Cognitive Psychology最新文献

In economic decision-making there is a trade-off between deliberation time to make a good decision and opportunity costs of other rewarding activities. Recent theories describe how the optimal strategy of evidence accumulation for this problem depends on the environment. If the utility difference between two options is known a priori, but not the identity of the better option, decision-makers should accumulate evidence according to a drift diffusion model with constant decision boundaries. If this difference is unknown beforehand, collapsing boundaries should be used. The exact position of the boundaries depends on the opportunity costs. In two experiments, we examined whether people can adaptively adjust their decision bounds. Participants rated and chose between risky lotteries, while we varied prior information about the utility difference. We also varied opportunity costs, by imposing time limits on task blocks. We found that participants used collapsing boundaries in all examined conditions, even in those where constant boundaries would have been optimal. This means they reduced their target strength of evidence during the choice process, even when they should not. In contrast, participants were sensitive to opportunity costs, deciding faster when choice time was more costly. In sum, people adapted to opportunity costs but not to prior information about utility differences.

Working memory updating is the process that replaces outdated content in working memory by new content. This requires removing outdated information and encoding new information. It is still unclear whether removal and encoding run sequentially or simultaneously. We explored this question in two experiments investigating the time course of removal and encoding and their consequences for the functioning of working memory. The updating task we used involved three phases: the initial encoding, the processing, and the retrieval phase. Across four conditions, we manipulated whether the processing phase involved encoding, removal, neither, or both (i.e., updating). In Experiment 1, processing time was self-paced, and we measured processing times in each condition. In Experiment 2, we measured accuracy as a function of available processing time. After the processing, participants were asked to recall the final item for each position in the retrieval phase. In combination, the results of the two experiments show that the time required for updating was shorter than the sum of encoding and removal time. Moreover, it was nearly the same as the time taken for either the encoding or removal process, indicating that encoding and removal are concurrent processes during updating. Additionally, we analyzed the proportion of correct responses and of different error types with a memory measurement model to investigate the effects of encoding and removal for information held in working memory. The analysis revealed that removal involves unbinding the outdated information from its context. However, despite the weakened bindings of information to its initial context, the outdated information still remains activated in working memory. Other information held in working memory benefitted little from removal of outdated information.

Discriminatory acts (i.e., harmful acts motivated by the victim's group membership) have outsized consequences for the victim and for society relative to similar harms committed for other reasons. Here, we investigated the development of children's evaluations of discrimination. Specifically, we asked whether children in the U.S., like adults, perceive discriminatory acts as distinctly harmful-that is, more harmful than identical acts that are not motivated by the victim's membership in a particular group. Across 4 studies, we examined children's (N = 588; ages 4-9 years) and adults' (N = 623) perceptions of discriminatory acts versus identical acts motivated by other, personal reasons (Studies 1 and 2). In contrast to adults, children-particularly younger ones-rated the discriminatory acts as least harmful. In addition, whereas adults rated discrimination motivated by the victim's membership in an unfamiliar social category (similar to gender or race) as more harmful than discrimination motivated by membership in an unfamiliar task-based group (a sports team), children did not (Study 3). Finally, both adults and older (but not younger) children rated discrimination against a member of a lower-status (vs. equal-status) group as more harmful (Study 4). These findings advance theory on the development of sociomoral cognition and provide new insight into how children perceive instances of discrimination and bias in their everyday lives.

Episodic memories, particularly temporal order memory, are influenced by event boundaries. Although numerous theoretical and computational models have been developed to explain this phenomenon, creating a model that can explain a wide range of behavioral data and is supported by neural evidence remains a significant challenge. This study presented a new model, grounded in ample evidence of position coding, to account for the impact of event boundaries on temporal order memory. The proposed model successfully simulated various behavioral effects in previous experiments measuring temporal order memory. Our model outperformed the context-resetting model in fitting all the data and capturing the full set of effects in previous and newly conducted experiments, including the boundary effect, the distance effect, the local primacy effect, and the absence of boundary number effect. These findings underscore a novel mechanism in which event boundaries affect temporal order memory by resetting the local position coding of events.

Some seemingly irrational decision behaviors (anomalies), once seen as flaws in human cognition, have recently received explanations from a rational perspective. The basic idea is that the brain has limited cognitive resources to process the quantities (e.g., value, probability, time, etc.) required for decision making, with specific biases arising as byproducts of the resource allocation that is optimized for the environment. While appealing for providing normative accounts, the existing resource-rational models have limitations such as inconsistent assumptions across models, a focus on optimization for one specific aspect of the environment, and limited coverage of decision anomalies. One challenging anomaly is the peanuts effect, a pervasive phenomenon in decision-making under risk that implies an interdependence between the processing of value and probability. To extend the resource rationality approach to explain the peanuts effect, here we develop a computational framework-the Assemblable Resource-Rational Modules (ARRM)-that integrates ideas from different lines of boundedly-rational decision models as freely assembled modules. The framework can accommodate the joint functioning of multiple environmental factors, and allow new models to be built and tested along with the existing ones, potentially opening a wider range of decision phenomena to bounded rationality modeling. For one new and three published datasets that cover two different task paradigms and both the gain and loss domains, our boundedly-rational models reproduce two characteristic features of the peanuts effect and outperform previous models in fitting human decision behaviors.