Reading Comprehension and Rate: One Column vs. Three Columns

Introduction

Background

The widespread use of the Internet and World Wide Web has lead to the introduction of many pages of text filled with information.  Because of the massive amounts of information currently available for users to freely view, screen layouts must be efficient.  The user should be able to scan or read a text document and have a general understanding of the document in the fastest possible time.  The problem is “How can a screen layout be more efficient?” One approach to building a more efficient textual layout of the screen is to examine the columnar format of data presented on pages. This experiment examines the influence screen layout has on reading comprehension and speed by testing a one-column text screen layout against a three-column text layout.

Psychological Theories

In this experiment, we try to determine whether one-column layout or three-column text layout  yields higher reading rate and better reading comprehension. Before we are conducting the experiment, there are several psychological issues needed to take into account with the design of multi-columns screen layout . These issues and theories are as follows:

Short-term Memory Load:  Assuming people read a line of text and put it into their short-term memory before moving onto the next line, a multiple column screen layout would require less short term memory load (more columns on a fixed total width will make each column shorter). Conversely, a single-column layout that spans the length of the screen would require more short-term memory load (length of each line is longer), which can take longer to be processed.

Data Chunking: Another relevant issue is the idea of chunking of data for easier memorization and processing.  With multi-column formats, data can be grouped into small chunks for easier memorization of data. This data can be easier to access because it is in several smaller chunks rather than a single big chunk.

Cognition: Since information from a screen must be recognized and stored in the subjects’ brain, many cognitive processes are working. The increased distance for eye movement in single-column formats and shorter distance for three-column formats can affect the subjects’ cognition of the information. The subject’s perception and recognition of the information in the layout may differ with different screen layouts.

Speed Reading: The tendency of speed-readers to focus on the center of a line, rather than the edges, is an issue that can have dramatic effects on efficiency of different columnar formats.  Since a three-column format has three times as many ends and centers as single column formats, speed-readers would have to scan the document three times more with three-column format.  The theory would suggest that speed-readers would take longer to read a three-column document than a one-column document.

Previous Experiments

Current display technologies allow for greater than 80 characters per line, and it is plausible to apply multi-column format from the printed media to the computer screen.  However, limited amount of research has been done on the relative legibility of one-column and multi-column formats, and the results are inconclusive.  There are advantages to using a narrow column, such as reduced load on short-term memory and ease of transitioning from the end of a line to the beginning of the next line.  There are also advantages for using wide columns, such as faster reading rate and few disruptions from transitioning from one line to the next.

A previous experiment conducted by Hartley et al. showed no significant difference between a single and two-column format on reading comprehensions and reading rate.  However, a  study done by Dyson and Kipping on the effects of one column and three-column text on the reading rate and comprehension has more interesting results.  The reading rate for the fixed one-column text is higher for readers between the age of 18-24, while slightly lower for readers older than 25.  On the other hand, no significant differences in overall comprehension between one-column and three-column text were found. 

In this experiment, we eliminate the scroll bar on both layouts to control the reading rate.  Both layouts will be a fixed size screen filled with text.  

Department of Computer Science: Direct questions and comments to the student editorial team