Back...

Literature Review Part II: Working Memory

In order for end users to effectively transfer visual information, researchers have claimed that they need to build schemata in long-term memory. However, some

information can be transferred and processed automatically to long-term memory, while

some information cannot. Studies argued that not all novel information can be processed

automatically due to the limited capacity in the working memory. The following section of this blog will describes how other studies have defined functionality of working memory and how the limited working memory capacity affects one’s ability to construct meaningful schemata in the long-term memory.

What is Working Memory?

All new information goes through working memory before it reaches long-term

memory. Even though not all information can be processed at the same time, one can

describe working memory as a placeholder for information. Baddeley and Hitch

specifically describes working memory as, “The system or systems that involve the

temporary maintenance and manipulation of information, and it supports human thought

processes by providing a linkage between perception, long-term memory and action”

(Baddeley & Hitch, 1974, p. 829). Based on this definition, some researchers questioned

the differences between working memory and short-term memory (Hebb, 1948; Miyake,

Friedman, Rettinger, Shah, & Hegarthy, 2001). Miyake et al. (2001) acknowledged that

short-term memory and working memory are similar to the extent that they both require

temporary storage of information. Nevertheless, they differ in the position that working

memory requires a lot of controlled attention in order to transfer information to long-term memory. Short-term memory might be able to perform working memory functions, but working memory can be impacted by selective attention during the stages of information processing.

To further examine how information is processed and transferred from the

working memory to long-term memory, Baddeley and Hitch introduced three components

of the working memory model in 1974. Their working memory model consists of a

visuospatial sketchpad, central executive and phonological loop system. Even though Baddeley and Hitch revised their model 20 years later with other subcomponents added onto the existing working memory, their first working memory model is still being used as the fundamental guideline for others who have shown interest in the working memory research. The revised models include information regarding working memory and intelligence.

Visuo-Working Memory

Baddeley defined the visuospatial sketchpad from his three-component working

memory model as, “Temporary storage with capacity to maintain and manipulate

visuospatial information” (Baddeley, 2001, p. 88). Philips (1974) defined visual working

memory as a temporary storage for visual imagery and a place where one can hold and

manipulate visuals in his or her mind. Both terminologies could be simultaneously used as it both describes how end user perceives menu, sub menu, and any images from the web interface. Therefore Baddeley and Hitch’s model, visuospatial sketchpad could be identify as  subcomponent of working memory. 

Visual a information is usually accessed through one’s senses (e.g., perception, motor, tactile, and haptic senses). Once information is perceived from an external stimulus, the visuo-working memory goes through a series of stages before transferring the information to long-term memory. First, the organization of perceptual information is made when an external stimulus occurs. In the second stage, the organized information is moved and stored in a “domain specific short-term buffer.” Finally, the central executive processes are initiated. The central executive processes including the maintenance and manipulation of information, react to the content of the incoming information from the short-term buffer (Baddeley, 2001; Leonard, Ibanez, & Giannakopoulos, 2002). However, similar to working memory, the visuo-working memory is severely constrained when the visual information is new and/or too complex. The information could be easily lost when the visuo-working memory tries to interpret these new and complex materials and maximize processing capacity. Therefore the web interface developer try to make an effort to ease the complex process step that end user must go through between visual stimuli and working memory. The organization of information needs to be structured in most efficient manner. By doing so, we can minimize the working memory stroage capactiy and process time.

Visuo-Working Memory Capacity

First, the working memory system includes a storage component with a limited storage capacity, which determines how much information can be presented in the system at one time. Second, the working memory systems include processing component (the central executive in Baddeley’s model), this component may be limited in its processing capacity, which refers to the amount of information that can be processed in a given period of time (Gold, et al., 2003, p. 61).

The visuo-spatial working memory capacity can be described in terms of a fixed number of objects that can be stored or by the total amount of visual information that can be stored (Alvarez & Cavangh, 2005; Baddeley, 2003; Luck & Vogel, 1997). Based on this description of visual working memory capacity, the visual working memory deficit could be caused by both limited storage and processing capacity. Studies indicated that there are approximately FOUR visual objects could be stored in visual working memory for each individual. Once the storage and processing capacity is met or exceeded in visual working memory, one’s task performance tends to decline (Gilhooly, Wynn, Phillips, Logie, & Sala, 2002).