Friday, October 26, 2018

Satisficing: A Skimming Technique for Identifying Important Information Quickly

Duggan, G. B., & Payne, S. J. (2009). Text Skimming: The Process and Effectiveness of Foraging Through Text Under Time Pressure. Journal of Experimental Psychology, 15(3), 228-242.


The purpose of the study at hand is to determine whether readers that skim through longer texts effectively allocate their limited time for identifying important information. Of the different skimming techniques that have been developed, Duggan and Payne identified satisficing as the skimming strategy of interest for their study. Satisficing is an information-foraging technique in which a reader moves on to the next paragraph or section once the reader perceives him or herself to have gathered sufficient information. 

Duggan and Payne conducted a study to examine the effectiveness of stratisficing as a skimming technique to find important information in a long text. The experiment examined two groups: the skim condition and the half-text condition. The skim condition group was told that they won't have enough time to read all of the text and to allocate and prioritize time accordingly. The half-text condition group was instructed to read at a normal pace, however one half of the text was missing (the first or second half). The research design allows for studying of two types of readers: those that skim to finish a full document and those who read at a normal pace and subsequently only get through half of the document. 

The authors hypothesized that memory for meaning would be greater when skimming compared to those who read half the text at a normal pace. The results supported the hypothesis that readers correctly recognized more of the important sentences after skimming than reading half texts and retained more information about the structure of the text. However, it is important to note that skimming did not improve understanding of unimportant sentences or inferences about the information. 

Duggan and Payne go on to suggest that readers under time pressures can achieve improved understanding of the main elements of the text by satisficing. The satisficing technique is particularly effective under time constraints because it mediates the common problem that "pages and documents suffer from diminishing returns in terms of rate of gain of valuable information". The phenomenon of diminishing returns occurs because there is a degree of redundancy of information in longer texts and writers typically organize important information by separating them into different paragraphs and sections. Thus, the longer a reader remains in a single paragraph or section, the more redundant and less valuable the single paragraph or section generally becomes. In this way, the authors support the use of satisficing when reading under time constraints to find the most important information. However, the authors do point out that satisficing does not aid memory of less important details and does not facilitate the inferences about information from the text. 


The authors of this article make a compelling argument for the use of satisficing as a skimming technique to find important information in a long text. If under time constraints as an analysts, this can certainly be a useful technique for finding the most important information in the fastest way possible. However, my main concern with the use of satisficing lies with the authors' findings that satisficing did not facilitate the inferences about information from the text. In other words, this means that readers can reliably identify what is important, but why the information is important and how it fits into the bigger context remains out of the readers' comprehension. The bigger picture and the inferences about the information is arguably the most important part of the analyst's job and without good inference the quality of analysis suffers. As a result, I believe analysts are justified in using satisficing under time constraints, but should appropriately account for satisficing by reducing analytic confidence in their estimates. 


  1. Tom, I agree with your critique and share your concern about satisficing analysts and their ability to make inferences. How did the half-text group fair in that category? I ask because if the results show analysts making weak inferences in both groups, then we might as well miss quickly if we're going to miss either way.

    1. Hi Jillian,
      The ability of the half-text group to make inferences was also adversely affected. Without having access to the full text, even at a short glance, the readers could not get a grasp of the bigger picture in the text.

  2. Tom, your critique was right on point. There may be pieces of information that an analyst can skip over that may be extremely relevant to cases, draw connections to other information, lead to a conclusion, or give direction.

  3. Tom interesting article! I agree with both Jillian and Chelsie about your critique. You talked about groups reading the text at a normal pace, but reading either the top half or the bottom half. Was there a difference in the results for those that read the top half verses those that read the bottom? To me it seems like the bottom half would be missing the context leading into it. Also, when skimming I often spend less time on the bottom part.

  4. Hi Alyssa,
    Great question but the authors did not examine how the top half group's performance compared to the bottom half group. The key takeaway from the results is that missing either half of the text leaves out important contextual information.