We rely heavily on the contents of long-term memory to furnish judgmental and behavioral responses. Through recollections of the past, we provide ourselves with knowledge that we use to guide future judgments and actions. Sometimes judgments are made after an extensive search in memory and an elaborate analysis of the retrieved information. On other occasions, judgments are made relatively effortlessly on the basis of knowledge that comes to mind most readily. Effortful or not, in both cases the contents of our recollections constitute the input on which future decisions on courses of action are based. Admittedly, this notion that judgmental outcomes are affected by knowledge retrieved from memory is so obvious that it comes across as fairly trivial. After all, it would be a waste not to use our past experiences and the contents of long-term memory.

Still, recent research suggests that it is not necessarily the content of our recollections per se that determines judgmental outcomes. When we busy our minds to arrive at a particular judgment, we retrieve instances pertaining to these judgments. Apart from the contents, however, our judgments can be affected by the experienced ease or difficulty of retrieval that accompanies this retrieval process. In other words, the perceived ease or difficulty of executing a mental activity can be an important mediator between the memorial recollections and the judgmental outcomes, because the experienced ease or difficulty itself serves as an informational cue for judgment.

As a further elaboration of the cognitive mechanism underlying Tversky and Kahneman’s (1973) availability heuristic, Schwarz et al. (1991) demonstrated that judgments can be based on the perceived ease or difficulty with which individuals bring instances to mind, rather than the content of their recall itself. Schwarz et al. (1991) asked participants to list either 6 or 12 instances in which they behaved assertively. Pretests indicated that recalling six examples was experienced as relatively easy, while recalling 12 examples was experienced as difficult. After retrieving the examples, participants rated how much difficulty they had experienced retrieving the examples and evaluated their assertiveness on a 10-point scale. If participants were merely to rely on the content of recall, they would report higher assertiveness after recalling 12 rather than 6 examples. As Schwarz et al. (1991) expected, this is not what happened. Self-ratings of assertiveness showed that participants perceived themselves as less assertive after recalling 12 rather than 6 examples of assertive events. Apparently, participants who were asked to retrieve 12 instances concluded that they were not very assertive simply because retrieving the instances was perceived to be very difficult. To support this argument, Schwarz et al. (1991) showed that participants’ ratings of ease of retrieval were negatively correlated with the self-ratings of assertiveness. Thus, the more difficult participants found the recall task, the lower their self-ratings of assertiveness.

The work of Schwarz et al. (1991) has inspired some researchers to test the experienced ease of retrieval effect in other areas, such as attitude judgments, stereotyping and impression formation. The data of this impressive but small set of available studies clearly provide evidence for the fact that experienced ease of retrieving material from long-term memory can serve as an informational cue to guide judgments of a variety of targets, such as “Are men impolite?”, or “Do I like riding a bike?”.

The present study aims to further demonstrate the effects of experienced ease of retrieval on judgments in a different domain, namely, reported frequency of personally performed past behavior. Instead of establishing the influence of experienced ease of retrieval on evaluations of social or nonsocial targets, we focus on the influence of experienced ease of retrieving instances of a given behavior on the reported number with which that behavior has been performed itself. It is our intuition that experienced ease of recalling different examples of past actions not only affects evaluations of the self (as in the Schwarz et al., 1991 study), but also plays an important role in the reported frequency of occurrence of these same actions in the past. The question “How often did you behave assertively?” or “How often did you travel by bicycle last month?”, will elicit a search in personal memory for relevant past actions and events to subsequently produce an answer revealing a summary quantity. The subjective experiences accompanying this search in memory will affect the reported frequency of this behavior, and, following the logic of Schwarz et al. (1991), if we experience difficulties in recollecting instances of bike use, we should draw the conclusion that we did not use the bike very often (resulting in lower frequency judgments).

The fact that we believe that judgments about frequency of past behavior is subject to the experienced ease of retrieving instances of that behavior is based on the following consideration. When requested to report on the frequency of actions we engage in relatively often, the recall of all the instances usually requires much cognitive effort and time. Research on the cognitive operations underlying the generation of self-reports of frequent behavior suggest that persons make no attempt to search for and retrieve all episodic instances. Instead, they estimate the frequency of behavior or use a general impression. We believe that, in many cases, individuals will use the availability heuristic in which the estimate is based on the experienced ease of retrieving instances. In a task context, then, where the demands of memory retrieval are varied (i.e., easy versus difficult), we anticipate that subjective ease of recall will affect frequency estimates of past behaviors. Specifically, the number of reported past actions would decrease as the difficulty of retrieving examples of past actions from memory increases.

As an important extension to previous work, we expect that the ease of retrieval effects on frequency estimates of past behavior to be moderated by the individual’s motivation to be accurate. Accuracy motivation is generally thought to influence judgment through the initiation of a more systematic and effortful mode of information processing . In other words, upon setting the goal to arrive at correct and objectively valid judgments persons turn to strategies in which they concentrate on and scrutinize relevant information, which attenuates the impact of information that is not (or less) germane to making the judgments. For instance, dual-process models of persuasion and attitude change posit that individuals evaluate persuasive messages on the basis of argument quality when they are motivated to hold accurate attitudes. Consequently, they are less influenced by peripheral cues (e.g., an attractive source) or abandon the use of simple decision rules (e.g., “experts can be trusted”). In short, people do not (or to a lesser extent) rely on heuristics to arrive at judgments when they are instigated to be accurate and precise. The same logic may be extended to the perceived ease of retrieval effects on judgments such as frequency estimates of past behaviors. The experienced ease of retrieving information from long-term memory can be conceived of as a heuristic cue that guides judgmental outcomes in certain situational contexts. Consistent with this line of reasoning, then, we expect that ease of retrieval effects on self-reported frequencies of past behavior will be manifest when individuals are not motivated to be accurate. Ease of retrieval effects will be absent when self-reports are driven by accuracy motivation.