Museum experience: why do visitors spend longer with some artworks?

An art encounter begins before someone enters the gallery. A visitor may recognise the artist, have encountered a reproduction on social media, be accompanying a friend or expect an enjoyable family afternoon. These expectations influence what seems relevant. Then attention is allocated: which works enter the visual field, where does the person stand and what earns another look? Interpretation follows, although interpretation can also direct subsequent looking.

For research it helps to separate five questions: what did the visitor expect, notice, interpret, do and remember? They are analytic distinctions rather than five discrete brain mechanisms. A visitor may repeatedly move between looking and interpreting. A conversation in the next room can change what an earlier object means. A later memory can give a short encounter more significance than a long one.

Leder and colleagues' model of aesthetic processing brings perception, classification, attempts at understanding and evaluation into a coherent research framework. It does not prescribe one identical path for every visitor. Leder et al., 2004. Pelowski and colleagues review alternative approaches and show why an art experience may include changes in understanding rather than a simple pleasure response. Pelowski et al., 2016.

A study that measures only stops and durations therefore captures part of the journey. Add questions about meaning at the appropriate moment and, when relevant, ask what visitors remember later. The artwork that receives the longest look need not be the artwork they recall most vividly.

An empty station can suggest freedom to one visitor and loss to another. This does not mean the painting contains one objectively correct message. It means elements in the painting may become connected with existing experiences, memories, beliefs and expectations. A sculpture made from cables may evoke craftsmanship for an engineer, play for a child and an inaccessible art world for someone else.

An association is not simply an opinion given in response to a direct question. Some links can emerge quickly and are difficult to describe. That does not license a researcher to label every unexplained reaction 'subconscious'. The person may have been asked the wrong question, may be uncertain or may be attending to something the researcher overlooked. The responsible approach is to measure associations in a defined task and compare them with what people do and say.

An artwork's appearance is only one source of meaning. Its title, its neighbours on the wall, the exhibition theme, the information supplied by a curator and the actions of other visitors may all prompt additional connections. An installation that seems obscure on its own can become meaningful when placed in an exhibition about migration. Yet the same explanation may narrow a visitor's interpretation when the institution supplies only one possible reading.

The practical question becomes specific: what meanings arise before explanation, which emerge after viewing or reading and which remain later? That is how a museum distinguishes attracting attention from enabling an encounter to develop meaning.

A large painting at eye level near a gallery entrance has a different chance of being seen than a small work hidden beside a crowded doorway. A school group standing around an object can attract interest while simultaneously preventing others from approaching it. A bench encourages time and conversation. A narrow passage may make stopping uncomfortable or cause people to linger while others move past.

These factors are not all inconvenient noise. The room is part of what visitors encounter. But they do change the interpretation of a finding. Are we studying the attractiveness of one artwork, the effectiveness of its display or the overall quality of the visitor journey? A single dwell-time figure cannot answer all three questions.

In the 2014 study, participants viewing works in a museum spent longer with them on average than participants viewing them in a laboratory. Related research from 2015 found differences in valuation and memory across settings. Those findings support the importance of context, but do not isolate a single cause, such as authenticity, architectural setting, expectation or spatial memory cues. Brieber et al., 2015.

For practical museum research, record display height, sight lines, crowding, label position, distance and sequence alongside time. Otherwise a team may move an artwork when the original problem was the placement of a text panel. Experience design starts with a precise diagnosis of what visitors were actually able to see and do.

An observed pause can support several competing explanations. This matrix helps a museum avoid deciding in advance what longer viewing 'must' mean. The examples are hypotheses rather than diagnoses of individual visitors.

Graf and Landwehr describe how aesthetic pleasure and interest can arise from different modes of processing. A challenging stimulus may attract sustained attention because a person wants to understand it, without immediately feeling pleasure. Graf & Landwehr, 2015.

The value of the table lies in keeping multiple possibilities alive. A researcher should only narrow them after collecting the information needed to distinguish them.

Experience with art can change which characteristics a visitor notices and how quickly references, techniques and styles are recognised. This does not mean experts always look better or newcomers have inferior experiences. They may take different routes towards understanding. An expert may need only a brief look to recognise an established technique. A newcomer may spend longer because the same work raises unfamiliar questions.

Vogt and Magnussen compared eye-movement patterns and visual memory in artists and non-artists. Their findings support the relevance of expertise to visual exploration, but should not be turned into a universal hierarchy of visitors. Vogt & Magnussen, 2007.

For museum research, associative differences may be more informative than a crude split by age or formal art education. Some people seek recognition, some challenge, some emotional connection and some an opportunity for conversation. These motivations can coexist or change from one gallery to the next. Thinking in terms of an association network helps describe connections between meanings without turning them into fixed human types.

Begin with what a particular artwork means to a visitor. Then ask whether similar patterns emerge across people under comparable conditions. This protects both the analysis and the visitor's freedom to have an unexpected response.

A title is not merely an inventory label. It can tell a visitor where to look, suggest a story or make an unusual symbol easier to interpret. A short text may offer a useful starting point. A longer explanation may support deeper investigation. But a panel can also impose one reading, disrupt a visitor's pace or attract more attention than the artwork itself.

Leder, Carbon and Ripsas examined how different kinds of titles affect the understanding and appreciation of paintings. Under some experimental conditions, more elaborate titles improved understanding of abstract works without an equivalent rise in aesthetic appreciation. Leder et al., 2006. Reitstätter, Galter and Bakondi studied how visitors used exhibit labels with mobile eye tracking and other methods. Their findings describe patterns in which people move between artwork and text, and the different ways those patterns unfold. Reitstätter et al., 2022.

A newer museum study by Reitstätter, Pesen and Bakondi in 2026 compared altered labels and found that reading behaviour varied considerably, with longer texts associated with higher dropout and particular kinds of image-related information attracting attention. Those findings apply to the studied setting; they are not a rule that every museum should use short labels. Reitstätter et al., 2026.

The testable question is which information helps which visitors encounter which artwork. Try a specific visual-detail question against a more descriptive label and compare interpretation as well as reading and looking.

Consider an entirely fictional temporary exhibition called 'In Transit'. The first room contains a large, colourful painting of travellers on a crowded platform. The second contains a small installation of intertwined thread and old suitcases. The final room presents a restrained painting of an empty station. These imagined works are illustrative; no museum, artist, visitor measurements or client results are being represented as real.

During an exploratory walkthrough, a curator notices that some people spend a long time with the suitcase installation, while others pass it immediately. The museum team wonders whether it is more emotionally powerful than the larger painting. But the installation sits near a narrow passage, is accompanied by a long label, and occasionally becomes a waiting point for groups. These circumstances are plausible alternative explanations for the time spent there.

The researcher therefore keeps several hypotheses open. Perhaps suitcases and thread evoke memories of leaving, belonging or loss. Perhaps visitors are simply trying to work out how the object was made. Perhaps the bottleneck slows movement. Perhaps the title prompts a kind of reflection that the artwork alone does not. No one treats those possibilities as observed results.

The team designs a potential test: compare a label that states a single authoritative interpretation with a label drawing attention to an observable detail and inviting a personal response. The primary outcome is not more seconds. It might be whether visitors identify the detail, can describe their interpretation independently and show curiosity without creating further congestion.

Mobile eye tracking can record where a participant looks, how visual attention shifts between artwork and label and how patterns unfold along a route. With suitable calibration, reliable recording conditions and carefully defined areas of interest, this provides information that a simple stopwatch cannot offer. It may reveal, for example, whether a label attracts attention before or after the artwork.

The technology has constraints. Eyewear, reflective glass, low light, changing viewing distances, head movement and calibration error can affect recording quality. Small details may be hard to resolve. A fixation does not demonstrate comprehension, appreciation or even conscious recognition of the thing being looked at. Pupil size can vary with illumination, cognitive demands and other physiological factors; it cannot serve as a universal emotional meter.

In a 2026 review, Walker discusses findings from mobile eye tracking studies in art museums and notes the contribution of artwork characteristics, exhibition design and visitor differences. He also discusses the remaining challenge of translating such research into practical exhibition decisions. Walker, 2026.

An impressive heatmap can still answer a very narrow question. Use it to investigate a specific observation: which details visitors miss, how they move between object and text, or whether a design variant changes visual exploration. Then gather evidence about interpretation separately.

An art museum does not need to fit every visitor with an EEG cap. It needs a research design that addresses the curatorial decision. Different methods can reveal different aspects of experience, and their limitations should be documented before collection begins.

Thébault and colleagues' museum study published in 2024 argues against relying exclusively on timing and tracking. They combined observed behaviour with visitors' reported experiences to investigate the role of labels and interpretive encounters. Thébault et al., 2024.

Mobile EEG raises additional issues with movement, artefacts, participant comfort and privacy. A recorded brain signal is not an off-the-shelf measure of aesthetic pleasure or memory. Triangulation, meaning the careful use of complementary evidence, is more important than the number of sensors on a participant.

It is tempting to compare artwork A with B as though the works were isolated products. Yet apparent differences can depend on where each is displayed, when someone visits, the people accompanying them and what came earlier in the exhibition. A label may help one artwork and interfere with another. If the order or placement changes, a previously observed difference may change too.

Decide what the unit of analysis will be. If the goal is to understand individual works, observe multiple visitors under comparable conditions. If the goal is to improve an exhibition, treat movement through the space as part of what is being evaluated. If the goal is to test new labels, use a comparison that avoids confounding the text change with crowding or a different visitor group whenever possible.

Record the essentials: time and route, stops and label interactions, visitor characteristics relevant to the question, crowding and disruptions. A small number of well-chosen open questions can reveal whether an apparently important detail was understood. Delayed recall can add a distinct outcome. An exploratory study is useful for finding plausible mechanisms, but does not justify precise population effect estimates without an adequate design and sample.

There is also a selection problem. Visitors willing to complete an interview may be those already most engaged. A heatmap averaged across participants can conceal fundamentally different viewing strategies. Report the distribution and uncertainty rather than a single average whenever possible.

An association map can show the meanings that cluster around a work: recognition, surprise, craft, loss, confusion, anger or belonging. The map does not vote on the 'correct' interpretation. Its value lies in identifying where the curator's intention and visitors' experiences might diverge.

Suppose the fictional suitcase installation is often associated with being trapped or confused. If the exhibition intentionally aims to evoke discomfort, it would be inappropriate to optimise away every negative word. A better possibility is to offer a way into the work without dictating what the visitor must feel. A question pointing to a concrete detail might help visitors construct their own interpretation while preserving the tension.

The design translation is specific. Change one main component when possible, such as label wording or placement. Choose the primary outcome beforehand: visitors' ability to identify a detail, the variety of interpretations, or a meaningful return from reading to viewing. Keep dwell time as a supporting measure rather than the definition of success. Track possible side effects, including congestion, longer queues or a reduction in time available for other works.

This is how associative research informs exhibition design: by converting a plausible explanation into a careful test, not by presenting an association score as the final answer.

A museum can start with a modest project. What matters is preserving the link between the question, the evidence and the decision. The following sequence is designed for a realistic exhibition setting rather than a technology showcase.

1. Define the curatorial decision. Is the museum choosing a label, rethinking a route or trying to make one artwork more accessible? Select one decision that the evidence can inform.
2. Document the current setting. Observe stops, gaze opportunities, group behaviour, movement, light, labels and bottlenecks before treating longer viewing as meaningful.
3. Investigate visitors' associations. Ask what they recognise, enjoy, avoid or find unclear. Use open accounts and structured associative methods where suitable.
4. Bring different forms of evidence together. Compare behaviour, gaze data and visitors' explanations, noting agreement, disagreement and alternative causes.
5. Design one explicit intervention. Prepare two defensible label formulations or placements, with the principal difference specified before testing.
6. Measure the intended outcome and side effects. Record understanding, exploration or recall as appropriate, alongside crowding, accessibility and potential unintended reactions.
7. Decide with uncertainty visible. Report what the data support, what remains unresolved and how confidently the result can be applied in other rooms or visitor groups.

This workflow moves the conversation from 'people seem to look for longer' towards 'this change appears to support a defined museum objective under the conditions we tested'. That is a more useful foundation for experience design.

The artworks that receive the longest looks are not necessarily the artworks visitors like most. Conversely, a brief encounter can be memorable or deeply meaningful. Viewing time is a piece of behaviour that needs context, interpretation and a clear purpose before it becomes evidence for a design decision.

For curators, the opportunity is to use looking behaviour as a clue rather than a ranking. Ask what visitors recognised, what confused or interested them, whether labels supported their own interpretation and what they later remember. Combine methods where they add genuine information, without pretending that a heatmap or EEG signal reveals the complete experience.

The goal is not to keep every visitor in front of every artwork for as long as possible. The goal is to create the conditions in which visitors can find, question and develop meaning for themselves. The broader context is explained in our introduction to associations and experience and the experience series overview.