Free Publishing Rights - You may publish this article as long as you include the following author bio/blurb at the end of the article:
Bob Wayland is the author of "The Mind's Eye Guide to Composition" that explains how the interface of the exploding new fields of cognitive neuroscience and digital photography can guide your digital image workflow to great photos. Go to his web site for a quick introduction to cognitive photography
Aside: This post is meant for all levels of photographers, starting with serious beginners and developing to the level of serious professional: stop at the level that you are comfortable with. Editors may shorten the article as long as the message of the article is not distorted or misrepresented.
Good Photographic Composition using
Visual Working Memory
Put simply, Visual Working Memory (VWM) represents a limited–capacity store of visual information that allows the performance of mental operations on its contents. The information can either originate from visual sensors, your eyes, or it can be retrieved from long-term memory. VWM is used in the processing, integration, disposal and retrieval of visual information. It should therefore come as no surprise that it is important to visual learning and attention. The purpose of this post is to establish how the limitations of VWM affect your ability to create good compositions. Visual perception requires your mind to perform complex manipulations on the signals coming from your eye; these manipulations require you to hold information in memory for a short period of time while processing other bits of information. This short term memory bank is called VWM, and it actively maintains visual information to serve the needs of the subsystems that store and manipulate visual images. VWM has limited capacity that varies depending on the type of information to be stored. VWM’s size is different from one person to another and is also age dependent, decreasing, as you get older.
There are two general ways to determine the characteristics of an object from a visual inspection. Feature searching is the process of finding something different based on its unique visual features, like size, shape, and color (for us geeks this is a bottom-up approach: see glossary). The other is conjunction searching, in which any single unique visual feature does not define an object, but instead by a combination of two or more features (for us geeks this is a top-down approach: see glossary). This is the way you ID an object in an image. People naturally concentrate their attention by spot focusing on a particular location. Anne Treisman’s Feature Integration Theory (FIT) suggests that when you first look at an image, visual separable features like color, line and form are perceived early and automatically without you knowing it. It is the process, in VWM, that directs your attention to interesting locations in a photograph: color and orientation are the hooks that allow the viewer to more fully appreciate the details in the image. As you process this information, it is used to create a prioritized ranking of objects in the image for attention. The hottest object, the one that has something special, like extra sharp focus, gets your attention: you can use this to start the story of your photo.
The easiest way to explain how the VWM works is to think about how you remember individual objects. There are a limited number of objects that can have full detailed information stored in VWM; once this limit is reached, information about other objects is poorly retained and with reduced accuracy. (Hence the advice to not clutter your images with too much detail) The way this works is discussed more fully in the technical section below. In addition, VWM is one of the central executive functions that coordinate how the subsystems that store and manipulate object information work together. Some people think of the executive functions like they think about the conductor of an orchestra; the executive functions make sure the subsystems know exactly what to do and when to do it. Lets look at a few images with this information in mind.
In the first example image there are three distinct objects: the man using the camera, the train passengers looking out the windows, and the rafters on the river. There are many secondary objects, trees, rocks, etc., that complete the image and set a backdrop for the image but contribute little to the storyline of the photo. This is where Treisman’s prioritized ranking of objects in the image comes into importance. As a photographer you can identify these objects by making their colors less saturated with soft focus. The photo is easy to comprehend. If there had been four objects, some viewers would have found it difficult to grasp, and if more than four it would have been confusing for almost all. Recall that Nmax varies from one person to the next.
But more important is that the story asks a question, what is the picture the fellow with camera takes? This lets you tell a short story. The next photo gives the answer.
Of course, not every photo you take falls into such a simple design, but knowing how your VWM works helps keep you from making a mistake as in the next image: it leaves you asking, “What am I supposed to be looking at?” the rocks in the foreground, the mountains in the back ground or the three sunlit rocks with their distinctive shapes?
Remembering VWM and the crop tool provides a quick solution; notice how your attention is drawn to the morning sunlit rocks, their suggestive shapes.
But sometimes you have a chance for an incredible image that has a lot of VWM ambiguity, like this photo of the two butterflies.
Actually my attention was on getting a photo that shows both sides of the butterflies’ wings, but my VWM came to the rescue.
Please, don’t think that this places anything magic about the number three into making good photographs; it’s just the way your mind works.
The More Technical Side
There is a wonderful review of VWM by S. J. Luck and E. K. Vogel that I have summarized below for the more technical side of this blog.
Visual Signal Processing
So how does the VWM work? First, it probably is not a memory system but seems to be a visual representation system that temporally stores information about an object for short times. It is active for visual search and multiple object tracking. Younger viewers are usually limited to three or four bundles of information, but as you would expect there is also considerable variation across the population. Complex objects reduce the accuracy with which a visual task can be performed. Second, the most successful theory is one where the objects are discreet clearly identified. There are a limited number of objects, Nmax that can have information stored in VWM. When the number of objects exceeds
Nmax, only Nmax have information, the rest have little or no information stored and the viewer is confused, making random guesses to identify what the objects are. Nmax for most people is about 3 or 4. The hypothetical case shown in the above graph is for a
person with Nmax= 3. So unless your goal is confusion, it is best to have three or at most four clearly distinct objects.
Of course, it is not this black and white, there is a transition from being 100% correct to a level that represents how accurately a person is guessing about the objects’ information, as shown here for the above person: note that pure guessing would be at 50%.
Let me borrow a flowchart from my book, The Mind’s Eye Guide to Composition, to remind you of how visual perception occurs in your mind.
It is during these actions that the VWM is actively processing information to determine what objects are in the image. There is a more complete discussion of each stage of processing in my book, but all you need for what follows is contained in the above very simplified flowchart. VWM works to fill in details about the relation between short-term and long-term memory in this over-simplified flowchart. VWM acts as an executive control, a control that if you carefully use it will make, and tell, the story of your photograph. But the catch is the VWM’s limitations. The dominant limit is its capacity and its effect as illustrated above.
Limited capacity visual working memory is one of your mind’s eye guides that is hard wired. To ignore it places stress on your viewer; it is better used to help tell the story of your image using at most three clearly distinct objects. Because your mind naturally rank orders the objects, if you can choose or modify your image in processing, so that the dominate object is established by color, shape, sharpness, etc.to be first seen, you will have made an important start in telling the story of your image.
Because you are wired to do this, it is a good idea to follow your instincts. Of course, the more good experiences you have as a photographer the better your instincts.
© J. JRobert Wayland 2014
For a general overview please see:
B. Wayland (2014), The Mind’s Eye Guide to Composition (Expanded Second Edition), BookBaby.
If you are interested in the full technical details, I would suggest for a general overview of the Working Memory concept:
M. S. Gazzaniga, R. B. Ivry and G.R. Mangum (2009), Cognitive Neuroscience: The Biology of the Mind (Third Ed.), W. W. Norton and Co.
and for the technical literature on VWM,
S. J. Luck and E. K. Vogel, (2013), Visual working memory capacity. Trends in Cognitive Sciences, 17, pgs. 391-400.
A good reference for FIT is
Anne Treisman (1988), Features and Objects: The Fourteenth Barlett Memorial Lecture, The Quarterly Journal of Experimental Psychology, 40A(2), pgs. 201-237.
Bottom-up process: Processing in which stimuli are responsible for visual awareness.
Conjunction searching: the process of finding an object that is not defined by any single unique visual feature, but by a combination of two or more features.
Feature Integration Theory (FIT): a theory developed by Anne Treisman to explain how an image is scanned.
Feature searching: the process of finding an object that differs by unique visual features.
Long-term memory: the retention of information over hours to years.
Nmax: maximum number of objects a person can identify in their visual working memory.
Short-term memory: the storage of information over seconds to minutes.
Top-down process: Processing in which cognitive operations are responsible for visual awareness.
Visual working memory: the limited–capacity store of visual information that allows the performance of mental operations on its contents. See my web site for images VWMFinal.pdf
I am in my second career, after a faculty position at Texas A&M and 26 years at a national research lab, pursuing my dream as a professional photographer. What a wonderful time with digital photography on a runaway train to one exciting development after another. As I race to keep up, it became clear that another new discipline had caught fire, a discipline that all to often is ignore by photographers, but one that has a direct impact on our avocation, namely cognitive neuroscience, i.e., the true basis of psychology. By understanding how our brain evolves into our mind and the impact upon our photography we are on an grand adventure (See ) . The knowledge gained indicates what your mind’s eye wants to see in an image to have good composition.