Do Smarter People Employ Better Decision Strategies? The Influence of Intelligence on Adaptive Use of the Recognition Heuristic

Within the adaptive toolbox approach, it has repeatedly been shown that, on average, people tend to adapt their decision strategies to the decision context. Building upon these results, we investigated whether individuals systematically differ in their ability to successfully adapt to the situation when applying the fast‐and‐frugal recognition heuristic (RH). In decisions between recognized and unrecognized choice objects, individuals can base their choices solely on recognition, as predicted by the RH, or they can use further knowledge retrieved from memory. Since intelligence has been conceived as the ability to successfully adapt to different situations, we expected intelligence to influence the degree of adaptive use of the RH. To test this hypothesis, we first re‐analyzed data that referred to a decision domain for which RH‐use is known to perform well. As expected, individual RH‐use increased with general intelligence. Next, we designed an experiment addressing individual RH‐use in two new decision domains, one domain for which RH‐use was less effective than knowledge‐use and another domain for which both strategies were about equally effective. In addition, we tested whether fluid or crystallized intelligence best predicts adaptive use of the RH. RH‐use was found to decrease with fluid but not crystallized intelligence when RH‐use was less effective than use of further knowledge. In contrast, there was no significant association between either type of intelligence and RH‐use when none of the two strategies was optimal. We conclude that adaptive use versus non‐use of the RH is moderated by fluid intelligence. Copyright © 2017 John Wiley & Sons, Ltd.     

A matter of time: Antecedents of one-reason decision making based on recognition

The notion of adaptive decision making implies that strategy selection in both inferences and preferences is driven by a trade-off between accuracy and effort. A strategy for probabilistic inferences which is particularly attractive from this point of view is the recognition heuristic (RH). It proposes that judgments rely on recognition in isolation-ignoring any further information that might be available-and thereby allows for substantial effort-reduction. Consequently, it is herein hypothesized that and tested whether increased necessity of effort-reduction-as implemented via time pressure-fosters reliance on the RH. Two experiments corroborated that this was the case, even with relatively mild time pressure. In addition, this result held even when non-compliance with the response deadline did not yield negative monetary consequences. The current investigations are among the first to tackle the largely open question of whether effort-related factors influence the reliance on heuristics in memory-based decisions.     

Judgments of learning reflect encoding fluency: conclusive evidence for the ease-of-processing hypothesis

According to the ease-of-processing hypothesis, judgments of learning (JOLs) rely on the ease with which items are committed to memory during encoding--that is, encoding fluency. Conclusive evidence for this hypothesis does not yet exist because encoding fluency and item difficulty have been confounded in all previous studies. To disentangle the effects of encoding fluency and item difficulty on JOLs, we used a variant of the learner-observer-judge method in which participants observed the study phase of another participant and indicated his or her JOLs. At the same time, the to-be-studied word pairs were concealed by strings of symbols. Our experiment revealed that participants use self-paced study time as a cue for JOLs when they themselves have studied and recalled word pairs before. This metacognitive monitoring of study time provides strong support for the ease-of-processing hypothesis.     

The revelation effect for autobiographical memory: A mixture-model analysis

Participants provided information about their childhood by rating their confidence about whether they had experienced various events (e.g., “broke a window playing ball”). On some trials, participants unscrambled a key word from the event phrase (e.g., wdinwo—window) or an unrelated word (e.g., gnutge—nugget) before seeing the event and giving their confidence ratings. The act of unscrambling led participants to increase their confidence that the event occurred in their childhood, but only when the confidence rating immediately followed the act of unscrambling. This increase in confidence mirrors the “revelation effect” observed in word recognition experiments. In the present article, we analyzed our data using a new signal detection mixture distribution model that does not require the researcher to know the veracity of memory judgments a priori. Our analysis reveals that unscrambling a key word or an unrelated word affects response bias and discriminability in autobiographical memory tests in ways that are very similar to those that have been previously found for word recognition tasks.     

What kind of bias is hindsight bias?

When people are asked to recollect a former response after having received feedback information, their recollection tends to approach the feedback answer. This phenomenon is referred to as hindsight bias. Recently, Erdfelder and Buchner proposed a multinomial model designed to independently measure the contributions of reconstruction and recollection processes. On its basis, they only found firm evidence for the contribution of reconstruction biases to hindsight effects. In the present study, we tried to experimentally enhance the probability of recollection biases by (a) reducing the depth of processing of the original estimate, (b) minimizing the distinctness of the original estimate and feedback, and (c) combining both treatments. The empirical data were analyzed using a variant of the Erdfelder and Buchner model which allows for the experimental manipulation of the feedback answer. This model is shown to adequately describe the data of our experiment. Moreover, although both treatments were effective in that the first one reduced correct recollections in general and the second one diminished differences between original estimates and recalled estimates, the parameter estimates indicate that the probability of recollecting the original estimate is not hampered by feedback information, even under conditions which are thought to enhance memory impairments. Received: 10 December 1995 / Accepted: 29 October 1997     

GPOWER: A general power analysis program

GPOWER is a completely interactive, menu-driven program for IBM-compatible and Apple Macintosh personal computers. It performs high-precision statistical power analyses for the most common statistical tests in behavioral research, that is,t tests,F tests, andχ ² tests. GPOWER computes (1) power values for given sample sizes, effect sizes andα levels (post hoc power analyses); (2) sample sizes for given effect sizes,α levels, and power values (a priori power analyses); and (3)α andβ values for given sample sizes, effect sizes, andβ/α ratios (compromise power analyses). The program may be used to display graphically the relation between any two of the relevant variables, and it offers the opportunity to compute the effect size measures from basic parameters defining the alternative hypothesis. This article delineates reasons for the development of GPOWER and describes the program’s capabilities and handling.     

A stochastic lie detector

Surveys on sensitive issues provide distorted prevalence estimates when participants fail to respond truthfully. The randomized-response technique (RRT) encourages more honest responding by adding random noise to responses, thereby removing any direct link between a participant’s response and his or her true status with regard to a sensitive attribute. However, in spite of the increased confidentiality, some respondents still refuse to disclose sensitive attitudes or behaviors. To remedy this problem, we propose an extension of Mangat’s (Journal of the Royal Statistical Society: Series B, 56, 93–95, 1994) variant of the RRT that allows for determining whether participants respond truthfully. This method offers the genuine advantage of providing undistorted prevalence estimates for sensitive attributes even if respondents fail to respond truthfully. We show how to implement the method using both closed-form equations and easily accessible free software for multinomial processing tree models. Moreover, we report the results of two survey experiments that provide evidence for the validity of our extension of Mangat’s RRT approach.     

Encoding, maintenance, and retrieval processes in the lag effect: a multinomial processing tree analysis

Short-term studies on repeated learning of verbatim material have typically revealed an overall benefit of long lags compared to short lags between repetitions. This has been referred to as the lag effect. On educationally relevant time scales, however, an inverted-U-shaped relation between lag and memory performance is often observed. Recently, Cepeda et al. (2009) showed that the optimal lag for relearning depends heavily on the time interval between the last learning session and the final memory test (i.e., the retention interval; RI). In order to explore the cognitive mechanisms underlying this result in more detail we independently manipulated both the lag and the RI in a 3×2 experimental design and analysed our data using a multinomial processing tree model for free-then-cued-recall data. Our results reveal that the lag effect trends are mainly driven by encoding and maintenance processes rather than by retrieval mechanisms. Our findings have important implications for theories of the lag effect.