In the field of cognitive ergonomics, it is generally acknowledged that individual differences play a role in the performance of information processing tasks. Studies by others indicated that some individuals prefer to encode information in a pictorial format while others appeared to prefer a linguistic format, regardless of whether the stimulus was presented as a word or a picture. Such earlier research tended to rely on the ranking of reaction time patterns for one variation of a linguistic-pictorial task as a means of identifying and grouping subjects according to encoding strategies. Problems were apparent in this method of classifying subjects.

The present work consists of fundamental research focusing on one parameter which the researcher refers to as "Cognitive Encoding Style" (CES), and which is defined as "the habitual or preferred format (i.e., linguistic or pictorial) by which one encodes and manipulates information in working memory." The purpose of this research was to attempt to verify that CES is a distinct dimension of cognitive style displaying the following characteristics:​ the two poles of the dimension (which are labelled "pictorial encoders" and "linguistic encoders") differ in terms of strategy usage and performance on the criterion task;​ individuals are consistent in their use of those cognitive strategies associated with a particular polarity;​ the criterion task is an indication of strategy selection and not subject ability;​ and the dimension is distinct from similar dimensions of "cognitive style".

In order to test the use of self reports as a means of identifying and classifying subjects according to their CES and to verify CES as a dimension of cognitive style, a series of three experiments was designed and carried out using variations of a linguistic-pictorial verification task as well as self reports. The results confirm CES as a dimension of cognitive style conforming to the stated characteristics. The results also confirm that, for purposes of CES research, self reports have merit as a method of classifying subjects according to CES.

Some implications of this research for information display design are discussed and suggestions made for further study in this area of cognitive ergonomics