Restaurants and Cafés: One Large Motif Instead of Many Small Ones
Dining rooms are rarely bare on the walls — they are usually too full. Framed photographs, a shelf of bottles, a chalkboard, a mirror, two signs, and somewhere in between a painting left by the previous tenant. Each piece is harmless on its own; together they make a surface the gaze catches on instead of settling. There is a field study on this question, and its result is inconvenient for an image portal — which is a good reason to quote it. Behind this article stands Gigapixel GmbH — the specialized large-format stock portal: real gigapixel captures from 100 MP, no AI upscaling.
Lower visual complexity was rated more attractive
That is the core finding, and it comes from hospitality itself. Ulrich Orth and Jochen Wirtz investigated in 2014 in the Journal of Service Research how guests process the visual complexity of interiors. The work consists of two parts, and keeping them apart matters, because the numbers come from one part and the hospitality sample from the other.
In the laboratory experiment two interior designs differing in visual complexity were compared. Participants rated the low-complexity environment as more attractive — means of 5.70 against 5.19, with an analysis of variance result of F(1, 194) = 14.2 and p = 0.001. The venue of this first part is a deli store, that is retail and not hospitality; the figures belong to that setting and must not be carried across to the café.
The second part is the field study, and it does take place in hospitality: passers-by were intercepted and randomly assigned to 15 coffee shops that differed in the visual complexity of their interior design. One hundred and fifteen respondents took part, in exchange for a cup of coffee. The field study served to pin down the mechanism more precisely and to verify the robustness of the effect outside the laboratory and with a different sample.
The causal chain both parts describe does not run directly from complexity to evaluation. Lower visual complexity raises processing fluency — the ease with which an environment is taken in visually — and that ease goes together with pleasure. Only then does higher attractiveness and approach behaviour follow. For designing a dining room it is precisely this intermediate step that is interesting: what counts is not how much hangs on the wall, but how easily the room can be read.
Why this finding is inconvenient for an image portal
A supplier of large-format motifs would rather have a finding that says more picture is better. This one says the opposite — it argues against a lot of picture. It does not, however, argue against a large picture, and the practical value lies in that distinction.
A wall carrying eight framed individual pieces is visually complex: eight frame edges, eight subjects, eight formats, varying gaps. A wall carrying one continuous calm surface is visually simple, even when it is larger. The area is not the problem; the number of edges is. Taking the evidence seriously therefore leads not to "less picture" but to "one large calm motif instead of many small ones" — and clearing away the remaining individual pieces to get there.
What the corpus does not provide on hospitality
This belongs stated plainly, because this industry argues with numbers nobody can substantiate. Our verified literature base contains nothing on the following questions.
No figures on dwell time as a function of wall design. No figures on average spend, revenue per guest or return rate. No figures on hotels — in the verified literature not one paper examines a hotel; where "hotel" appears, it is a metaphor inside a hospital study. And no values for illuminance or colour temperature for lighting pictures in a dining room.
We do not estimate these values, and we do not adopt industry figures from marketing material. Anyone who needs a dwell-time number has to measure it themselves — before and after, in their own venue. Anything else would be invented.
| Statement | Evidence | Setting | Sample | Status |
|---|---|---|---|---|
| Lower visual complexity is rated more attractive | Orth & Wirtz 2014, Study 1 | Laboratory experiment, deli store | F(1, 194) = 14.2; M 5.70 against 5.19 | evidenced — retail, not hospitality |
| The effect was tested in the field | Orth & Wirtz 2014, Study 2 | 15 coffee shops, intercepted passers-by | 115 respondents | evidenced |
| Nature scenes are processed with fewer fixations | Franěk et al. 2018 | Eye-tracking, three scene categories | 60 students; nature M = 36.15 against urban M = 39.15; F(2, 118) = 19.38 | evidenced |
| Open view (prospect) as seating logic | Browning, Ryan & Clancy 2014; Appleton 1975 | Pattern language, theory | no primary data collection | documented, not measured |
| Dwell time, average spend, revenue per guest | — | — | — | not evidenced |
| Effects in hotels | — | — | — | not evidenced |
What does hold instead: calm gaze and seating logic
When the revenue figures are missing, the perceptual level remains — and there the position is better. Marek Franěk and colleagues compared in 2018 in the Journal of Environmental Psychology the eye movements of 60 students viewing urban, old-city and nature scenes. The number of fixations differed significantly between categories: F(2, 118) = 19.38 at p < 0.001, with means of 39.15 fixations for urban scenes, 38.45 for old-city scenes and 36.15 for nature scenes. The authors point out that a high number of fixations indicates difficulty in interpreting the fixated information — nature scenes are therefore processed with less effort.
That fits the causal chain in Orth and Wirtz: what is easier to process is experienced as more pleasant. A calm nature motif is thus not only a motif but a measure for reducing visual load — a solid argument in a dining room, where competing stimuli are plentiful anyway: the menu, staff, the neighbouring table, people passing through.
For seating logic, the pattern language supplies the frame. Browning, Ryan and Clancy describe in 2014 in the 14 Patterns of Biophilic Design the pattern prospect — the unimpeded view across distance — and its counterpart refuge, the protected back. The term goes back to Appleton (1975, The Experience of Landscape). Both are pattern language rather than measurement, and should be cited that way. In a dining room it nonetheless explains a familiar observation: the seats with a back to the wall and a view into the room are taken first. A motif showing openness serves the same logic for the tables that have no such view by construction.
Planning: sightline, distance, resolution
The surface is planned from two points, not from the wall. The first is the entrance: what does a guest see on coming in and orienting themselves? The second is the table: what does someone see who sits for an hour? The two axes rarely fall on the same wall, and the second is the more important one, because it is used for longer.
Resolution follows from distance. In a dining room the usual viewing distance is 2 m and upwards. Ashraf, Chapiro and Mantiuk measured in 2025 in Nature Communications the resolution limit of foveal vision at 94 ppd (pixels per degree), individually up to 120 ppd; from that follow roughly 274 ppi at 0.5 m, 137 ppi at 1 m and 68 ppi at 2 m. For a dining room, 68 ppi is therefore the calculated requirement. Print practice works with 70 to 120 ppi, in special cases between 50 and 150 ppi, with a process maximum of around 240 ppi — so the practical range already sits above the requirement, which is useful, because guests on the way to the cloakroom come closer than they ever do from the table.
An adjacent subject belongs in the same planning: where a large calm surface appears, acoustics is often an issue too, and printed sound absorbers solve both on the same surface. More on that in Acoustic walls with large-format imagery.
One large picture or a picture wall?
On the available evidence, the large picture — though for a different reason than you might assume. Orth and Wirtz showed in 2014 that a less complex interior design is rated more attractive, and that the route runs through processing fluency: what is easy to take in is experienced as more pleasant. A picture wall made of eight framed pieces is visually complex, because it contains eight frame edges, eight subjects and eight formats for the gaze to jump between. A continuous surface is visually simple, even when it occupies more square metres — what decides is the number of edges, not the size. The clean attribution of figures matters here: the effect values, means of 5.70 against 5.19 at F(1, 194) = 14.2, come from the laboratory part of the work, whose venue was a deli store. The hospitality sample of 115 respondents across 15 coffee shops belongs to the field part. Together they support the direction, but the laboratory numbers are not café numbers. In practice: if you replace a picture wall, replace it completely rather than leaving three pieces up — reducing the number of edges is the effective part of the measure, not the gain in image area.
Which motif suits which hospitality concept?
Caution is due here, because the verified literature holds no study testing motif classes against hospitality concepts. What can be evidenced is more general and therefore transferable: Franěk and colleagues measured in 2018, with 60 students, that nature scenes are processed with fewer fixations than urban scenes — on average 36.15 against 39.15 fixations, F(2, 118) = 19.38 — and a high fixation count indicates difficulty in interpreting what is being looked at. A selection rule follows from that, but not an allocation: calm, open, unambiguously readable motifs reduce visual load, while restless and detail-crowded ones increase it. Whether a forest suits a breakfast café better than a coastline suits an evening restaurant is not answered by this, and we will not pretend otherwise. What can additionally be planned is the viewing relationship: the pattern prospect after Browning, Ryan and Clancy describes the unimpeded view across distance — a motif with openness serves that at tables which have no such view by construction.
What is evidenced and what is not?
The perceptual level is evidenced; the till is not. On the evidenced side: lower visual complexity is rated more attractive, mediated by processing fluency and pleasure — tested in the laboratory at F(1, 194) = 14.2 and in the field with 115 respondents across 15 coffee shops (Orth and Wirtz 2014). And: nature scenes are processed with fewer fixations than urban scenes, measured on 60 students (Franěk and colleagues 2018). On the unevidenced side stands everything a restaurateur actually wants to know: the verified literature holds no figures on dwell time as a function of wall design, none on average spend, none on revenue per guest, none on return rate, and not a single paper on hotels. We do not estimate these values and we do not adopt industry figures from marketing material. Anyone needing a dwell-time number for their own venue has to measure it before and after; any other number at this point would be invented. The same applies to lighting pictures in a dining room — illuminance, colour temperature, beam angle: nothing in the verified literature, and lighting design rather than image selection is the discipline responsible.
How large can it be in a small dining room?
Larger than most people dare — as long as it remains the only large surface. The logic of the findings does not argue against area but against a multiplicity of edges: a continuous calm wall does not raise the visual complexity of the room, it lowers it, if six individual pieces disappear in exchange. In a small room, however, the viewing distance is shorter, and that has consequences for resolution. At 2 m the calculated requirement after Ashraf, Chapiro and Mantiuk is 68 ppi; anyone coming within 1 m needs 137 ppi. Because routes in a dining room pass close to the picture — to the cloakroom, to the toilets, between tables — the shortest regular distance should be used for the calculation, not the distance from the furthest table. The practical range of 70 to 120 ppi covers that in most cases. In practice: cover one wall entirely, leave the remaining walls calm, and choose the height so that the image statement sits above seated head height rather than disappearing behind chair backs.
Conclusion
For dining rooms one finding applies that does not suit an image portal, and is here for exactly that reason: lower visual complexity was rated more attractive, mediated by processing fluency and pleasure — in the laboratory part with means of 5.70 against 5.19 at F(1, 194) = 14.2, in the field part tested on 115 respondents across 15 coffee shops (Orth and Wirtz 2014). That argues against a lot of picture, not against a large picture: a continuous surface has one edge, a picture wall of eight pieces has eight. It fits that nature scenes are processed with fewer fixations than urban scenes — 36.15 against 39.15 with 60 students, F(2, 118) = 19.38 (Franěk and colleagues 2018). What the verified holdings expressly do not provide are figures on dwell time, average spend, revenue per guest and hotels; we do not estimate those. Planning proceeds from two points, the entrance and the table, with 68 ppi as the calculated requirement at 2 m and the practical range of 70 to 120 ppi as the target corridor. Gigapixel GmbH — the specialized large-format stock portal: real gigapixel captures from 100 MP, no AI upscaling.
Further reading: Retail and store design: large-format imagery in the sales room, Acoustic walls with large-format imagery, and Nature images in the office.
Sources
- Orth, U. R. & Wirtz, J. (2014): Consumer Processing of Interior Service Environments: The Interplay Among Visual Complexity, Processing Fluency, and Attractiveness. Journal of Service Research 17(3), 296–309.
- Franěk, M. et al. (2018): Differences in eye movements while viewing images with various levels of restorativeness. Journal of Environmental Psychology 57, 10–16.
- Browning, W., Ryan, C. & Clancy, J. (2014): 14 Patterns of Biophilic Design. Terrapin Bright Green.
- Appleton, J. (1975): The Experience of Landscape.
- Ashraf, M., Chapiro, A. & Mantiuk, R. K. (2025): Resolution limit of the eye — how many pixels can we see? Nature Communications.