PSYCH 018 Index > Working Forwards vs. Working Backwards

Working Forwards vs. Working Backwards

When you design your “design phase only” experiment, observation, and survey, you have a choice to either work forwards or work backwards. What does that mean?

Working Forwards

When you work forwards, you come up with a general area of interest, see what other people have done, and then use that to come up with an idea for your research. For example, I’m interested in what makes things legible on computer displays. You might ask what this has to do with psychology—it’s part of industrial psychology and ergonomics.

I went to the library database and did a search for legibility computer, making sure to have selected full text and peer reviewed journals. The search came up with 40 hits; only the first two seemed to be relevant:

The second one looked like a good one to read, but I needed more information, so I expanded my search to just legibility, and came up with this title:

Aha! Typography—that has something to do with fonts. I started reading the article, and it was mostly related to artistic theory, but I came upon this sentence: “Although the human organism is eminently capable of reading perpective distorted texts in Real Life, David Small notes that when it comes to screen-based 3D there are significant shortcoming in this regard” (Ayiter, Yazicigil, Çetin, & Türkmen, 2013, p. 210).

OK, that sounded interesting, so I looked in their reference list and found this reference (which I have converted to APA style):

Small, D. (1996). Navigating large bodies of text. IBM Systems Journal, 35(3/4), 514-525.

I went back to the library basic search and entered “small navigating large bodies of text,” but all the database had was an abstract. Time to use a different search engine:

I went to and tried the search. Success! The full text was available there (links in the right-hand column are links to the full text). The article didn’t look promising until page 9, when Small says:

If a change in typeface, such as boldface or italic, is used it can be difficult to see when viewing an entire scene or act. Dynamic highlighting, such as blinking can be effective when the selected text is small and could be swamped by other information; however, it can also render illegible just that informa- tion that one wishes to make visible. To avoid these problems, I used brightness to cue the dialog of a character. The contrast between the selected and unselected text was continuously adjusted to account for changes in scale. (Small, 1996, p. 9)

OK. Now I had an idea for an experiment. I could have text in those three different conditions: boldface, blinking, and brightness, and see which of those truly is most legible. How would I measure legibility? A search at for “measuring legibility” gave me further references that I could use, the best one of which was:

Mustonen, T., Olkko, M, & Häkkinen, J. (2004, April). Examining mobile phone text legibility while walking. Paper presented at CHI 2004, Vienna, Austria.

You get the idea; when you work forwards, you start from a general topic are, see what other people have done, see what appeals to you, then decide on the experiment you want to do.

Working Backwards

When you work backwards, you have the idea for the experiment already in mind, and you must then look for research that is relevant to your topic. For example: I want to find out if people comprehend reading better when listening to no music, listening to music without lyrics, or listening to music with lyrics. My hypothesis is that listening to music with lyrics will be the worst because of interference between the words the people are reading and the words they’re listening to.

Now I had to figure out what area of psychology this is part of—it’s probably cognition—and I had to find relevant research. The one thing I did not do was try to find an experiment exactly like the one I want to do. I would not do a search for “music lyrics comprehension”; it probably wouldn’t turn up anything useful.

Instead, I had to start casting about, trying searches such as “lyrics psychology.” That came up with quite a few interesting articles, but none of them at all related to what I want to do. One of them did sound interesting, though:

Ahmadi, F. (2011). Song lyrics and the alteration of self-image. Nordic Journal Of Music Therapy, 20(3), 225-241. doi:10.1080/08098131.2010.522718

At this point, I could have changed my mind and started exploring that instead (and I would then be working forwards again), but I decided to keep researching my original plan. After a few more searches coming up empty, I finally tried “linguistic interference” and came up with one article that seemed like a winner:

Gautreau, A., Hoen, M., & Meunier, F. (2013). Let’s all speak together! Exploring the masking effects of various languages on spoken word identification in multi-linguistic babble. Plos ONE, 8(6), 1-9. doi:10.1371/journal.pone.0065668

Was this exactly the same as what I wanted to do? No. Was it reasonably well related to what I wanted to do? Yes. That means I could use it as a reference. The references at the end of that article now gave me more possibilities. I also saw the words “masking effect” and used that as a search term in “music masking effect”; that gave me this article:

Shi, L. (2009). Normal-hearing English-as-a-second-language listeners’ recognition of English words in competing signals. International Journal Of Audiology, 48(5), 260-270. doi:10.1080/14992020802607431

You’ll notice at this point I’m working forwards again, and I would eventually come up with enough references to write a solid introduction and form reasonable hypotheses.

Which Way Should You Work?

I have found that working forwards is far easier than working backwards. If I start with a large topic (but not too large; I wouldn’t start with “behavior”!), I feel as if I have more choices available to me, as I am not tied down to a specific goal. I don’t have to freeze my ideas too early.

No matter which way you work, remember that any article that is reasonably closely connected to your research is usable. Even if the article you are reading is about a survey, you can still use its results to build up the background information on the topic and the hypotheses that you will make in your introduction.