The definitions and distinctions already outlined provided an important starting point for the PAM theory (Smith, 2003; Smith & Bayen, 2004, 2005, 2006; Smith, McVay & McConnell, Hunt, 2007). These definitions lead to an important role for consciousness in the performance of delayed intentions. An individual must be conscious of the plan to perform an action (in other words, the plan is in the focus of attention) at both encoding, when the plan is formed, and at retrieval, when the action is performed. The endorsement of the practitional copula is a conscious activity. Furthermore, this endorsement occurs along with a conscious recollection of the prior plan of action.
In addition, intents are volitional and volition is a decision. Thus, an intention is a decision about action that consciously references a prior plan. The importance of the decision making rests in a key aspect of prospective memory tasks: The agent must often remember to perform the action when other activity is underway (Einstein & McDaniel, 1996). Refer again to the characteristic that distinguishes delayed intentions and immediate intentions. Immediate intentions are performed as the intention is formed. In contrast, delayed intentions, by definition, require that the plan of action leaves the focus of attention for some amount of time between the decision to perform the action and the time at which the action is to be performed. In this interim, ongoing activity will occupy the focus of attention.
How is it that the intent is returned to focal attention? One might argue that the return to focal attention of the intent could be achieved through automatic processes (e.g., Guynn, McDaniel, & Einstein, 2001; McDaniel & Einstein, 2000). These automatic processes could take several different forms. For instance, the "associative linkage between target and intended action" could lead to "activation of the intended action upon encounter of the target event, with such activation presumably being reflexive and nonstrategic" (McDaniel & Einstein, 2000, p. S130). McDaniel and Einstein also argued that an environmental event can lead to "spontaneous and relatively resource-free retrieval of the action" (McDaniel & Einstein, 2000, p. S131). In both the automatic associative activation case and the spontaneous and resource-free retrieval of the action case the prospective memory task is performed automatically and a decision process would not be required. Careful application of the definitions outlined earlier would exclude these two cases of automatic performance as reflexive action that is not subject to conscious control and is therefore not an intention. Alternatively, one would require different but equally precise definitions of intentions to include these cases.
McDaniel and Einstein (2000) also suggested that there are cases in which an environmental event spontaneously captures attention (e.g., because the target is salient) due to a noticing process, perhaps because of a differential ease of processing for the target event. The attention capture is followed by a "controlled search for memory" (p. S131), as would be the noticing of a target. In these cases the initial orienting to the target event may be automatic, but this is followed by a controlled decision process about how to respond to the stimulus. In this alternative situation an individual must devote capacity to active decision making (McDaniel & Einstein, 2000). Similarly, in this sort of situation PAM would predict that the attention capture is followed by preparatory attentional processing, with the preparatory attentional processing likely in the focus of attention.
When prospective memory tasks involve targets that are not going to spontaneously capture attention, which thereby force active decision making, there must be other means for preparing to respond to a target. To perceive the target as such, the individual must be prepared in some way for the potential occurrence of the target (Smith, 2003). The PAM theory argues that some of our limited cognitive resources must be devoted to evaluating the environment and responses to the environment for an event to be recognized as a target event. This particular point arises from the basic assumption of transfer appropriate processing (Morris, Bransford, & Franks, 1977). The intended action plan will return to the focus of attention through reinstatement of the processes that occurred at encoding. For the same processes to be reinstated, the retrieval context must sufficiently match the encoding context.
With respect to the external context, or environment, there will always be a difference in time and often in other factors. For example, if you plan in the morning at home to stop and pick up dry cleaning after work, the location at encoding and at retrieval will be different. What might match are the internal context of retrieval and the internal context that was part of the plan. The same external information can be viewed in very different ways depending on an individual's own information and experience. This is perhaps one reason that laboratory prospective memory tasks do not always suffer from ceiling effects, despite very good physical matches between encoding and retrieval contexts. The internal context has been changed sufficiently by the ongoing task to reduce the chances of reinstating the processing necessary for completion of the prospective memory task.
Reinstatement of the processing that will lead to the performance of a delayed intention depends on the match between the external or internal context at retrieval and the internal context at encoding. This requires that the agent's internal context be ready for the reinstatement to occur. If the agent is devoting all available resource capacity to some ongoing activity, the external environment will only match the context that existed at encoding if the ongoing activity sets the same context. Given that the intent of the background task is by definition not the same as the delayed intention, the match will always be less than ideal. This problem can be overcome by devoting some amount of capacity to evaluating the environment and the appropriateness of your responses to events in the environment. (The importance of the relationship between the external context at retrieval and the internal expectations about that context are highlighted nicely in a recent study by Marsh, Hicks, & Cook, 2006; see also Cook, Marsh, & Hicks, 2005; Nowinski & Dismukes, 2005.)
As alluded to earlier, there are cases in which preparatory attentional processes need be engaged for only a short time prior to performing the delayed intention. For instance, if one sets a loud alarm as a reminder, the alarm, if sufficiently obnoxious, may intrude into one's focus of attention and demand consideration. This would lead to an evaluation of how one is to respond to the stimulus or what the meaning of the stimulus is; that is, preparatory attentional processes are engaged. It is still the case that the task involves resource-demanding preparatory processes; however, in this case they would be initiated in response to an attention-capturing event in the environment.
Figure 2.2 presents a depiction of how a delayed intention is performed according to the PAM theory. First an individual decides to perform an action. This is the formation of an intention,2 shown in Figure 2.2 as the point at which the individual decides to perform a particular action, "I will o^." The action could occur immediately, "I a/' (an immediate intention, arrow 1), or it could be delayed, in other words, "I engage in some other action, a2." In the case of a delayed intention in which we engage in some ongoing activity after forming the intent to perform a1, we would have a different intent, "I a2," and the planned action a1 leaves the focus of attention (arrow 2).
2 It should be noted that until an action is carried out, there is no intention in a strict sense. An intention is a kind of intent and intents do not exist in the absence of action because they are the cause of action. Thus, until an intention is acted out we cannot know for sure that it is an intention. However, I use the term intention with the understanding that at encoding this represents something that could be an intention.
FIGURE 2.2 Diagram of the possible stages in the performance, or failed performance, of an intention. The process begins with the formation of an intention to perform action a1. Following arrow 1 illustrates the case of an immediate intention. The action is carried out with the plan leaving consciousness. Following arrow 2 illustrates the case of a delayed intention. The intention leaves consciousness and capacity is devoted to the ongoing task a2. If insufficient capacity is devoted to preparatory attentional processing (arrow 3), or if retrospective recognition of the targets or retrospective recall of the action fails (arrow 9), then the individual will move on to some other action a4. Selecting either arrow 5 or 8 will lead to performance of the delayed intention. Following arrow 5, once the delayed intention is performed, the individual could continue with action a3 if the ongoing task is incomplete (arrow 7). If there are multiple target events to which the individual needs to respond with action a1, the individual could continue with preparatory attentional processing (arrow 6). Following arrow 8, the individual stops performing the ongoing task and devotes capacity only to deciding what to do.
For the intention to be executed, I must devote some amount of my limited cognitive resources to making decisions about how to respond to my environment, at the point at which the intended action can be carried out. In other words, I must be prepared for the possibility that I will make a decision to change from an ongoing action to some other action. This does not mean that considering the change to a specific action is the focus of attention. Rather, this simply means that I am prepared for the possibility in a general sense represented by aA. The processes involved in considering the possibility of a change in a general sense, aA, are referred to as preparatory attentional processes. Although these preparatory attentional processes need not be the constant focus of attention, they do require resources and they prepare the individual for the return of the intended action to the focus of attention.
The PAM theory proposes that preparatory attentional processes are required during the performance interval, but does not require constant engagement of these processes prior to the performance interval. That is, if we must remember to stop at the store on the way home, we must engage in preparatory processing during the interval when the opportunity to stop at the store is possible, but we need not engage in preparatory attentional processing throughout the entire day.
What might lead to the initiation of preparatory processing? It is possible that through developmental processes we learn to reevaluate our behavior and potential responses to the environment at various points during our daily lives. For instance, when finishing one task, but before starting the next, we might consider whether we can focus only on that task or must also accomplish something else, with the latter leading to the engagement of preparatory attentional processes. Event-based prospective memory tasks that can be accomplished at such switch points, called combination time- and activity-based prospective memory tasks by Kvavilashvili and Ellis (1996), may have an advantage. These tasks must be performed at natural points for initiation of preparatory attentional processes, and these processes can be engaged without also devoting resources to an ongoing task. Kvavilashvili and Ellis suggested that the fact that these combination time- and activity-based prospective memory tasks can be performed without interrupting an ongoing activity may contribute to an advantage in performance success rates for these sorts of tasks.
Referring again to Figure 2.2, assume that the performance interval or opportunity for a1 is associated with an ongoing activity a3. The individual must engage preparatory attentional processes during the performance interval to have successful prospective memory performance. The individual need not engage in preparatory attentional processing outside of the performance interval, in this case, during the performance of a2. This would be similar to saying to participants that they should remember to press the Enter key if they see the word dog during a lexical decision task, but then having them perform a filler task of math problems between the time they get the instructions and the time they perform the lexical decision task. Participants may or may not also engage in preparatory attentional processing during the math problem task. PAM theory proposes that they must be engaging in preparatory attentional processing when the target occurs during the lexical decision task to be successful at this task.
It is possible that the individual does not engage in preparatory processing during the performance interval and fails to perform the delayed intention (arrow 3). Alternatively, as suggested earlier, participants may routinely evaluate potential changes in activity or the need to be prepared for such changes when switch points occur in the environment. This would lead to the initiating of preparatory attentional processing while engaging in the action a3 associated with the performance opportunity for the intended action a1 (arrow 4). If the target occurs and is correctly identified and the action correctly recalled, then the individual can perform the delayed intention (arrow 5). Execution of the delayed intention could be followed by additional preparatory processing, in the case where participants anticipate opportunities to perform the task again (arrow 6), or participants may stop engaging in preparatory attentional processing and continue with the ongoing task only (arrow 7).
Another way to achieve success is to engage only in preparatory attentional processing, which would be accompanied by complete neglect of the ongoing task (arrow 8). This could lead to successful prospective memory performance assuming accurate target recognition and action recall, but this would be at the expense of the other activity.
Even if participants engage in preparatory attentional processing, they may fail to recognize the target event as a performance opportunity or may fail to recall the intended action. Both of these cases would lead to a failure to perform the intended action. In this case the performance of a3 would be completed and followed by the next action a4 (pathway 9).
What is the relationship between the preparatory attentional processes and the processing needed for the ongoing task? The PAM theory proposes that we switch back and forth between actions a3 and aA. This is thought to be similar to situations in which secondary tasks are imposed during encoding. The switching between the perception and comprehension processes involved in encoding information and the processes required for the secondary task, according to Craik, Govoni, Naveh-Benjamin, and Anderson (1996), is controlled and must therefore require capacity. The allocation of capacity to aA, the preparatory attentional processes, is necessary for stimuli to be interpreted; if the situation for action is correctly interpreted, this will lead to the case of "I a1." Thus, capacity is involved in the initiation of a delayed intention, not just in the execution of the action.
The likelihood of initiating a delayed intention to a particular target will be determined in part by the amount of capacity devoted to preparatory attentional processes. The capacity available for preparatory attentional processes is a function of the capacity devoted to ongoing activities, motivation, and overall available capacity. Furthermore, even though some minimal amount of capacity might be devoted to preparatory attentional processes, as mentioned earlier, these processes need not be the focus of attention; that is, participants need not be explicitly checking for performance opportunities. In fact it is more likely the case that preparatory attentional processes will operate outside of the focus of attention and instead on the periphery of awareness. The distinction between focal attention and peripheral awareness idea is analogous to Wundt's (1912/1973) distinction between apprehension and apperception. This distinction was drawn to explain the relationship between the contents of focal attention and the contents of consciousness. Consciousness is limited, but focal attention is even more restricted. When the contents of attention reach full-blown awareness, they are apperceived. In contrast, contents that are above the threshold of consciousness, but of which we are only minimally aware or perhaps unaware, are apprehended. Although outside the focus of attention, apprehended contents occupy some of our limited cognitive capacity. Analogous to Wundt's ideas, the preparatory attentional processes can be the focus of attention but also can operate outside of focal attention, and are therefore called preparatory attentional processes.
Importantly, neither the target nor the action is occupying capacity; rather, it is the processes of evaluating how to respond to our environment that occupy capacity.
Although resource-demanding preparatory attentional processes are required, they are not sufficient to guarantee correct performance. In addition to being prepared to change responses to the environment, one must also correctly recognize when to change the response to the environment. Performing the intention correctly depends on recognizing the target as a target. The probability of correctly recognizing a target event is likely to depend on a number of factors, such as how well the target is processed at the time the intention is formed or the number of different target events to be remembered. Accurate recall of the intended action is also required for successful prospective memory performance. The processes involved in the recognition of target events and recall of the action are thought to be similar to the processes involved in retrospective recognition and recall tasks. It is the requirement for preparatory attentional processes that distinguishes prospective memory tasks from retrospective memory tasks.
The PAM theory is similar in some respects to the multiprocess framework (McDaniel & Einstein, 2000), which proposes that sometimes retrieval of an intention will require resource-demanding strategic processes, and in these cases the predictions made by the multiprocess framework and the PAM theory will generally parallel each other. However, the two explanations differ when it comes to some types of prospective memory tasks. As explained earlier, the PAM theory proposes that prospective memory tasks will always require resource-demanding preparatory processes for successful performance. In contrast, the multiprocess framework proposes that there are some cases in which the intention is retrieved automatically. Specifically, the multiprocess framework proposes that if the action is simple, the target is salient, there is a strong association between the target and the action, and the focal processes involved in the ongoing task include processing of the target, then the intention will be retrieved automatically on presentation of the target. In these select cases, the PAM theory and the multiprocess framework will make different predictions regarding the automaticity of the prospective memory task.
Was this article helpful?