Working Storage

Th e inference processes build up the contextual overview or situation awareness in working storage.

Th is is not the same as the short-term memory, but short-term memory is an important limit to perfor-mance and is discussed fi rst.

7.2.4.1 Short-Term Memory

Figure 7.25 shows some typical data on how much is retained in short-term memory aft er various time intervals. Memory decays over about 30 s, and is worse if the person has to do another cognitive task before being tested on what the person can remember.

Th is memory decay is important in the design of computer-based display systems in which diff erent display formats are called up in sequence on a screen. Consider that the user has to remember an item from one display, which should be used with an item on a second display. Suppose, the second display format is not familiar, then the person has to search for the second item: Th is search may take about 25 s. Th e fi rst item must then be recalled aft er doing the cognitive processes involved in calling up the second display and searching it.

Th e memory data suggest that the person might have forgotten the fi rst item on 30% of occasions.

Th e practical implication is that, to avoid this source of errors, it is necessary to have suffi cient dis-play area so that all the items used in any given cognitive processing can be disdis-played simultaneously.

Minimizing non-task-related cognitive processes is a general HF/E aim, to increase processing effi ciency.

In this case, it is also necessary to reduce errors. Th is requirement emphasizes the need to identify what display items are used together, in a cognitive task analysis.

7.2.4.2 The Overview in Working Storage

Although there are good reasons to argue that the cognitive processes in complex dynamic tasks build up a contextual overview of the person’s present understanding and plans (Bainbridge 1993a), not much is known about this overview. Th is section makes some points about its capacity, content, and the way items are stored.

Capacity. Bisseret (1970) asked the air-traffi c area controllers, aft er an hour of work, about what they remembered about the aircraft that they had been controlling. Th ree groups of people were tested:

trainee controllers, people who had just completed their training, and people who had worked as con-trollers for several years. Figure 7.26 shows the number of items recalled. Th e experienced controllers could remember on average 33 items. Th is is a much larger fi gure than the 7 ± 2 chunk capacity for static short-term memory (Miller, 1956) or the two items capacity of running memory for arbitrary material (Yntema & Mueser, 1962). Evidently, a person’s memory capacity is improved by doing a meaningful task and by experience. A possible reason for this is given later.

Content. Bisseret also investigated on the items that were remembered. Th e most frequently remem-bered items were fl ight level (33% of items rememremem-bered), position (31%), and time at fi x (14%). Leplat and Bisseret (1965) had previously identifi ed the strategy that the controllers used in confl ict identifi cation

100.0

90.0

80.0

70.0

60.0

0 10 20

Time (s) Record Add Classify

Percentage recalled correctly

30

FIGURE 7.25 Decrease in recall aft er a time interval with diff erent tasks during the retention interval. (From Posner, M.I. and Rossman, E., J. Occup. Accidents, 4, 311, 1965.)

(checking whether aircraft s are at a safe distance apart). Th e frequency with which the items were remembered matched the sequence in which they were thought about: the strategy fi rst compared the aircraft fl ight levels, followed by position, time at fi x, and so on.

Sperandio (1970) studied another aspect (Figure 7.27). He found that more items were remembered about aircraft s involved in confl ict than those that were not. With regard to nonconfl ict aircraft s, more was remembered about the aircraft s that had been in radio contact. With respect to confl ict aircraft s,

35

30

25

20

15

10

5 8

Number of aircraft present at one time

Average number of items recalled

11 Experienced controllers

Trainee controllers Controllers

FIGURE 7.26 Number of items recalled by air-traffi c controllers. (Data from Bisseret, Personal communication;

based on Bisseret, A., Ergonomics, 14, 565, 1971.)

50 50

60

40 40

30 30

20 20

10 10

0 0

1 2 3 4 1 2 3 4

Average number of items recalled per aircraft Nonconflict aircraft

Conflict aircraft

Average number of items recalled per aircraft

Percentage of aircraft in this category Percentage of aircraft in this category

5 6 7 5 6 7

In radio contact Not yet in radio contact

No action Action made Action chosen but not yet made

FIGURE 7.27 Recall of items about aircraft in diff erent categories. (Based on Sperandio, J.C., Charge de tra-vail et mémorization en contrôle d’approche (Report No. IRIA CENA, CO 7009, R24), Institut de Recherche en Informatique et Aeronautique, Paris, France, 1970.)

more was remembered about the aircraft s on which action had been taken, and most was remembered about the aircraft s for which an action had been chosen but not yet implemented.

Th ese results might be explained by two classic memory eff ects. One is the rehearsal or repetition mechanism by which items are maintained in short-term memory. Th e more frequently the item or aircraft has been considered by the controllers when identifying the potential collisions and acting on them, the more likely it is to be remembered. Th e fi ndings about the aircraft s in confl ict could be explained by the recency eff ect, that items that have been rehearsed most recently are more likely to be remembered. Th ese rehearsal and recency mechanisms make good sense as mechanisms for retaining material in real as well as laboratory tasks.

7.2.4.3 The Form in Which Material Is Retained

Th e controllers studied by Bisseret (1970) remembered the aircraft s in pairs or threes: “Th ere are two fl ying towards DIJ, one at level 180, the other below at 160,” “there are two at level 150, one passed DIJ towards BRY several minutes ago, the other should arrive at X at 22,” or “I’ve got one at level 150 which is about to pass RLP and another at level 170 which is about 10 min behind.” Th e aircraft were not remem-bered by their absolute positions, but in relation to each other. Information was also rememremem-bered relative to the future; many of the errors put the aircraft too far ahead. Th ese sorts of data suggest that although rehearsal and recency are important factors, the items are not remembered simply by repeating the raw data, as in short-term memory laboratory experiments. What is remembered is the outcome of working through the strategy for comparing the aircraft s for potential collisions. Th e aircraft s are remembered in terms of the key features that bring them close together—whether they are at the same level, or fl ying toward the same fi x point, and so on.

A second anecdotal piece of evidence is that air-traffi c controllers talk about “losing the picture” as a whole, and not piecemeal. Th is implies that their mental representation of the situation is an integrated structure. It is possible to suggest that experienced controllers remember more, because they have better cognitive skills for recognizing the relations between aircraft , and the integrated structure makes the items easier to remember.

Th e only problem with this integrated structure is that the understanding, predictions, and plans can form a “whole” that is so integrated and self-consistent, that it becomes too strong to be changed.

Subsequently, people may only notice information that is consistent with their expectations, and it may be diffi cult to change the structure of inference if it turns out to be unsuccessful or inappropriate (this rigidity in thinking is called perceptual set).

7.2.4.4 Some Practical Implications

Some points have already been made about the importance of short-term memory in display systems.

Th e interface also needs to be designed to support the person in developing and maintaining an over-view. It is not yet known whether an overview can be obtained directly from an appropriate display, or whether the overview can only be developed by actively understanding and planning the task, with a good display enhancing this processing but not replacing it. It is important in display systems, in which all the data needed for the whole task are not displayed at the same time, to ensure that there is a perma-nent overview display and that it is clear how the other possible displays are related to it.

Both control automation (replacing the human controller) and cognitive automation (replacing the human planner, diagnoser, and decision maker) can cause problems with the person’s overview.

A person who is expected to take over manual operation or decision making will only be able to make informed decisions about what to do aft er the person has built up an overview of what is happening. Th is may take 15–30 min to develop. Th e system design should allow for this sort of delay before a person can take over eff ectively (Bainbridge, 1983). Also, the data mentioned earlier show that a person’s ability to develop a wide overview depends on experience. Th is indicates that, to be able to take over eff ectively from an automated system, the person needs to practice building up this overview. Th erefore, practice opportunities should be allowed in the allocation of functions between computer and person, or in other aspects of the system design such as refresher training.