With the use of technology continually rising, it is definitely no surprise that the number of Internet users, already millions, continues increasing. Though the Internet is useful for many things, it is my belief that the most beneficial task that the Internet can be used for is to do research. There are many search engines online that can help users find almost anything they want to know. In a survey that was done in 2004 (see Figure 1), Google, Yahoo, MSN, AOL, and Ask Jeeves were the top five search engines on the Internet, with Google being number one (Sharma, 2004).

Figure 1 (29771)

Figure 1: CNET News: Google ranks number one on search survey.

What is it that makes Google the number one choice of search engine, though? Surely, there is something that makes Google unique among other search engines, a reason why is it the most popular for Internet users. Even a search engine so preferred, however, can and does have flaws in its design and much room for improvement.

System Description:

When an Internet user wants to find some type of information, all the person has to do is go to the Google homepage, Once the Google page has loaded, the user's eyes are immediately directed to the colorful "Google" logo and, therefore, to the search box directly below it (see Figure 2). At this point, there are a few options to what can be done next, which will depend on what exactly is being searched for.

Figure 2 (29773)

Figure 2: Google homepage

Let's say that we have one user who is going to use Google to find some general information on global warming. Once the person has gone to, they simply have to click in the search box below the Google logo, type in "global warming," and either press the enter key on the keyboard or click the "Google Search" button that is seen in Figure 2. The search usually takes less than one second to complete (0.13 seconds to search global warming), which is very convenient because most Internet users just want to get their information as fast as possible. Once the search has been completed, all the websites that have information about global warming are listed in order, beginning with the closest match to the keywords searched (Figure 3).

Figure 3 (29774)

Figure 3: Global warming search results and sponsored links

There is also what Google calls "Sponsored Links" listed down the right side of the screen and the most popular sponsored links are at the very top of the actual results listed and have a light pink background, making those target links easier to detect (see Figure 3). Each listing has a title, which you click on to go to that site, and a short description of the information on the website, so as to make it easier to find the exact information sought after. However, not all the websites about global warming can be seen at once, making it necessary to increase search time by having to scroll up and down. Also, there can only be ten results listed on a single page, which means that if the first ten results do not have what is being sought for, the user must click on the "Next" button on the bottom of the Google results page or click on any of the numbers to the left of the next button to browse through more results (see Figure 4). With so many results listed, the user can feel overwhelmed and frustrated to have to sort through all the material. However, the user can select a related search to global warming, which Google nicely lists at the bottom of the results page, by clicking on it (see Figure 4).

Figure 4

Figure 4: Related search options and navigating to more pages of results

Now, let's assume that someone needs to do research for a class assignment on amnesia, but the teacher has specific instructions to use only scholarly sources. Searching for this in the same way as global warming would be problematic because, not only would the person have to search for the right information, but they would also have to go through and make sure that it is a scholarly source. There is an easier way to do this on Google, though, as long as the user is trained well enough to know how. In the upper left-hand corner on the Google homepage, as seen in figure 2, there are different options that a user can click on to find results in those categories, such as images or news. By clicking on the "more" option, a drop-down list appears with further category choices. From here, the user can click on "scholar" in order to search through scholarly sources only (see Figure 5).

Figure 5

Figure 5: Search for scholarly sources by clicking on "more" and then on "scholar"

Now on the Google Scholar homepage, typing amnesia into the search box and clicking search will result in only relevant scholarly sources being shown. The results page for Google Scholar is designed slightly differently than the regular Google results pages. Firstly, there are no sponsored links on the right side of the page and instead of sponsored links at the top of the search results page, there is an option to see "All Articles" or just "Recent Articles," so that a user can get only the newest information on the topic if desired. Also, there is no related search options listed at the bottom of the screen, which has been replaced by "Key authors." Still, however, the user has to scroll up and down and click through numerous pages to see all of the results.

System Critique/Information Processing Constructs:

The interactions between humans and systems can be analyzed using many different information-processing constructs. In interacting with Google, there are certain constructs that are much more relevant than others that can help explain how fast and easily a person can use the system. A user's navigational ease and accuracy through can be affected by signal detection and the vigilance decrement, attention and visual search, and egomotion.

When performing a search on Google, it is important to be able to find the right information (the signal) easily. Although the results page shows up in the exact same way for each person, an individual's sensitivity to be able to detect the signal and their biases influencing their decision, or response criterion, factor in to the ease of signal detection. The state of the user's sensory system and internal noise (neuronal firing) greatly affect their detection sensitivity and their motivations and expectancies also come into play with biases in response criterion (Wickens & Hollands, 2000). It can still be difficult to detect the correct signal, even without the individual differences being considered, though, because of the problem of scrolling up and down the results page, as well as having to click through pages just to see all the options. Searching through the millions of results for an extended period of time can cause a person to become cognitively fatigued and, therefore, susceptible to the vigilance decrement. The vigilance level decreases due to sustained attention and, when this happens, sensitivity and the response criterion are also negatively affected, causing the ability to detect the signal to become much more difficult (Wickens & Hollands, 2000).

As pointed out by the vigilance decrement, attention also has a large role in navigation of Google. With human memory limited in its capacity, it sometimes becomes necessary to limit attention too. The information that Google can help a user find seems infinite and, by using selective attention in the search process, the user can focus on the immediate task at hand. When viewing a Google results page, such as shown in Figure 3, attention is automatically directed to the sponsored links' box with the light pinkish background. This is a good example of a "pop-out" in the search process, which is beneficial in a parallel search. Although a parallel search is more efficient because fewer scans are necessary than in a serial search, it can only be helpful if that is the target information that is being searched for and only "when the target can be defined using a simple rule" (Wickens & Hollands, 2000, p. 79). Going through Google results for some information is definitely not that simple because many of the results options can be very similar, causing the target to blend in. Because of this, it is necessary to perform a serial and self-terminating search, in which "each item is inspected in turn" and "the search stops when the target item is found" (Wickens & Hollands, 2000, p. 79). Doing a serial search through Google results can be problematic also, though, because of the fact that there are so many results to search through and that some of the descriptions under the titles of each result are not very informative.

The speed and ease in which a user can navigate through a visual search strongly depends on how the information is presented and dispersed across the visual field. The search can be performed most efficiently if the visual display presented is compatible with the user's egomotion, meaning that the display should "capitalize on the visual cues people naturally use to perceive their motion through the environment" (Wickens & Hollands, 2000, p. 159). Perhaps the most important type of a user's egomotion for the results display to be compatible with is the optical flow. When searching through Google results, the user's speed is directly affected by the optical flow of the page. In the serial search of scrolling down one page, Google's optical flow has few obstacles and is quite good. However, the optical flow gets ruined when the user reaches the tenth result and has to then click to a new page to see more results, causing the search time to greatly increase. Another problem with timing on Google is seen when a user clicks on the title of a desired result and there is a delay in the response. "When we act upon objects in the real world, there is typically very little delay from the time the action is initiated until motion occurs" (Wickens & Hollands, 2000, p. 172). In order to keep this virtual world compatible with human egomotion, there needs to be a closed-loop interaction and very little delay in response to control inputs.

System Improvement Suggestions:

Even though Google rates number one on a search survey, as shown in Figure 1, there are obviously some flaws in its design and, therefore, some improvements that need to be made (Sharma, 2004). As we have seen, vigilance seems to be an issue when searching through Google results because there are many options to choose from. In order to help improve vigilance, it becomes necessary to improve individual sensitivity and response criterion. To make people more sensitive to the target being searched for, it would be helpful to improve the salience of the target (Wickens & Hollands, 2000). For instance, when searching for global warming, the salience can be increased by having "global warming" highlighted or colored in the results so that the user's attention is directed towards those words, making it easier to locate. For some older adults with their sensory systems in a lower state, making the font size larger and/or bold so that these individuals can see the target more easily can help signal detection. Also, training user's to search on Google would be effective so that they can know what they are looking for, which helps improve response criterion.

Since it seems difficult to do a parallel search on Google, unless the user really wants the sponsored links, serial searching is necessary, but can take a great amount of time and effort. A user wants to get their search done as quickly as possible and one way to help shorten the search time would be to have better descriptions saying exactly what type of information that result has under every title option. Doing this would make the serial search faster and easier. Having the user practice searching on Google would also be beneficial because, since Google's mapping is consistent, automaticity can be developed and the person could switch to doing a parallel search instead of a serial search (Wickens & Hollands, 2000). Being able to do a parallel search instead would greatly reduce the search time and effort also.

When searching through a visual field, like Google, navigation can be disrupted when the optical flow goes against the user's egomotion. With only ten results listed per page, Google's flow gets interrupted by having to click through more pages. This makes the search take much more time to be completed. To make the flow of the page continuous, it would be optimal to have all the results listed on a single page. However, each individual's computer system and Internet provider can affect the speed of the Internet, so having all the results on one page could make their search even more difficult. With so many different computer systems and types of Internet connections, Google should put the option of having as many results per page as the user would want. This can be accomplished, for example, by having a "Results per page" option listed in the very top right corner of the results page with options of 10, 50, 100, etc. That way a person can search with the greatest efficiency that their computer system and Internet connection will allow. Also, with this as an option, a user can make sure that they have a closed-loop interaction with Google by choosing the number of results per page that will make for the smallest amount of delay possible in response to control inputs.

Although Google is the number one search engine online, there are some problems that affect signal detection. Critiquing the system to account for the vigilance decrement, serial searching, and optical flow, as well as for individual differences would make Google even better for users. By using highlighting and larger font, improving site descriptions, putting options for number of results per page, and training users, the searching experience on Google can be greatly improved, making for a faster, easier, and more efficient search process.


Sharma, Dinesh C. (2004). Google tops the search charts. CBS Interactive Inc. 2008. CNET News.

Wickens, C. D., & Hollands, J.G. (2000). Engineering Psychology and Human Performance (3rdEdition). Upper Saddle River: Prentice Hall.