difficult stimulus may become enjoyable when effort is meaningful and insight is attainable. The monograph therefore evaluates the strongest causal evidence, the alternative mechanisms that can explain the same observations, and the situations where the outcome must be revised or narrowed. Related constructs such as general emotional reward, visual composition, artistic expertise, and aesthetic interest appear where they constrain this dossier's central causal question.
Historical Formation of the Research Question
The scientific study of preferences grew from psychophysics, the experimental analysis of relationships between stimulus properties and psychological responses. Its early development established the possibility of measuring aesthetic judgment through structured comparisons rather than appealing solely to normative philosophy. Modern processing-fluency theory was formed through several intersecting traditions: research on familiarity and repetition, affective judgments made without explicit recognition, recognition and categorization efficiency, Gestalt organization, and metacognitive experiences of knowing or understanding. The historical sequence matters because the evidence originally addressed different outcomes before investigators proposed a common explanatory process.
Robert Zajonc's Zajonc (1968) placed repeated contact with a stimulus at the center of a testable prediction: exposure itself can make attitudes more favorable. In Zajonc (1980), Zajonc elaborated the separation of affective discrimination from detailed cognitive recognition. The historical contribution lies in demonstrating the importance of affective evaluation under conditions that did not require articulate reasons. The precise mechanisms remained disputed. Familiarity, lowered uncertainty, affective conditioning, memory, and increased processing ease can all participate in exposure-related preference.
Bornstein (1989) examined the exposure-and-affect research accumulated during the two decades after Zajonc's early work. It documented methodological moderators, including duration, recognition, stimulus complexity, and temporal spacing. That synthesis gives a crucial warning to any later single-mechanism account: repeated exposure is a family of experimental procedures, and its effects must be interpreted jointly with stimulus and task parameters. The link from repetition to fluency is an explanatory hypothesis, while the link from repetition to liking is an experimentally studied relation.
The modern fluency account acquired sharper causal leverage with Reber et al. (1998). Their three experiments varied matching primes, figure–ground contrast, and presentation duration. More fluently processed stimuli received more favorable aesthetic evaluations under the tested conditions. These manipulations are historically decisive because they facilitate perception without relying on prior exposure alone. Reber et al. (1999) further showed that ease of reading could influence truth judgments, placing fluency within a larger class of judgments that use subjective processing as information.
The Reber et al. (2004) linked the earlier results into processing-fluency theory of aesthetic pleasure. Subsequent research broadened the object domain, distinguished stages of processing, and tested competing interpretations. Leder et al. (2004) described aesthetic appreciation as a multistage process involving classification, cognitive mastery, and evaluation. Graf et al. (2015) later distinguished rapid pleasure from motivated interest. The historical trajectory culminates in research on relative gains, prediction errors, epistemic goals, and the conditions under which ease reinforces or amplifies an existing reaction.
The Construct of Processing Fluency
Processing fluency is the subjective ease associated with performing cognitive operations on a stimulus. It can reflect perceptual processes, such as detecting contours, identifying letters, or segregating figure from ground, and conceptual operations, such as resolving a word's meaning, recognizing a category, or understanding the relation between a painting and its title. It is a metacognitive experience because the observer may use something about the processing operation as information when evaluating the object. Alter et al. (2009) argue that diverse manipulations share this subjective dimension despite operating at different levels of cognition.
A fundamental distinction concerns objective processing efficiency and felt processing ease. Reaction time, error rate, thresholds, and eye movements characterize the performance of a system under a particular task. A fluency judgment reports an experienced quality of the process. Although the two can covary, neither is a general substitute for the other. Reber et al. (2004) found that contrast and font affected different stages of word processing while both influenced subjective fluency. This supports the proposition that fluency experience can summarize several operations, and it also emphasizes the necessity of isolating the stage actually manipulated.
Perceptual fluency concerns early discrimination, recognition, legibility, and efficient organization of the sensory input. Conceptual fluency concerns meanings and categorical relationships rendered readily accessible by knowledge or context. An artist's descriptive title can facilitate one level while leaving lower-level visual operations essentially unchanged. A highly legible image may remain conceptually obscure, while an intricate image can be semantically recognizable to an expert. The conceptual distinction is operationally useful because experimental manipulations can cross physical clarity with semantic congruence.
Fluency can be measured as a state tied to a trial, while the skills and learning history supporting it vary across observers. A trained viewer may decode visual conventions with little explicit effort; a novice may need prolonged exploration. The same pattern may be effortless in one presentation context and difficult amid competing distractors. Research should therefore treat fluency as an interaction between stimulus, observer, and task. A universal property assigned to an object alone cannot capture the relevant mechanism.
Oppenheimer (2008) places fluency beyond overt ratings: processing ease can affect memory access, attentional allocation, and choice of cognitive strategy. This breadth creates an identification problem. If a fluent stimulus is liked more, its effect might pass through positive affect, through perceived familiarity or certainty, or through a changed strategy of evaluation. An adequate model represents these alternatives explicitly and tests whether participants actually experience and use the proposed cue.
Aesthetic Pleasure and Its Dependent Variables
Aesthetic pleasure is the positive affective dimension of engaging with a perceived or imagined object under conditions of aesthetic attention. In an experiment, investigators commonly operationalize it through immediate liking, pleasantness, attractiveness, or beauty ratings. These scales are related but ask different questions. Liking registers a personal response, beauty invokes an aesthetic evaluation that may draw on learned criteria, and artistic value can involve knowledge about innovation, authorship, historical significance, or technical achievement. The existence of positive associations between their scores does not establish their interchangeability.
An experiment can request a binary choice between figures, an ordered ranking, a continuous rating, or a judgment after extended study. As Palmer et al. (2013) document, task formats influence which aspects of preference are observable. A forced-choice trial compares two local alternatives, whereas an isolated rating evokes internally calibrated standards. This methodological distinction becomes especially important when testing relative fluency. Effects that require comparison among fluent and disfluent stimuli may diminish when participants see only one processing condition.
Pleasure also differs from interest. A highly familiar composition may be comforting and immediately pleasant; a puzzling composition may command curiosity while producing little initial pleasure. The latter can reward prolonged attention if the observer believes comprehension is possible. Silvia (2005) identified novelty and perceived coping potential as appraisal determinants of interest. The implication for this dossier is that researchers should sample at least two dependent dimensions when a stimulus invites exploration: immediate liking and interest in further processing.
A further distinction separates a report of felt fluency from an evaluation of the target. Asking directly how easy a stimulus felt can sensitize a participant to processing cues and change their subsequent interpretation. Conversely, measuring only liking can leave the mediator unobserved. A rigorous design therefore considers timing, measurement reactivity, and the validity of each response instrument. A positive liking effect after priming is evidence of evaluative change, but its full causal explanation requires appropriately designed measures and tests.
Finally, an aesthetic encounter can be emotionally mixed. Novelty, uncertainty, pleasure, tension, and release may coexist or occur in sequence. Silvia (2005) shows why reducing the emotional field to a single pleasantness factor misses theoretically important distinctions. The scientific question is which outcomes fluency predicts at which moments and by what mediating interpretation.
Causal Architecture of Hedonic Fluency
The classic hedonic-marking hypothesis proposes a chain from facilitated processing to positive affect and from that affect to a more favorable evaluation of the stimulus. In simple causal notation, experimental facilitation F changes the processing state P; P changes the experienced valence A; and A contributes to a judgment J. The experimental aim is to distinguish this route from direct effects of stimulus properties, prior associations, semantic meaning, and demand. Reber et al. (2004) use the shared behavioral consequences of disparate fluency manipulations as evidence for an integrative mechanism.
The model has an attribution component. If a perceiver experiences ease without knowing its experimental cause, the positive aspect of the experience can be attached to the object. If the perceiver recognizes that the stimulus is readable merely because its contrast was artificially adjusted, the experience may be interpreted as a technical property of the display rather than a source of beauty. Attribution is therefore itself a psychological process involving awareness of sources, relevance of the current question, and implicit judgment rules. It makes the prediction conditional rather than unconditional.
The proposed causal chain can be evaluated by manipulating fluency while holding other stimulus features as constant as possible. Priming a target with a matching image is more diagnostic than merely comparing naturally symmetric and asymmetric pictures, because symmetry may have affective, semantic, or ecologically relevant consequences independent of recognition speed. The same principle applies to comparisons between familiar and unfamiliar stimuli. Random assignment and independent measures of processing benefit reduce—but do not automatically eliminate—the possibility of alternative routes.
Forster et al. (2015) addressed subjective mediation more directly by manipulating feedback interpreted as a physiological sign of processing ease, alongside presentation duration. Their experiments showed that interpreted ease and compatibility of the feedback manipulation could influence liking. This strengthens the claim that an experienced or inferred fluency cue is causally relevant, while also demonstrating that interpretations attached to the cue are part of the mechanism.
An alternative model treats fluency as an amplifier of evaluations already present. A positively evaluated stimulus could become more positive when easier to process; a negatively evaluated one could become more negative. This mechanism predicts a different interaction between initial valence and facilitation. A later preregistered research program by Landwehr et al. (2020) compared hedonic uplift and bidirectional amplification and found indications that both mechanisms could operate. The causal architecture is therefore best described as plural and testable: processing change can contribute an affective signal, modify the strength of an existing evaluation, or alter how evaluation proceeds.
Repetition and the Mere-Exposure Tradition
Repeated presentation is the most historically prominent manipulation associated with fluency. Repetition often improves perceptual and conceptual availability, increases familiarity, and can modify subsequent preference. It can also induce boredom or habituation at higher exposure levels. The mechanism cannot be inferred from exposure count alone because repetition changes more than one variable: memory strength, expectation, uncertainty, affective association, and the distribution of attention may change together. A clean experimental test distinguishes these possibilities whenever possible.
Zajonc (1968) established the research program by relating repeated exposure to shifts in attitude, including stimulus classes that lacked well-developed semantic content. In the historical debate, the possibility of affective preference without accurate recognition suggested that explicit identification is not required for all forms of liking. A modern processing account can interpret this as improved subthreshold perceptual efficiency; another account can appeal to implicit memory or familiarity. The evidence of an exposure effect by itself adjudicates none of these completely.
Bornstein (1989) synthesized methodological variation across the early literature and found that effect magnitude depends on important design decisions. This matters for aesthetic explanation because a brief unfamiliar geometric figure and a frequently heard melody have different information structures and different baseline representations. A later meta-analysis by Montoya et al. (2017) estimated 268 exposure–response curves from 81 articles and identified a positive linear and negative quadratic pattern. Their findings support a context-sensitive, sometimes inverted-U trajectory rather than unlimited improvement with repetition.
The methodological interpretation should remain precise. The evidence summarized by Montoya and colleagues concerns aggregated exposure–liking relationships under specified task and stimulus conditions. It does not show that every individual exposure curve is inverted-U shaped, nor does it verify that fluency is the sole mediator. The meta-analytic pattern is consistent with initial processing benefits accompanied by eventual satiation or shifting expectations; additional discriminating experiments are necessary to establish the relative contribution of each process.
The exposure tradition supplies two decisive lessons. First, aesthetic preference can change through history of encounter even when an object's physical features remain constant. Second, a theory of fluency and pleasure must explain the temporal boundary at which greater familiarity stops improving liking. Repetition thereby becomes both supporting evidence and a constraint: the theory must account for positive effects, their moderators, and situations in which curiosity or novelty becomes a more important evaluative goal.
Visual Priming, Contrast, and Presentation Duration
The most direct test of fluency-based liking varies how readily a visual target is processed while attempting to preserve the target's identity and evaluative content. Reber et al. (1998) made this approach central to empirical aesthetics by investigating matching visual primes, figure–ground contrast, and presentation duration. Across three experiments, stimuli prepared for easier processing were judged more attractive under the tested conditions. The inferential significance lies in the convergence of several manipulations that influence access to perceptual information through different immediate routes. If the effect were tied only to repeated exposure, contrast and duration should not produce parallel results.
Matching primes facilitate detection or interpretation of a subsequently presented target. A design should separate identity or shape priming from general alerting and from changes in semantic association. A matching prime can make a target more expected, not merely easier to perceive, while an incongruent prime may simultaneously produce prediction error. Masked presentation helps limit conscious recognition but does not guarantee the absence of awareness in every participant or trial. Forced-choice awareness tests and sensitivity analyses are consequently important. Under controlled exposure, the fluency account predicts that a facilitation manipulation can shift liking even when objective target quality is held fixed.
Contrast is another useful manipulation because it changes the quality of available sensory evidence. High figure–ground contrast typically improves discrimination or identification of the target, yet perceived attractiveness may depend on brightness preference, visual comfort, and the visual composition itself. Researchers must accordingly counterbalance contrast assignments, use identical target structures across conditions, and measure legibility or accuracy separately. A physical manipulation that eases perception can also change the aesthetic properties of the image, making convergent designs especially valuable.
Presentation duration influences how much perceptual information can be accumulated before judgment. Longer inspection may enable more complete processing, although prolonged exposure can also reveal defects, enable semantic reinterpretation, or diminish curiosity. The critical comparison is between facilitation at the same level of available information and additional time that changes the processing stage. Reber et al. (2004) showed that contrast and font-related changes can operate at distinguishable stages even when subjective ease reports show similar tendencies. This helps explain why equal subjective fluency ratings do not identify an identical causal computation.
A recent experimental continuation directly addresses stage and duration. Jiang et al. (2026) used masked priming to dissociate perceptual and conceptual fluency in aesthetic liking. Their evidence indicates that perceptual facilitation of liking weakened as target duration increased, whereas the conceptual contribution was more persistent across the tested duration contrast. This result supports level-specific hypotheses about the timing of aesthetic evaluation rather than a timeless, undifferentiated processing-ease variable. The experiments should be interpreted within their stimulus, masking, and exposure conditions; they do not establish an invariant effect for every object and population.
Prototypicality and Category Learning
Prototypicality refers to how well a stimulus represents the central tendencies or characteristic structure of a learned category. A prototype can be recognized efficiently because its features are congruent with many learned category instances. Category typicality thus supplies a potential bridge between cognitive organization and aesthetic response. Prototype preference is especially revealing when categories are novel and artificially learned, because investigators can separate the effects of familiarity with individual examples from a generalization process that favors a central representation.
Winkielman et al. (2006) investigated the appeal of prototypical stimuli across experimental materials and found support for a link between classification ease and positive evaluation. Their work employed reaction-time and affect-related indicators as part of a broader attempt to identify why category-central examples can seem more attractive. Because learning a category changes representation at the group level, an observer may prefer an unseen prototype over a previously encountered unusual exemplar. The resulting preference is more informative than a simple association between repetition count and liking.
Category learning also embeds psychological expectations. An object can feel right because it matches a familiar category, whereas deviation can signal creativity, rarity, or risk. In ordinary environments, the category norm can change with culture, expertise, and historical fashion. An architect's prototypicality judgments about a façade or a musician's judgments about a harmonic progression are shaped by acquired knowledge, and what is fluent for one community may be unfamiliar to another. Accordingly, prototype-based aesthetic pleasure must be understood as observer-relative, including the social conditions under which a category is learned.
A standard causal design teaches participants controlled exemplars, tests their recognition or classification of new instances, and then elicits attraction or beauty ratings. Ideally, individual exposure histories and perceptual distances from the category center are independently balanced. Statistical analyses can model whether classification speed mediates the relation between prototypicality and liking, while testing whether positive affect also follows from simple familiarity or expectancy confirmation. A mediation estimate is informative only insofar as the design supports the necessary assumptions about unmeasured confounding.
Prototypicality further raises a question about aesthetic originality. A stimulus can be easy to classify and pleasant yet receive a low rating of artistic innovation; a striking departure from a category can attract attention precisely because it violates conventions. A fluency theory of liking therefore should not be interpreted as a theory that predicts the merit of all creative products. Palmer et al. (2013) show that visual preference requires attention to the interplay between stimulus structure and observers' experience, rather than treating universal stimulus properties as sufficient explanations.
Symmetry, Regularity, and Visual Form
Symmetry and perceptual regularity provide intuitively powerful examples of fluency because they reduce the complexity of visual description and can facilitate organization. Reflectional symmetry, periodic patterns, aligned elements, and coherent grouping permit the perceiver to compress or predict relationships among features. Their subjective appeal has long occupied experimental aesthetics, but the explanatory inference must be cautious: symmetry also carries informational, biological, social, and design-related significance that can influence evaluation independently of processing effort.
A regular pattern may be liked because its structure can be rapidly grasped. The same pattern may also signal stability, workmanship, or environmental predictability. Conversely, asymmetric compositions can become engaging through balance, directionality, or controlled surprise. Preferences vary with stimulus class, task, cultural convention, and the degree of expertise. The scientific task is to estimate how much of a regularity preference is due to a change in perceptual fluency rather than to alternative perceptual or semantic properties. Reber et al. (2004) included perceptual regularity among the central demonstrations motivating their broader account.
Methods range from parametric modification of symmetry and contour to controlled exposures and priming of unchanged targets. Orthogonal manipulations are particularly informative: if identical asymmetric patterns become more fluent through task training, one can test whether liking changes while the image remains physically asymmetric. Likewise, comparing more and less symmetrical patterns at equivalent measured processing efficiency helps separate preference for regularity as content from preference for ease of recognition as experience.
Contour preference illustrates the complexity of this problem. Bar et al. (2006) reported preferences for curved relative to sharply angled visual objects. A fluency-based explanation is possible in some contexts, but perceived threat, learned associations, and intrinsic features of contours are plausible competitors. An elegant pattern of results across different shape categories is therefore the beginning of mechanistic analysis rather than proof of a single process.
For complex visual art, perceptual organization can unfold over time. A painting may initially resist easy grouping, and exploratory attention may reveal new relations among elements. The evaluation can improve during this discovery even though the object's physical symmetry or complexity has not changed. This observation connects regularity research to insight, predictive reduction, and the dynamics of understanding. A framework sensitive only to absolute simplicity would overlook the pleasure associated with finally discovering order in a difficult composition.
Conceptual Fluency and Semantic Meaning
Conceptual fluency arises when an object's identity, category, narrative, symbolism, or semantic relationships become easier to access. A familiar icon can have clear meaning despite poor sensory presentation; an unfamiliar artwork can be perceptually crisp yet difficult to interpret. The distinction allows experiments to separate seeing from understanding. In aesthetic settings, a title, label, cultural context, or prior explanation can change the meaning of an unchanged image, creating an opportunity to test whether conceptual comprehension alters pleasure, curiosity, and value judgments.
Leder et al. (2006) investigated the effects of titles on understanding and appreciation of paintings. Their results underscore that increases in comprehension and changes in preference should be measured independently. Descriptive or elaborative titles can help participants interpret a work without guaranteeing that their liking changes in the same way. This is an important constraint on claims that any successful cognitive operation automatically produces positive aesthetic value.
Belke et al. (2010) examined high-level processing dynamics in art appreciation, extending the fluency discussion to relationships between explanatory context, interpretative activity, and liking. Their work is especially relevant to more representational and meaning-rich material, where conceptual recognition may be easier to achieve than for unfamiliar abstraction. It suggests that the same intervention can have different evaluative consequences according to the representational and interpretive affordances of the artwork.
The temporal order of comprehension and pleasure is a key research variable. A title provided before the image can prime interpretation, while one delivered after an initial judgment can create a measurable experience of revision. An experiment that collects liking at both moments can assess the trajectory of change and identify whether appreciation rises when uncertainty resolves. Muth et al. (2013) studied the aesthetic Aha experience produced by perceptual insight, linking the emergence of understanding with a more favorable evaluation of previously ambiguous material.
Semantic congruity can also produce cultural variation. Knowledge of mythology, historical events, genres, or symbols may make particular images easy to understand for one observer and conceptually opaque to another. A title that clarifies meaning in one language may introduce ambiguity when translated. Cross-cultural experiments should therefore validate what participants understood rather than infer conceptual fluency from the presence of a label alone. The scientific contribution of semantic-fluency research lies in revealing that aesthetic evaluation changes with knowledge and interpretation, not simply optical form.
Relative Fluency and Processing Dynamics
The hedonic significance of ease may depend on a comparison standard. An observer can process two clear stimuli with comparable objective efficiency yet experience the first as surprisingly easy after a run of difficult trials. An experiment in which fluent and disfluent targets are intermixed differs psychologically from a between-participant design in which every target is easy. This distinction gives rise to relative-fluency accounts: evaluative consequences depend partly on the contrast between current processing and an expected or locally available baseline.
Forster et al. (2015) compared processing-fluency manipulations under within-participant and between-participant arrangements. Their findings distinguished subjective experiences from evaluations: clarity-related ease could be detected in different arrangements, but a liking effect was particularly evident where fluent and disfluent stimuli were available for comparison within the same observer's experience. This is a powerful warning that a subjective mediator need not carry the same evaluative weight across task contexts.
Relative change also connects fluency research to perceptual learning and insight. When an initially difficult image suddenly becomes intelligible, the magnitude of improvement can contribute to pleasure even though the final state remains objectively more demanding than an easy image encountered immediately. Muth et al. (2013) provide evidence relevant to this change-based mechanism. The rewarding event may be the reduction of uncertainty experienced across successive states rather than the final processing speed considered in isolation.
A useful conceptual representation is that evaluated fluency depends on current processing efficiency, anticipated efficiency, and change from a recent or learned baseline. These influences can be crossed experimentally: hold final legibility constant while varying prior difficulty, or hold prior difficulty constant while varying resolution. A preregistered interaction analysis could then test whether liking tracks absolute ease, the gain in ease, or the surprise produced by an unexpectedly fluent outcome. Such a design discriminates mechanisms that make identical predictions in a simple easy-versus-hard comparison.
Forster (2020) emphasizes the role of relative processing change in aesthetic engagement. Its interpretation helps reconcile the pleasure of mastery with laboratory liking effects produced by simple clarity. The two cases may share an evaluative response to improved accessibility while differing in temporal complexity and observer goals. This possibility is analytically promising but should be tested with measurements of momentary expectation and temporal processing, rather than assumed from similarity of phenomenological descriptions.
Competing Explanations of the Fluency–Liking Relation
The simplest hedonic-marking theory proposes that fluency contributes a positive experiential signal. Yet the same positive association between ease and liking could emerge through other paths. Fluent objects may seem familiar, safe, true, predictable, typical, or easy to justify. Each attribute may influence evaluation in its own right. A sophisticated experiment therefore tests specific mechanisms instead of treating all benefits of familiarity and recognition as evidence for positive affect induced by processing ease.
The fluency-amplification account proposes that ease can increase the intensity of an existing response regardless of its initial valence. If a stimulus is already evaluated positively, greater ease can make the positive assessment stronger; if it carries negative meaning, greater ease could support stronger negative evaluation. Landwehr et al. (2020) explicitly tested amplification against hedonic marking with exploratory and confirmatory studies, finding evidence for contributions from both. The distinction matters because the effects of facilitating recognition will depend on stimulus valence, not merely task difficulty.
Affective misattribution provides another explanatory route. A perceiver may encounter a pleasant processing experience, experience familiarity, or receive a cue of successful prediction, then interpret that cue as a property of the target. This interpretation depends on context and awareness: processing ease may be discounted when its source is obvious or when the current evaluation task draws attention to different criteria. Alter et al. (2009) locate the role of fluency in metacognitive judgment across several domains, showing why aesthetic application must specify the evaluation rule through which the signal enters a beauty or liking response.
Prediction-based accounts foreground expected sensory or conceptual structure. A stimulus that closely matches a prediction can be processed easily, yet it may be boring. An initially surprising stimulus can be interesting, and the resolution of that surprise can generate reward. Van de Cruys et al. (2011) proposed a link between predictive error dynamics and reward in art. Their account invites empirical studies of dynamic error reduction, but it remains important to distinguish computationally plausible explanatory models from experimentally established causal pathways.
The models need not be exclusive. A stimulus may become easier to recognize, generate a weak positive feeling, evoke greater certainty, and simultaneously amplify an existing evaluative association. What matters is whether a research design can independently vary those pathways and assess their contribution. Predictions diverge when a fluent stimulus has unpleasant semantic content, when difficulty is valued as an intellectual challenge, when the cause of ease is disclosed, or when subjective certainty and objective processing efficiency dissociate. The most informative future experiments target these cases rather than repeatedly demonstrating that clear, familiar images are often preferred.
Pleasure, Interest, and Cognitive Mastery
A central development in the psychology of aesthetic response distinguishes the immediate comfort of fluent recognition from the interest generated by productive difficulty. Graf et al. (2015) proposes that different processing modes and motivations yield different consequences for preference. Under conditions of effortless early processing, fluency can generate positive evaluations; when the observer undertakes more controlled interpretation, the prospect of overcoming difficulty can make complexity desirable. This distinction gives the theory explanatory reach across everyday design and challenging art.
Interest is a motivated emotion involving a willingness to allocate resources to an object. Silvia (2005) developed an appraisal account in which novelty and complexity acquire the potential to elicit interest when an observer judges the stimulus comprehensible or manageable. An unfamiliar painting may be compelling to someone with relevant expertise because its ambiguity promises insight. The same ambiguity may produce disengagement for a viewer who sees no viable route to understanding. What changes is the relation between perceived task demand and the observer's coping potential.
The pleasure–interest distinction requires time-sensitive measurement. Immediate liking collected after a short glance may principally reflect easily accessible visual features, whereas an evaluation after several minutes permits interpretation, memory retrieval, and exploration. The same object may shift from low liking and high curiosity to greater liking after a meaningful discovery. Muth et al. (2013) show why an aesthetic insight event merits attention as a process that can alter appreciation during an encounter.
A research design should measure initial reaction, interest, exploratory effort, subsequent comprehension, and final evaluation independently. Some participants may enjoy challenge despite receiving no immediate solution; others may prefer rapidly intelligible material. A model that simply aggregates their final beauty scores can obscure distinct motivational pathways. Even when fluency contributes to enjoyment in both groups, its role differs: one group values ease already available, while the other values the transition toward achievable understanding.
The conceptual consequence is that processing difficulty and pleasure can be positively associated under identifiable conditions. Difficulty becomes aesthetically productive when the goal is mastery, when progress is perceptible, and when the observer chooses to invest effort. Such situations set limits on any universal principle that lower processing cost must always correspond to higher aesthetic value. They also give researchers a concrete agenda for experimentally manipulating observer goals alongside stimulus complexity.
Predictive Processing and the Reward of Resolution
Predictive-processing approaches treat perception and cognition as processes shaped by learned expectations about incoming information. In this framework, recognition is facilitated when sensory evidence conforms to a useful internal model, while mismatches require updating or further inference. Aesthetic pleasure may track not only the current fit between prediction and input but also the progress made in reducing error. Van de Cruys et al. (2011) articulated a theoretical account connecting art engagement, prediction errors, and reward.
The explanatory advantage of a dynamic account is that it can accommodate both an immediately recognizable object and an initially obscure object whose organization becomes apparent. In the first case the prediction is already suitable; in the second case a changed interpretation improves fit over time. Both encounters may generate appreciation, though the second involves surprise, exploration, and a subjectively rewarding discovery. Their psychological trajectories differ in ways that static fluency ratings may fail to capture.
A predictive explanation also clarifies why perfectly expected stimuli can eventually become less attractive. As further repetitions add little information, an observer's motivation may shift toward obtaining novelty or reducing uncertainty elsewhere. The aggregate exposure literature, including Montoya et al. (2017), is consistent with the possibility that familiarity has diminishing or reversing evaluative returns under some conditions. Those findings should not be interpreted as direct confirmation of a particular neural prediction-error model, because other habituation and attentional mechanisms remain plausible.
Experimental tests could present a sequence whose structure participants learn and then introduce an event that initially violates but eventually clarifies the sequence. Separately measuring surprise, processing time, perceptual resolution, and pleasure would help isolate whether appreciation follows uncertainty itself, successful correction, or a broader shift in attentional motivation. Computational models are useful when their latent predictions are specified before observing the preference data and compared against alternative predictors.
A distinctive challenge is the possibility that inferred ease, expectation fulfillment, and reward are partly circularly defined if all are measured by a single liking score. Independent behavioral indices and, where justified, physiological measures are needed. The predictive-processing perspective is a productive theoretical framework for designing such tests; its specific reward claims require direct empirical discrimination.
Expectation, Surprise, and Epistemic Motivation
Recent fluency research integrates what observers expect a processing event to feel like and what they want to accomplish while processing. Yoo et al. (2024) examine how dynamics, expectations, and epistemic goals shape aesthetic evaluation. The work develops the important point that identical fluency can have different subjective meanings when it is surprising, attributed to a different source, or experienced in pursuit of different goals.
An easy processing episode is not necessarily rewarding merely because it consumes fewer resources. An observer who expects a work to be straightforward may treat its ease as unremarkable; an unexpectedly accessible interpretation may feel satisfying; and a task that asks observers to search for a hidden pattern may make initial difficulty desirable. Ease, surprise, and task goals should therefore be represented as distinct experimental variables. Mixing them can produce apparently contradictory findings when different laboratories use different instructions or trial structures.
Epistemic goals concern acquiring knowledge, resolving uncertainty, and determining whether a representation is accurate or complete. They can alter the evaluation of an unfamiliar stimulus. In an information-seeking task, a complex artwork may offer an appealing opportunity for discovery; in a quick choice task, the same ambiguity may be costly or unpleasant. This effect can arise through changes in attention, persistence, interpretation, and affective appraisal. Aesthetic pleasure depends partly on the goal the observer brings to the encounter.
The 2024 framework also suggests interactions between expected and actual fluency. When high fluency violates an expectation of difficulty, it may register as a meaningful achievement; when low fluency violates an expectation of ease, it may create frustration or prompt renewed effort. The precise effect is conditional on whether the observer values accurate comprehension, speed, novelty, social approval, or another outcome. Such predictions are scientifically valuable because they require explicit manipulation of expectations and goals.
Winkielman et al. (2025) extended the synthetic discussion of fluency, feelings, interpretations, expectations, and goals. Together these contributions suggest that aesthetic evaluation is generated through interaction between an evolving processing state and an interpretive context. The historical principle that easy processing often favors liking remains informative, but the relevant explanatory unit is increasingly the observer's situated experience of processing change.
Individual Differences and Cultural Context
Observers differ in visual expertise, category knowledge, perceptual ability, need for cognition, ambiguity tolerance, familiarity with a cultural tradition, and interest in particular forms of art. These differences affect how quickly a stimulus can be understood and whether effort is experienced as a burden or opportunity. The fluency contribution to aesthetic pleasure therefore has both within-person and between-person components. A stimulus that affords easy pattern recognition to an expert may remain opaque to a novice.
Expertise can shift the balance between immediate pleasure and active exploration. Experienced viewers may have efficient recognition routines and richer interpretive repertoires, enabling engagement with complex forms that novices find overwhelming. Leder et al. (2012) compared expertise-related dimensions of appreciating art and highlighted the importance of knowledge and appraisal. Such evidence constrains any claim that visual simplicity alone establishes an invariant preference.
Cultural learning contributes conventions of beauty, category centrality, perceptual expectation, and aesthetic meaning. These conventions affect which stimuli appear typical and which associations are made accessible during evaluation. Zhang et al. (2022) investigated aesthetic preference and conceptual metaphors with attention to cultural context, including participant comparisons involving the United States and Iran. The research underscores that apparently formal visual preferences can reflect learned symbolic mappings and differences in conceptual accessibility.
Differences in the context of encounter matter as much as differences among people. A symmetrical pattern displayed as architecture, scientific information, a decoration, or an artistic experiment can invite distinct standards of judgment. The same observer may prefer one presentation for clarity and a different one for novelty or expressiveness. Context may change how fluency is attributed: ease can signal competence, predictability, conventionality, or tediousness. Experimental tests should vary framing independently of the stimulus while identifying whether the evaluation task is beauty, liking, originality, or usefulness.
Cross-cultural research must avoid equating national labels with psychological mechanisms. Sampling should document language proficiency, arts exposure, education, cultural identification, and prior experience with the stimuli. Translation, screen calibration, and response-style differences also deserve scrutiny. Where a fluency manipulation replicates, the result supports generalization across the sampled conditions; where it differs, the discrepancy can motivate tests of learned categories, observer goals, or different interpretation rules. Neither outcome should be reduced to an unexamined claim about a whole culture.
Experimental Measurement and Identification
Scientific explanation begins by operationally separating stimulus manipulation, processing state, affective experience, and explicit aesthetic judgment. Researchers can change contrast, prime congruence, category structure, repetition, presentation time, or semantic labeling. They then measure recognition latency, accuracy, perceived ease, facial electromyography, pupil size, viewing time, confidence, and aesthetic ratings where appropriate. These indicators serve different functions and have different inferential limits. A reaction-time advantage indicates behavioral facilitation; it does not by itself show a positive experience or explain a beauty judgment.
An ideal factorial design manipulates fluency and candidate moderators independently. For example, targets of controlled baseline valence can be presented in high- versus low-facilitation conditions while epistemic goals or task framing are randomized. Such a design can test hedonic marking against fluency amplification by examining whether the effect of facilitation differs for initially favorable and unfavorable targets. Randomization improves causal inference about the manipulated factor, though it does not itself make an unmeasured mediator causally identified.
The distinction between within-participant and between-participant conditions is essential. Forster et al. (2015) demonstrated that the arrangement of fluency contrasts can matter for liking. Mixed models should account for both participant and stimulus heterogeneity, with random effects suited to the sampling design. When researchers use thousands of trials but only a small set of participants or stimuli, the number of observations should not be mistaken for independent replication across people or object categories.
The temporal structure of measurement can change the studied process. A prompt to rate fluency before liking may call attention to ease and make participants use it explicitly, while asking only for beauty may permit an implicit contribution. Repeated measures of pleasantness may also interrupt the aesthetic experience. Researchers can use counterbalanced task order, distinct participant groups, or unobtrusive timing measures, explaining the compromises each introduces. A claim about automatic affect requires evidence beyond a participant's post-task narrative about their preferences.
Measurement validity and statistical uncertainty must be reported transparently. Reaction-time distributions should be inspected for outliers and condition-specific errors, and preregistered decisions should govern exclusions, transformations, and inferential contrasts. A significant main effect of facilitation does not establish a psychological mechanism; mediation, discriminant outcomes, moderated effects, and independent replication progressively strengthen the inference. The most informative studies report effect sizes and uncertainty intervals rather than only directional verbal claims.
Psychophysiology and Neural Inference
Physiological recordings can extend self-report by probing processes that participants might not easily articulate. Facial electromyography may index components of affective reactions; pupil changes and autonomic signals may covary with attention or arousal; electroencephalography can track the timing of perceptual or semantic processing. Each measure is multiply determined. A physiological difference should therefore be interpreted according to the experimental manipulation and convergent evidence, not treated as a direct readout of aesthetic pleasure.
Winkielman et al. (2006) used converging approaches in the study of attractive prototypes, including processes related to categorization and affective response. Such multimethod work is valuable because it reduces dependence on a single rating. However, physiological activity associated with liking can also reflect attention, familiarity, task difficulty, or arousal. Its explanatory value emerges from a carefully planned pattern across methods and control conditions.
The link between subjective ease and an interpreted bodily cue was explored by Forster et al. (2015). The research used feedback construed as a signal of processing ease to ask whether interpreted fluency contributes to liking. The findings encourage the view that a subjective inference about one's processing can influence evaluation. This does not imply that the experimental feedback measured a natural neural marker of fluency; it was an intervention affecting participants' interpretation of their state.
Neuroaesthetic studies often identify systems associated with valuation, sensory representation, emotion, and learned meaning during artistic evaluation. These associations offer context for a fluency account but do not specify its unique neural implementation. When activation covaries with beauty ratings, one still needs to ask whether it reflects perceptual efficiency, pleasure, salience, task demands, or semantic appraisal. A domain-general explanation of fluency and liking requires manipulations that separate these processes within the same participants and stimuli.
The strongest next step is a temporally resolved design coupling independent indices of early sensory facilitation, later semantic comprehension, subjective ease, and moment-by-moment evaluations. Such research can test whether fluency contributes to an initial rapid response or a subsequent reinterpretation of the object. It can also address whether the affective consequence of fluency depends on the extent to which participants understand or believe they understand the event. The neural evidence currently enriches hypothesis generation; the causal assessment remains anchored in the design of the behavioral experiment.
Modalities and Ecological Generalization
The evidence base for processing fluency originated largely in visual experiments with simple figures, symbols, words, and other controlled stimuli. Generalizing this mechanism to music, spoken language, architecture, interfaces, and artworks requires understanding what counts as an easier processing event in each modality. Visual identification speed is not interchangeable with auditory predictability, semantic comprehension, or the felt effort of interpreting complex art. The general hypothesis—that fluent processing can enter aesthetic evaluation—must be tested with measures suited to the specific sensory and cognitive system.
Auditory examples include repeated sound patterns, familiar melodic contours, prosodic regularity, and predictability of musical sequences. Their dynamics are intrinsically temporal: pleasure may follow recognition of an expected event, rhythmic entrainment, or a satisfying deviation from a learned pattern. Popescu et al. (2026) reviewed 19 studies in 13 articles examining links between auditory fluency and aesthetic reactions. Their scoping review reported generally positive associations or effects but identified heterogeneous repetition protocols, limited examination of affective mediators, and insufficient attention to musical training and individual preferences.
This auditory literature makes clear why modality is a moderator rather than merely a different stimulus label. The same person can recognize a melody without liking its performance, enjoy an unpredictable musical development, or prefer a familiar piece because of autobiographical association. A simple repeated-exposure effect can therefore contain sensory, affective, episodic, social, and semantic influences. The causal contribution of fluency can be strengthened by manipulating prediction or processing demands while controlling the identity and associative meaning of the sound.
Ecological contexts introduce multiple, interacting sources of fluency. A visitor in a museum encounters lighting, scale, prior knowledge, crowds, explanatory text, and institutional framing. A person choosing between two app interfaces may value navigational clarity, while evaluating a poster for artistic originality can invoke an entirely different goal. Each context alters both the processing task and the criterion of a successful response. Transfer from randomized laboratory stimuli to natural aesthetic encounters therefore requires field experiments, repeated observations, and studies with realistic choices.
Digital presentation offers opportunities for controlled field investigations: interfaces can vary legibility, visual order, predictability, and explanatory labeling while measuring comprehension, satisfaction, and subsequent use. But behavioral engagement should not be conflated with aesthetic pleasure. A design that causes longer viewing time could be intriguing, confusing, distressing, or enjoyable. The strongest studies combine behavioral outcomes with validated experience measures and an explicit causal contrast.
Replication, Meta-Analysis, and Evidential Quality
A scientific synthesis of processing fluency and aesthetic pleasure requires careful separation of evidence about experimental effects from evidence about their psychological mechanisms. The laboratory record establishes that manipulations such as priming, contrast, repetition, and conceptual accessibility can shift liking judgments in particular tasks. Carbon (2010) further examines the longer-term cycle of aesthetic preference and its changing relation to familiarity. It does not automatically establish that the same unobserved affective mediator explains every result. Meta-analyses of exposure aggregate outcomes from diverse designs, while targeted experiments attempt to isolate the contribution of fluency itself.
Bornstein (1989) established a historical basis for studying methodological moderators of mere exposure. The subsequent Montoya et al. (2017) synthesized 268 exposure–response curves from 81 articles and characterized the aggregate relationship as involving positive linear and negative quadratic components. These data bear on repetition-related liking, not exclusively on aesthetic pleasure induced by subjective processing ease. Conflating the two would overstate mechanistic evidence.
There are important instances of theoretically informative constraints and partial dissociations. Forster et al. (2015) found that felt ease and certainty could vary under within- and between-participant manipulations, while liking was affected mainly by within-participant fluency variation. Landwehr et al. (2020) examined whether ease creates a positive shift, amplifies evaluation, or both. These findings complicate simple accounts while supporting more discriminating, explicitly conditional research programs.
Threats to validity include stimulus sampling, demand characteristics, confounding of repetition with familiarity or expectation, insufficient assessment of awareness in priming experiments, selective outcome reporting, inadequate statistical treatment of stimulus and observer variability, and reliance on a single aesthetic scale. The risk of publication bias is especially relevant where hypotheses predict small context-dependent effects. Scientific quality improves with preregistration, adequately powered designs, independent replication, transparent stimuli and data, precise manipulation checks, and hypotheses that specify the direction and moderators of the expected effect.
Some published studies are read here through their abstracts or accessible summaries, while others have publicly available full text; source access should be documented without claiming an exhaustive retrieval of all relevant studies. This monograph is a critical narrative synthesis. It does not have a prespecified search strategy, dual screening, formal risk-of-bias judgments for every study, or a new quantitative meta-analysis. Its role is to articulate the empirical record and expose the experiments needed to distinguish competing mechanisms, while preserving uncertainty where the designs cannot settle causation.
Current Research Through 2026
The newest stage of research retains the distinction between perceptual and conceptual facilitation while also examining the computational representations that may predict them. Jiang et al. (2026) reported two masked-priming experiments in which contour- and word-based matches were used to facilitate different levels of processing in aesthetic liking judgments about everyday objects. Matching primes increased liking under the tested conditions, and the perceptual effect weakened with greater presentation duration while conceptual priming remained more stable. The abstract supports a stage-sensitive account, but it does not determine the causal mediator in every condition.
A computational extension appears in Lin et al. (2026), who analyzed aesthetic experience through human-aligned deep convolutional neural networks. In their study, a representation-based integration measure predicted perceptual fluency for paintings particularly well at earlier network layers and fine spatial scales. Yet that measure was less successful in predicting liking. Deeper representations, incorporating higher-order information, were more predictive of liking. The partial dissociation provides an unusually explicit reason to measure perceptual processing and aesthetic valuation separately.
Computational fits are not direct demonstrations that artificial networks reproduce human pleasure. A model can correlate with human fluency judgments or beauty ratings while using different internal processes and without establishing subjective aesthetic experience. Its contribution is methodological: layer-wise representations offer testable proxies for transformations from image properties toward conceptual content. Researchers should ask whether model comparisons generalize across painting genres, observers, cultures, and ecological viewing conditions, and whether their advantage persists on genuinely held-out materials.
A theoretical proposal by Walther (2026), published online ahead of its scheduled 2027 print issue, develops a categorization-based interpretation of visual aesthetic pleasure. It suggests that successful classification across levels of the visual hierarchy can itself be rewarding and can help organize findings about prototypicality, symmetry, complexity, and self-relevance. As a conceptual review and proposal, it gives hypotheses to test rather than a universally demonstrated causal account. Its strength is a commitment to specifying the operations that make a visual stimulus recognizable.
The Yoo et al. (2024) framework and Winkielman et al. (2025) synthesis add another dimension: fluency is interpreted through expectation, goal, and experience. Together with the 2026 experiments and computational work, these contributions indicate a productive direction for the field. Future theories need to predict when easier processing yields pleasure, when it strengthens an existing evaluation, and when the reward arises primarily from discovering new structure rather than from recognizing it effortlessly.
Artificial Era and Human–Artificial Aesthetic Encounters
The Artificial Era expands the empirical environments in which humans encounter images, music, language, and designed experiences. Artificial systems produce visual objects, captions, rankings, recommendations, interactive forms, and generated variations that can change the ease of perceptual or conceptual processing. This establishes a domain for investigating human aesthetic judgments under new conditions of attribution, personalization, and adaptive presentation. The central psychological variables remain those measured in ordinary fluency research: accessibility, clarity, expectation, interpretation, observer goals, and evaluated pleasure.
A human observer may find an AI-generated image immediately legible because its composition follows familiar visual conventions. Alternatively, a generative system can produce unfamiliar combinations that reward prolonged interpretation. A descriptive caption could increase conceptual accessibility while decreasing surprise or originality. These mechanisms are experimentally approachable by varying image structure, label information, source attribution, and the time allowed for viewing in factorial designs. The relevant outcomes include liking, interest, perceived novelty, judged artistic value, and comprehension.
Attribution to a human creator or an artificial generator may affect evaluation even when the stimulus is identical. This is a possible social and conceptual pathway that must be distinguished from perceptual fluency: different source labels can alter expectations, perceived intentionality, or institutional meaning without changing the sensory image. A study that finds a source-label effect on liking has not thereby demonstrated a fluency effect. The two processes can be crossed and tested for interaction rather than grouped into one explanation.
Computational systems can also be scientific instruments for generating stimuli at controlled levels of ambiguity, typicality, contour structure, and semantic congruity. The approach promises greater stimulus diversity, but it can introduce hidden correlations between supposedly independent dimensions. Generated images should be screened for semantic content, familiarity, emotional valence, and unintended visual artifacts. Model-estimated fluency should be validated against human processing indicators rather than assumed from the generator's own scoring function.
The broader distinction between human aesthetic response and artificial information processing is experimentally essential. A computational network can model categorization and predict ratings, as in Lin et al. (2026). This establishes useful functional correspondences between measurable representations and human judgments. Whether an artificial system has aesthetic experience is a distinct research and philosophical question requiring its own operational framework. This dossier examines human aesthetic pleasure and computational models of its determinants, while situating their extension to artificial systems within a wider interdisciplinary research agenda.
Integrated Scientific Synthesis
The research can be integrated around five linked but separable elements: the stimulus and its learned significance; the processing operation; the subjective signal or interpretation associated with that operation; the observer's goal and comparison standard; and the final aesthetic response. Fluency manipulations influence one or more processing stages. Their effects on evaluation are conditional on whether ease becomes affectively positive, supplies usable metacognitive information, signals familiarity or category fit, amplifies prior valence, or supports a valued discovery.
The foundational evidence supports a relatively clear statement: facilitating perception or conceptual access can increase liking in controlled experiments, and exposure or categorical learning can change preference across encounter histories. Reber et al. (1998), Reber et al. (2004), and Winkielman et al. (2006) provide influential demonstrations and synthesis. Those results constrain the phenomenon but allow several interpretations of the mediating psychological processes.
Evidence from relative-fluency contrasts, positive and negative target valence, interest appraisals, and art interpretation shows that pleasure is contingent. The same processing facilitation can have different consequences according to comparison context or prior evaluation. Graf et al. (2015), Forster et al. (2015), and Landwehr et al. (2020) show why the field must move beyond a monotonic mapping from ease to liking. Their findings are compatible with fluency playing several causal roles, not with its explanatory disappearance.
A dynamic account gives the research a coherent temporal structure. Recognition may start with manageable sensory information, proceed toward interpretative success, and end with a revised appraisal. Interest may motivate continued work before pleasure emerges. Expectations and goals determine which processing changes are experienced as meaningful. The computational work of Lin et al. (2026) makes the same point from another angle by demonstrating that representations predictive of visual fluency do not necessarily best predict liking.
The most defensible integrated conclusion is conditional and mechanistic: processing fluency is one experimentally tractable contributor to aesthetic pleasure, operating within a larger system of expectations, interpretive activity, emotion, learned categories, and evaluation. A mature theory specifies the conditions under which each pathway dominates and measures the relevant stages independently. This conclusion preserves the explanatory power of classical fluency theory while allowing complex and sometimes demanding aesthetic experiences to be understood as systematic phenomena.
Research Agenda
Future research should prioritize direct comparisons among competing mechanisms under preregistered, sufficiently powered designs. A particularly revealing factorial experiment would cross processing facilitation, initial target valence, and the observer's epistemic goal while keeping target identity and sampling properties controlled. Hedonic marking predicts a positive shift associated with ease, fluency amplification predicts stronger effects conditional on existing valence, and interest-based models predict changes when processing is chosen as an intellectually rewarding challenge. Measuring all three outcomes allows the theories to be compared within the same population.
The field also needs time-resolved experiments. Researchers can vary prior disfluency, expected difficulty, and eventual resolution separately, then collect rapid measures of processing efficiency and later measures of felt ease, interest, and liking. This approach can discriminate benefits of absolute accessibility from benefits of improvement or successful surprise. Longitudinal exposure designs should record how preferences develop and potentially saturate, examining novelty seeking and category learning alongside habituation.
Cross-modal and cross-cultural generalization must be established through comparable but modality-appropriate measures. Music, painting, everyday design, and interactive interfaces differ in temporal structure, meaning, and behavioral consequences. Samples should be diversified across training histories and cultural contexts, with cross-linguistic validation of terms such as beauty, pleasure, fluency, and interest. Reports should publish stimuli, exclusion decisions, uncertainty intervals, and model checks when feasible. Only this kind of cumulative evidence can support general principles extending beyond narrow laboratory stimulus classes.
Computational work can help generate quantitative predictions about perceptual structure, representation depth, category similarity, and the expected trajectory of processing. Such models should be evaluated on independently sampled images and participants and compared with simpler baselines. Importantly, successful prediction of liking should be tested against prediction of fluency itself. The partial dissociation identified by Lin et al. (2026) is a reason to keep the two targets distinct during model development.
Finally, a transparent evidence synthesis with formal risk-of-bias assessment and predefined inclusion criteria would help establish the sizes and reliability of fluency-specific effects separately from mere-exposure effects. Large multi-laboratory replications could vary priming, stimulus category, measurement sequence, and within-versus-between-participant comparisons. Such research would permit stronger statements about transportability and the causal role of subjective experience than are possible from individual demonstrations and broad narrative reviews.
Conclusion
Processing fluency and aesthetic pleasure are connected through identifiable psychological processes. Easier perception, recognition, and comprehension can improve liking, while repetition, categorical typicality, contrast, priming, and explanatory context provide experimentally manipulable routes to processing change. The historical progression from mere exposure through the modern fluency account established an important explanatory program linking cognition and aesthetic evaluation.
Subsequent evidence sharpens the boundaries of the relationship. Felt fluency and liking can dissociate; initially difficult encounters may become pleasurable through insight; existing evaluations can be amplified by facilitation; and expert observers or motivated viewers may value complexity and active understanding. Recent research separates perceptual and conceptual mechanisms, models the representational stages involved, and examines the role of expectation, goals, and changing comparison standards. These findings support a layered, conditional account of aesthetic response.
The enduring scientific problem is to explain when a processing event becomes a pleasurable aesthetic encounter. Progress requires measurements that distinguish objective efficiency, subjective experience, emotional valence, interpretation, interest, and judgment. The strongest theory of processing fluency and aesthetic pleasure will account for their experimentally demonstrated connections alongside the conditions in which they diverge, thereby clarifying how mental processing contributes to the experience of art and to aesthetic life more broadly.
References
- Alter, A. L., & Oppenheimer, D. M. (2009). Uniting the tribes of fluency to form a metacognitive nation. Personality and Social Psychology Review, 13(3), 219–235. https://doi.org/10.1177/1088868309341564
- Bar, M., & Neta, M. (2006). Humans prefer curved visual objects. Psychological Science, 17(8), 645–648. https://doi.org/10.1111/j.1467-9280.2006.01759.x
- Belke, B., Leder, H., Strobach, T., & Carbon, C.-C. (2010). Cognitive fluency: High-level processing dynamics in art appreciation. Psychology of Aesthetics, Creativity, and the Arts, 4(4), 214–222. https://doi.org/10.1037/a0019648
- Bornstein, R. F. (1989). Exposure and affect: Overview and meta-analysis of research, 1968–1987. Psychological Bulletin, 106(2), 265–289. https://doi.org/10.1037/0033-2909.106.2.265
- Carbon, C.-C. (2010). The cycle of preference: Long-term dynamics of aesthetic appreciation. Acta Psychologica, 134(2), 233–244. https://doi.org/10.1016/j.actpsy.2010.02.004
- Forster, M. (2020). Processing fluency. In M. Nadal & O. Vartanian (Eds.), The Oxford handbook of empirical aesthetics (pp. 430–446). Oxford University Press. https://doi.org/10.1093/oxfordhb/9780198824350.013.21
- Forster, M., Fabi, W., & Leder, H. (2015). Do I really feel it? The contributions of subjective fluency and compatibility in low-level effects on aesthetic appreciation. Frontiers in Human Neuroscience, 9, Article 373. https://doi.org/10.3389/fnhum.2015.00373
- Forster, M., Gerger, G., & Leder, H. (2015). Everything's relative? Relative differences in processing fluency and the effects on liking. PLOS ONE, 10(8), e0135944. https://doi.org/10.1371/journal.pone.0135944
- Graf, L. K. M., & Landwehr, J. R. (2015). A dual-process perspective on fluency-based aesthetics: The pleasure-interest model of aesthetic liking. Personality and Social Psychology Review, 19(4), 395–410. https://doi.org/10.1177/1088868315574978
- Jiang, L., Liu, C., Gao, C., Hao, J., & Ding, J. (2026). Aesthetic liking in automatic processing: Distinct effects of perceptual and conceptual fluency. Attention, Perception, & Psychophysics, 88(3), Article 94. https://doi.org/10.3758/s13414-026-03237-4
- Landwehr, J. R., & Eckmann, L. (2020). The nature of processing fluency: Amplification versus hedonic marking. Journal of Experimental Social Psychology, 90, 103997. https://doi.org/10.1016/j.jesp.2020.103997
- Leder, H., Belke, B., Oeberst, A., & Augustin, D. (2004). A model of aesthetic appreciation and aesthetic judgments. British Journal of Psychology, 95(4), 489–508. https://doi.org/10.1348/0007126042369811
- Leder, H., Carbon, C.-C., & Ripsas, A.-L. (2006). Entitling art: Influence of title information on understanding and appreciation of paintings. Acta Psychologica, 121(2), 176–198. https://doi.org/10.1016/j.actpsy.2005.08.005
- Leder, H., Gerger, G., Dressler, S. G., & Schabmann, A. (2012). How art is appreciated. Psychology of Aesthetics, Creativity, and the Arts, 6(1), 2–10. https://doi.org/10.1037/a0026396
- Lin, Y., Verlinde, L., Wagemans, J., & Op de Beeck, H. (2026). Tracing aesthetic experience from perception and conception to appraisal using deep convolutional neural networks. iScience, 29(6), 116232. https://doi.org/10.1016/j.isci.2026.116232
- Montoya, R. M., Horton, R. S., Vevea, J. L., Citkowicz, M., & Lauber, E. A. (2017). A re-examination of the mere exposure effect: The influence of repeated exposure on recognition, familiarity, and liking. Psychological Bulletin, 143(5), 459–498. https://doi.org/10.1037/bul0000085
- Muth, C., & Carbon, C.-C. (2013). The aesthetic Aha: On the pleasure of having insights into Gestalt. Acta Psychologica, 144(1), 25–30. https://doi.org/10.1016/j.actpsy.2013.05.001
- Oppenheimer, D. M. (2008). The secret life of fluency. Trends in Cognitive Sciences, 12(6), 237–241. https://doi.org/10.1016/j.tics.2008.02.014
- Palmer, S. E., Schloss, K. B., & Sammartino, J. (2013). Visual aesthetics and human preference. Annual Review of Psychology, 64, 77–107. https://doi.org/10.1146/annurev-psych-120710-100504
- Popescu, A., & Holman, A. C. (2026; first published online 2024). Loop and enjoy: A scoping review of the research on the effects of processing fluency on aesthetic reactions to auditory stimuli. Psychological Reports, 129(3), 1833–1868. https://doi.org/10.1177/00332941241277474
- Reber, R., & Schwarz, N. (1999). Effects of perceptual fluency on judgments of truth. Consciousness and Cognition, 8(3), 338–342. https://doi.org/10.1006/ccog.1999.0386
- Reber, R., Schwarz, N., & Winkielman, P. (2004). Processing fluency and aesthetic pleasure: Is beauty in the perceiver's processing experience? Personality and Social Psychology Review, 8(4), 364–382. https://doi.org/10.1207/s15327957pspr0804_3
- Reber, R., Winkielman, P., & Schwarz, N. (1998). Effects of perceptual fluency on affective judgments. Psychological Science, 9(1), 45–48. https://doi.org/10.1111/1467-9280.00008
- Reber, R., Wurtz, P., & Zimmermann, T. D. (2004). Exploring "fringe" consciousness: The subjective experience of perceptual fluency and its objective bases. Consciousness and Cognition, 13(1), 47–60. https://doi.org/10.1016/S1053-8100(03)00049-7
- Silvia, P. J. (2005). Emotional responses to art: From collation and arousal to cognition and emotion. Review of General Psychology, 9(4), 342–357. https://doi.org/10.1037/1089-2680.9.4.342
- Silvia, P. J. (2005). What is interesting? Exploring the appraisal structure of interest. Emotion, 5(1), 89–102. https://doi.org/10.1037/1528-3542.5.1.89
- Van de Cruys, S., & Wagemans, J. (2011). Putting reward in art: A tentative prediction error account of visual art. i-Perception, 2(9), 1035–1062. https://doi.org/10.1068/i0466aap
- Walther, D. B. (2026, online ahead of 2027 issue). The sweet spot of seeing: A categorization-based account of visual aesthetic pleasure. Current Opinion in Psychology, 73, Article 102351. https://doi.org/10.1016/j.copsyc.2026.102351
- Winkielman, P., Halberstadt, J., Fazendeiro, T., & Catty, S. (2006). Prototypes are attractive because they are easy on the mind. Psychological Science, 17(9), 799–806. https://doi.org/10.1111/j.1467-9280.2006.01785.x
- Winkielman, P., Jasko, K., & Yoo, J. (2025). Fluency shapes evaluations: Feelings, interpretations, expectations, and goals. Psychology of Learning and Motivation, 82, 179–205. https://doi.org/10.1016/bs.plm.2025.03.001
- Yoo, J., Jasko, K., & Winkielman, P. (2024). Fluency, prediction and motivation: How processing dynamics, expectations and epistemic goals shape aesthetic judgements. Philosophical Transactions of the Royal Society B: Biological Sciences, 379(1895), Article 20230326. https://doi.org/10.1098/rstb.2023.0326
- Zajonc, R. B. (1968). Attitudinal effects of mere exposure. Journal of Personality and Social Psychology, 9(2, Pt. 2), 1–27. https://doi.org/10.1037/h0025848
- Zajonc, R. B. (1980). Feeling and thinking: Preferences need no inferences. American Psychologist, 35(2), 151–175. https://doi.org/10.1037/0003-066X.35.2.151
- Zhang, L., Atari, M., Schwarz, N., Newman, E. J., & Afhami, R. (2022). Conceptual metaphors, processing fluency, and aesthetic preference. Journal of Experimental Social Psychology, 98, Article 104247. https://doi.org/10.1016/j.jesp.2021.104247