The Four Phases of Prospective Remembering An Empirical Paradigm

Most studies on time- or event-based prospective memory have employed prospective memory tasks that consist of a single, isolated act that has to be remembered and performed at appropriate points in the experiment, for example, to press a designated key on the computer keyboard whenever a prospective memory cue appears or at a specific time (e.g., Einstein & McDaniel, 1990; Einstein, McDaniel, Richardson, Guynn, & Cunfer, 1995; Einstein, Smith, McDaniel, & Shaw, 1997; Marsh & Hicks, 1998; Maylor, 1996; Park, Hertzog, Kidder, Morrell, & Mayhorn, 1997; West & Craik, 1999). Because these paradigms comprise of a simple (although often repeated) intended action, they might not completely capture the phases of prospective remembering already described as they potentially obviate the need

FIGURE 9.1 Decomposing the process of prospective remembering (after Kliegel et al., 2002).

FIGURE 9.1 Decomposing the process of prospective remembering (after Kliegel et al., 2002).

to explicitly plan delayed actions and minimize the retrospective memory load of intention content during the delay phase. Thus, most of the paradigms applied in the literature focus on prospective memory initiation only (Ellis & Kvavilashvili, 2000).

To be able to study all four phases of prospective memory and their interplay, as well as their development, we have proposed a prospective memory paradigm that provides empirical scores for performance in all four phases by applying a modified version of the six-elements task (SET; see Kliegel, Eschen, & Thone-Otto, 2004; Kliegel, Martin, & Moor, 2003; Kliegel, McDaniel, & Einstein, 2000; see also Kliegel et al., 2002; Kliegel, Phillips, Lemke, & Kopp, 2005). The SET was initially proposed by Shallice and Burgess (1991) to assess multitasking performance in neurological patients. To perform the SET, participants have to remember to self-initiate six different, open-ended tasks in a limited time period. Therefore, they have to schedule the tasks efficiently and keep track of time (see also Burgess, Veitch, Costello, & Shallice, 2000; Groot, Wilson, Evans, & Watson, 2002).

Materials We kept the basic format of the original SET. In our modified SET, participants must remember to perform six tasks that are subdivided into two parallel versions of three types of tasks: words, arithmetic, and pictures. In all tasks, stimuli are presented on a sheet of paper and participants can move down the sheet from one item to the next to continue working on the task. In the word finding tasks (Parts A and B), participants must score out a word from a list of several possibilities (e.g., nonword distracters: haz-haif-house-hool). In the arithmetic tasks (Parts A and B), participants write down the answers to sets of arithmetic problems of varying difficulty. In the picture tasks (Parts A and B), participants must write down the names of line drawings of everyday objects. Total performance time is limited to 6 minutes. Two rules govern performance: (a) All tasks must be attempted, and (b) participants cannot attempt both parallel versions of the same task type sequentially, hence restricting the order in which tasks can be attempted.

Procedure We extended the original SET instructions to require participants to both explicitly state a verbal performance plan that they intend to follow when working on this multitask set and to delay the self-initiated execution of this plan (Kliegel et al., 2000). A score is generated for each component of the modified SET so that performance at each phase of prospective remembering can be evaluated (Kliegel et al., 2000).

1. Intention formation component. First participants plan their later performance. This verbal plan is recorded on a cassette tape and transcripts are scored according to how elaborate the plan is.1

2. Intention retention component. Second, participants have to retain their plan over a delay period. This delay is filled by heavily engaging activities

1 This plan is scored as follows: It is noted whether the participant makes an explicit plan or not. If a plan is made, does the participant include the intention to attempt all tasks? Furthermore, does the participant plan to adhere to the rule that constrains the order in which tasks can be performed? Finally, the number of executable substeps is analyzed.

(e.g., IQ subtests) that prevent participants from continuously rehearsing the prospective memory task in working memory. Moreover, intention retention itself is examined, as participants are required to recall their plan during this delay period. When compared to their original plan, this gives a measure of how accurately participants are able to retrospectively recall their intention content.2

3. Intention initiation component. At the end of this delay period, the paradigm requires that participants self-initiate performance of the multitask set on presentation of an appropriate cue.3

4. Intention execution component. When the multitask set is finally performed, participants' performance and adherence to task rules are scored.4

Importantly for the present purposes, these modifications allow us to disentangle the intention formation or planning phase of a prospective memory task from the execution phase, because intention formation itself is explicitly assessed and evaluated in the form of a verbal plan. Similarly, the intention initiation phase can be scored independently of the intention execution phase during which the multitask set is performed. As an additional point of interest, plan fidelity, the extent to which participants execute their original plan, can also be measured.

This modified SET was initially applied as a paper-and-pencil test to investigate complex prospective memory development throughout adulthood (Kliegel et al., 2002; Kliegel, Martin, & Moor, 2003; Kliegel et al., 2000; Martin, Kliegel, & McDaniel, 2003) and in clinical populations (Kliegel et al., 2004, Kliegel et al.,

2005). It has also been implemented as a computerized version, the Heidelberger Exekutivfunktionsdiagnostikum (HEXE; Kliegel & Martin, 2002), which has so far been used to examine the development of complex prospective memory in typically (Martin & Kliegel, 2003) and atypically (Kliegel, Ropeter, & Mackinlay,

2006) developing children. In the next section, we review the results obtained in these cross-sectional studies and examine all four stages of prospective memory at distinct phases of the human life span.

At this point we should comment that our complex prospective memory task enables explicit differentiation of the four phases of prospective remembering at the cost of being a fairly resource-demanding task. In their multiprocess framework, McDaniel and Einstein (2000; see also Einstein et al., 2005) argued that

2 This recalled plan is scored using the same elements as the initial plan generated in the intention formation phase. The two plans are then compared and points are awarded for correct recall of the original plan (i.e., matching elements).

3 The cue to begin the multitask set is when participants are asked to write their date of birth at the top of a questionnaire given to them by the experimenter. If participants fail to act on presentation of this cue, the experimenter reminds participants to begin performing the multitask set and a failure of self-initiation is recorded.

4 Multitask performance is scored according to the following elements. The number of tasks attempted within the given time limit and whether the task order rule is adhered to or broken. Additional scores of interest include whether participants adopt an optimal performance sequence by beginning with a single item from each task and the accuracy of their responses to items in each task.

the need for planning in prospective memory tasks increases the probability of strategic processing as the underlying mechanism of prospective retrieval. Moreover, in the execution phase of our task, participants have to keep track of time and schedule six different tasks, which according to the multiprocess view will result in conscious monitoring. Thus, we acknowledge that the task applied in the research and theoretical reasoning presented in this chapter represents only one—albeit fairly common—prospective memory scenario.5 Other, more automatically triggered prospective memory tasks are likely to show distinct developmental patterns (see McDaniel et al., chap. 7, this volume).

+1 0

Post a comment