Did you ever wonder why in the era of Google Art Project and virtual museums people still visit 'real' art galleries? We have access to almost everything  from the comfort of our living room and yet we still can be bothered enough to travel to and physically walk around some distant building. We can google all paintings of our favourite artists but we still prefer a curator to pick a subset and arrange it for us. And often we're even willing to pay for this!

But even if you're not much into art, art galleries are quite interesting. This is from purely architectural point of view. Art and space... or art in space, is the only thing that guides your exploration in these buildings. No arrows telling you where to go, like in an airport. No zones explicitly dedicated to different functions, like in a library. No scheduled classes happening in assigned rooms, like in a school. In an art gallery, you are the master of your own path. There's nothing and no one telling you where to go next, where to stop, what to look at. Or is there?

elodie_focus.jpg

You must have noticed that some museums are like a maze, or one long corridor. You walk through a series of rooms looking at art pieces in a sequence - one after another. Well, here is a shocker: someone designed these sequences for you. Someone wanted you to see those things in this particular order. Those people are called curators and they're usually rather fantastic at what they do.

But there's this other group of museums on the other side of the spectrum. The galleries from this group are either open-plan, or have multiple connections between rooms. So that you can walk absolutely everywhere you want, in any sequence you desire. 

Which of these types is better? Or maybe both are great, depending on the purpose they serve? The crucial thing is, that even if curators want to allow for absolutely unrestricted walking, they have to understand how space guides visitors' attention - even if only to avoid doing it. On the other hand, they might want to consciously use the space for their purposes: did you ever enter to a room where some impressive painting was suddenly revealed right in front of you? That's one of such situations! Simply because we tend to look in front of us while walking (that's how our body is constructed, right?), a curator can be almost certain that placing an object there will get it more attention. In my PhD I'm spending 3 years trying to find out which of these 'spatial tricks' can work and which don't.

 

Experiment 1

gallery-no_artsy.jpg

My first experiment took place in a pre-arranged art gallery. Susi Bellamy kindly provided some fantastic artworks for the study so that I could make sure 'my' art gallery is a real thing, with real art on truly white cube walls (many psychological experiments use artificially prepared neutral pictures, which has its advantages and disadvantages, but certainly is hard to pass as 'art'). The idea was simple: let people freely explore the space and enjoy the artworks. There was only one small departure from normality: I asked them to wear an eye-tracking device during this. Not a big deal - just a large pair of plastic glasses. But what it gives me in the end is the insight into what people really saw. Not what was in front of them. Not what they stood in front of. But what they saw, how many times they saw it, and for how long they looked at it.

 

Ok, and I must admit - there were some things the visitors of my gallery didn't know when they walked through it. The first such thing was that there happened to be two actual galleries. Half of the people went to the first arrangement, half of them entered the second one. Same space, same pictures, but placed on different wall locations:

click to enlarge

Right, and here is the second little detail: I would ask them which pictures they remember from the inside after they have finished exploring. Being a psychologist, I had my little, sneaky tests prepared for this occasion [click on the arrows bellow to see all of them]: 

So two things were measured:

    1. attention via the eye-tracker (what the visitors actually saw)
    2. memory  via the computer and the miniature test (what they remember from the inside)

    And of course, what draws people's attention would be both its spatial location AND how interesting the picture was itself. That's why I shuffled the sequence of the pictures inside the gallery, so each person saw them in a different order. If, despite that, some locations happened to be more noticeable than others - that'd be the influence of space that I was after. 

     

    Results: Attention

    The eye-tracker basically gives you a video clip with these red dots showing where eye position fixated for at least few milliseconds. These dots can be measured and counted. So the more dots appeared in total on a given picture, the more attention people devoted into it.

    And here is a video sample of it: 

     

    And here is the thing: the amount of attention is very tightly linked to where the pictures were located!  Let me explain the concept of isovist very quickly on the slideshow below:

    As you might have noticed, these isovists are simply geometrical figures - and as such can be described like every other geometrical figure. So, for example, we can calculate its area size. Or the length of its borders. Or how 'spiky' it is. Many things, really. Notice above how one isovist is bigger from the other, how 'spiky' it is compared to the smaller one... All this affects how we see things when walking around space. For example, if the isovist is more 'spiky' it means there're more things interrupting our view (like the column above here).

    click to enlarge

    Right, I'm sorry it's taking so long, but understanding the idea of isovist  is very important to describe what people look at in a white cube-styled art gallery. To let you wait no more: they simply look at.... things that are more visible! Sorry to disappoint you with my trivial answer after such a long introduction. But it's true and I can prove it. In fact - the graph on the left proves it. It shows correlation between eye-tracking data and the isovist area. As you can see, the larger the isovist area , the more glimpses a picture gets (on average). 

    click to enlarge

    Now... why do we bother with doing all this? Because if isovist area almost equals the amount of attention we get in the particular point in space, one really clever thing we can do is to calculate isovist areas for every single point of our layout. What we then get is a map showing the bits of our layout which are the most visible. They have the largest isovist areas and here are marked in red. Blue colour symbolises the opposite (small isovist areas). See how the two examples given in the previous slideshow differ on this map? And now, if you were an amateur artist exhibiting on an art show with others, where would you like your picture to be hanging?

     

    Results: Memory

    But attention - or what people saw - is only one part of it all. You know this feeling when you can stare at something for half an hour only to realise you have no idea what you've been looking at?

     

    What did people really take out of this visit? Did they simply remember those pictures which stand out more (because of their colours or content)? Of course not. We wouldn't be here if it was so simple. And let me be clear - no one goes to art galleries in order to memorise pictures, that's not the point. What matters is the fact that the more you process something, the more you actually attend to a piece of art, the better will be your memory of it when you leave. 

    click to enlarge

    First of all, people who visited the first version of the gallery (Condition 1) , had better results on the computer recognition test (they tended to answer faster). Because the arrangement of the pictures was the only thing that differed Condition 1 from Condition 2, it shows that there must be something about this arrangement that 'helps' us to remember things from the inside better.

     

    click to enlarge

    What is it? Well, the pictures in Condition 1 are separated from each other and only one of them was hung on each wall. Compare to that Condition 2 and you'll realise it's much more 'cluttered', with many pictures close to each other. It was in Condition 2 where pictures' salience (how much they stand out visually) started to decide about what people remembered. But the same factor had no influence whatsoever in Condition 1! The conclusion is this: when space is comfortable and allows easy viewing of single pictures, we will generally remember more. It is difficult (if not impossible) to say what this will be exactly - things that you like will be different from what I like, and we will all leave the gallery with completely different experiences in mind. But as the arrangement of pictures becomes cluttered and we get confused from looking at many things at once - that's when we will simply remember things which stand out more. Those artworks, which were able to 'fight' their way through the crowd, will stay in our memory for longer.


    Experiment 2

    In another experiment I wanted to find out if co-visibility of pictures matters for the way people look at art. Co-visibility is high when you can see many pictures at the same time, as opposed to a situation where pictures are isolated in their own little corners and you're meant to concentrate on one piece at a time. This is a commonly used curatorial tool and some studies already showed that people are in fact sensitive to what they see at the same time. I designed a space that was supposed to facilitate this one effect only. So it was a single art gallery in 2 versions. The same set of pictures in both of them, but one little detail that made the difference:

    panorama-with.jpg

    Half out of my 44 participants went into the space which had little walls built up, that were dividing the pictures from each other. The panoramic photo above shows how it looked like (it was a straight wall in reality). The idea is that nowhere in it you could see more than 2 pictures at once and for majority of the time you were only able to see a single piece.

    The other half faced a rather different situation: 

    panorama-without.jpg

    Each of both sides of the wall contained 6 pictures and you were forced, in a way, to look at them in a very different way.

    And I do realise of course that the space on the photos does not necessarily look like a 'proper' art gallery but it should not matter too much because we're comparing 2 groups which had the same viewing conditions, and the only thing that differed them was the presence or the absence of the dividers.

     

    Results

    So far I'm still analysing the results, so come back in the future for more. But the very preliminary analysis shows that people are very good at being able to focus on the pictures they desire. What does differ the two groups is how they remembered the relations between pictures. So if co-visibility is meant to be used for emphasising such relations (e.g. by the theme, artists, or some other criteria), it can really work when done properly.


    Case Study

    Let me guess your main concern about the experiments presented above: these are not real art galleries. It might look like I'm creating some artificial situations, measuring some abstract things and keep sitting in my comfortable sciency bubble. I was concerned too. So I run a case study. I asked 16 people to walk around a proper art exhibition in a working art gallery. People at BALTIC Centre for Contemporary Art in Gateshead were fantastic and I'm very grateful to Hayley Duff, Alessandro Vincentelli, and everyone else for their help. The exhibition I asked my participants to visit was Thomas Scheibitz's ONE-Time Pad.

    balticmillCOM.jpg

    But this data is still waiting for the analysis. The idea is to find out if the things that came up in the experiments can be confirmed in the way people experience a real art exhibition organised by a prestigious gallery.


    Not enough?
    There's more to come, but for now you can have a look at more detailed academic publications.