Effect of words highlighting on reading comprehension

Question

I'm interested if there are studies dealing with text understanding and POS (part of speech) coloring, or coloring syntactic/semantic information. The studies should solve the questions like:

• Which combination of colors improve/deteriorate understanding and reading of texts?

• How many colors should be in the scheme for positive/negative effects on reading and understanding of text? Which POSs should be preferably colored?

• Are there types of people who are more efficient with/without coloring?

I believe that a few colors with good combination could be beneficial. Has somebody conducted research to test the effect of highlighting on reading comprehension?

---

An example:

On this text you can see that many colors that don't go together can make the reading process even worse and distracting. Can this be changed to improve the reading process, instead?

A few events inspired me in the following order:

1. I started learning Chinese and their characters seem to have positive effect on my comprehension.
2. I've read a NL (natural language) article in English in a programming editor. I was used to colors for various programming languages.
3. I searched the Internet and found the article A case against syntax highlighting but don't agree with all the author's arguments or opinions. When you have only a few carefully chosen colors it's beneficial for programming so maybe could be for NL comprehension as well. A few colors could underline the basic thought and help to draw attention of the reader. When I read a scientific text for which I have not passion I'm often bored because don't know where it directs and colors and images help me a lot to concentrate and skimming over the basic thoughts could help with decision if it has sense to read it at all. This does not hold for fiction and poetry where we read for pleasure rather than getting information.

Answer 1

I can speak to this question somewhat from a cognitive psychology standpoint. We memory researchers would think of text highlighting like this in terms of distinctiveness. (An article by McDaniel and Bugg (2008) may shed some light.) Simply put, highlighting a word in a different color than the rest of the text draws what we call item-specific processing towards that word. You'll have a better memory for it, for sure -- and maybe this ties into goals more relevant when writing code.

Increasing item-specific processing, however, comes with a cost: It detracts from organizational processing, which is involved with interrelating sentence (or code) elements, memory for order, and so forth.

Also note that the more items that are made distinctive, the less of a benefit making an additional item distinctive provides (e.g., if words are multicolored already, making a new word red or green doesn't really do anything to the distinctiveness of that word).

So when you use a lot of different highlighting colors, like in the example image, two things are happening: (1) too much distinctiveness is being used, washing out benefits from distinctiveness; and (2) increasing item-specific processing for individual items damages memory and processing for the structure. In essence, the example given is the worst of both worlds.

To answer your question, it might make sense to only use highlighting (or colors, fonts, etc.) to highlight key points, etc.; much like what we already do in writing for the web. Like anyone who highlights a textbook knows, it doesn't do any good to highlight an entire chapter.

Answer 2

According to this study, layout and color can be used to improve text readability. You can take a look online at text formatted in this manner here.