Cognition最新文献

Classification performance is better for learned than unlearned stimuli. This was also reported for faces, where identity matching of unfamiliar faces is worse than for familiar faces. This familiarity advantage led to the conclusion that variability across appearances of the same identity is partly idiosyncratic and cannot be generalized from familiar to unfamiliar identities. Recent advances in machine vision challenge this claim by showing that the performance for untrained (unfamiliar) identities reached the level of trained identities as the number of identities that the algorithm is trained with increases. We therefore asked whether humans who reportedly can identify a vast number of identities, such as super recognizers, may close the gap between familiar and unfamiliar face classification. Consistent with this prediction, super recognizers classified unfamiliar faces just as well as typical participants who are familiar with the same faces, on a task that generates a sizable familiarity effect in controls. Additionally, prosopagnosics' performance for familiar faces was as bad as that of typical participants who were unfamiliar with the same faces, indicating that they struggle to learn even identity-specific information. Overall, these findings demonstrate that by studying the extreme ends of a system's ability we can gain novel insights into its actual capabilities.

Although humans acquire sophisticated and flexible tool-use skills rapidly throughout childhood, young children and adults still show difficulties using the same object for different functions, manifesting in, for example, perseveration or functional fixedness. This paper presents a novel model proposing bottom-up processes taking place during the acquisition of tool-use abilities through active interaction with objects, resulting in two kinds of cognitive representations of an object: a lower-level, action-centered representation and a higher-level, purpose-centered one. In situations requiring the use of an object to attain a goal, the purpose-centered representation is activated quickly, allowing for an immediate detection of suitable tools. In contrast, activation of the action-centered representation is slow and effortful, but comes with the advantage of offering wide-ranging information about the object's features and how they can be applied. This differential availability and activation of action-centered versus purpose-centered representations also contributes to a deeper understanding of the cognitive mechanisms underlying perseveration or functional fixedness during multifunctional tool use. When applied to the teaching and acquisition of tool use, the model indicates that the form in which object-related information is provided determines which of the two object representations is fostered, thereby either facilitating or complicating the flexible application of an object as a tool for different functions.

Retrieval-induced forgetting (RIF) experiments show that the act of retrieving some recently encoded items from a given conceptual category leads to greater forgetting of competing items from that same category. However, RIF studies using emotional stimuli have produced mixed results, perhaps due to the reinstatement of arousal or negative affect during retrieval practice. To induce forgetting of negative episodic memories more indirectly, we examined if retrieving neutral semantic memories leads to RIF of related negative memories. In two experiments, participants studied eight categorized lists comprised of an equal number of negative and neutral words (Experiment 1) or neutral words preceded by neutral or negative images (Experiment 2). To avoid re-exposing individuals to negative material during retrieval practice, participants then performed a semantic memory retrieval task in which they generated (i.e., completed word-stems for) new neutral words from half of the studied categories. We found that semantic retrieval, or word generation, induced forgetting of recently studied words irrespective of their emotional valence or original emotional context. Additionally, across both experiments, less successful word generation was associated with stronger RIF effects. In Experiment 2, the magnitude of RIF was also correlated with higher subjective ratings of retrieval effort during word generation. Together, these results suggest that even when retrieving neutral semantic memories, effortful retrieval may enhance inhibitory processes that lead to forgetting of both neutral and negative episodic memories.

Regular print exposure is thought to benefit reading and language processes: those who read more have a larger vocabulary and better spelling and comprehension skills. Yet, there is little or no direct evidence that exposure to print facilitates reading. Here, we used an ecologically valid design to test the impact of print exposure on the early stages of reading in skilled adult readers. Participants read a novel at their own pace. Reading was followed by a lexical decision task, in which the positive trials were words that were exposed in the novel, and matched controls not exposed in the novel. If exposure during reading had a positive impact on subsequent word recognition, exposed words would be processed more efficiently than not-exposed words (exposure effect). This effect was obtained in three experiments. In addition, the effect was not modulated by the amount of exposure (1 vs. 3 occurrences in the text; Experiment 1), or the timing between reading and the exposure test (immediately after reading vs. on the following day; Experiment 3). However, the effect was present only in low-frequency words (Experiment 3). Interpretations of the exposure effect in terms of activation threshold and lexical quality are discussed.

For decades, eyewitness memory research has had the worthy goal of minimizing the chances that an innocent suspect is falsely identified. However, this is not the only goal. Partial receiver operating characteristic (ROC) curves provide a way to identify lineup procedures that keep the false alarm rate low while also maximizing the hit rate. Recently, there have been attempts to extend the ROC curve into high false alarm rate regions that fair lineups are intentionally designed to avoid. These new full ROCs could provide a way for the police to circumvent the protections offered by fillers in a fair lineup. Moreover, these attempts to extend the ROC curve are not based on a mathematically coherent model of latent diagnostic signals. In this article, we empirically demonstrate how this lack of a solid foundation can lead to dubious conclusions, such as eyewitnesses possessing precognition and being able to reliably identify the person they will see commit a crime in the future.

Visual working memory content is commonly thought to be composed of a precise visual representation of stimulus information (e.g., color, shape). Nevertheless, previous research has shown that individuals represent this visual information in different formats, historically dichotomized into “verbal” and “visual” formats. With growing popular knowledge of aphantasia, or the absence of sensory mental imagery, recent studies have demonstrated that individuals with aphantasia perform similarly to individuals with typical imagery on visual working memory tasks. This suggest that the use of non-visual strategies may be sufficient to perform visual working memory tasks, which were previously thought to be strictly visual. To investigate the effects of different strategies on performance in a visual working memory task, we recruited individuals across the visual imagery spectrum and tested their ability to identify relatively small (3°), medium (6°), or large (10°) changes in the degree of orientation of gratings held in working memory. Subsequently, participants indicated the extent to which they used five different strategies: visual, spatial, verbal, semantic, and sensorimotor. Results revealed that individuals with aphantasia and typical imagery performed similarly to each other across all task difficulty levels. Individuals with typical imagery dominantly used visuospatial strategies, but surprisingly, individuals with aphantasia overwhelmingly preferred the use of non-visual spatial and sensorimotor strategies over verbal strategies. These results suggest that non-visual spatial and sensorimotor strategies can be adopted in visual working memory tasks and these strategies are equally effective as visuospatial strategies. This calls for a rethinking of the “visual” versus “verbal” dichotomy, and provides evidence for the use of other non-visual mental representations in working memory tasks.

This article will consider the disruptive impact of generative AI on moral beliefs and practices associated with equality, particularly equality of opportunity. It will first outline a framework for understanding the mechanisms through which generative AI can alter moral beliefs and practices. It will argue that actual and perceived cognitive ability is one of the determinants of social outcomes in modern information economies, and that one of the potential impacts of generative AI is on the distribution of this ability. Emerging, tentative, evidence suggests that generative AI currently displays an ‘inverse skills bias’, which favours those with less actual and perceived cognitive ability. This could have a disruptive impact on current norms of equality of opportunity, particularly with respect to the means and the purpose of such norms. The longer-term impact of generative AI on equality norms is less clear. Generative AI may shift the entire focus of equality norms or deprioritise the value of equality.

In typical adults, recognizing both spoken and written words is thought to be served by a process of competition between candidates in the lexicon. In recent years, work has used eye-tracking in the visual world paradigm to characterize this competition process over development. It has shown that both spoken and written word recognition continue to develop through adolescence (Rigler et al., 2015). It is still unclear what drives these changes in real-time word recognition over the school years, as there are dramatic changes in language, the onset of reading instruction, and gains in domain general function during this time. This study began to address these issues by asking whether changes in real-time word recognition derive from changes in overall language and reading ability or reflect more general age-related development. This cross-sectional study examined 278 school-age children (Grades 1–3) using the Visual World Paradigm to assess both spoken and written word recognition, along with multiple measures of language, reading and phonology. A structural equation model applied to these ability measures found three factors representing language, reading, and phonology. Multiple regression analyses were used to understand how these three factors relate to real-time spoken and written word recognition as well as a non-linguistic variant of the VWP intended to capture decision speed, eye-movement factors, and other non-language/reading differences. We found that for both spoken and written word recognition, the speed of activating target words in both domains was more closely tied to the relevant ability (e.g., reading for written word recognition) than was age. We also examined competition resolution (how fully competitors were suppressed late in processing). Here, spoken word recognition showed only small, developmental effects that were only related to phonological processing, suggesting links to developmental language disorder. However, in written word recognition, competitor resolution showed large impacts of development which were strongly linked to reading. This suggests the dimensionality of real-time lexical processing may differ across domains. Importantly, neither spoken nor written word recognition is fully described by changes in non-linguistic skills assessed with non-linguistic VWP, and the non-linguistic VWP was linked to differences in language and reading. These findings suggest that spoken and written word recognition continue past the first year of life and are mostly driven by ability and not only by overall maturation.

Humans can flexibly use metacognition to monitor their own knowledge and strategically acquire new information when needed. While humans can deploy these skills across a variety of contexts, most evidence for metacognition in animals has focused on simple situations, such as seeking out information about the location of food. Here, we examine the flexibility, breadth, and limits of this skill in chimpanzees. We tested semi-free-ranging chimpanzees on a novel task where they could seek information by standing up to peer into different containers. In Study 1, we tested n = 47 chimpanzees to assess if chimpanzees would spontaneously engage in information-seeking without prior experience, as well as to characterize individual variation in this propensity. We found that many chimpanzees engaged in information-seeking with minimal experience, and that younger chimpanzees and females were more likely to do so. In two subsequent studies, we then further tested chimpanzees who initially showed robust information-seeking on new variations of this task, to disentangle the cognitive processing shaping their behaviors. In Study 2, we examined how a subset of n = 12 chimpanzees applied these skills to seek information about the location versus the identity of rewards, and found that chimpanzees were equally adept at seeking out location and identity information. In Study 3, we examined whether a subset of n = 6 chimpanzees could apply these skills to make more efficacious decisions when faced with uncertainty about reward payoffs. Chimpanzees were able to use information-seeking to resolve risk and choose more optimally when faced with uncertain payoffs, although they often also engaged in information-seeking when it was not strictly necessary. These results identify core features of flexible metacognition that chimpanzees share with humans, as well as constraints that may represent key evolutionary shifts in human cognition.

When thinking about possibility, one can consider both epistemic and deontic principles (i.e., physical possibility and permissibility). Cultural influences may lead individuals to weigh epistemic and deontic obligations differently; developing possibility conceptions are therefore positioned to be affected by cultural surroundings. Across two studies, 251 U.S. and Chinese 4-, 6-, and 8-year-olds sampled from major metropolitan areas in Texas and the Hubei, Sichuan, Gansu, and Guangdong Provinces judged the possibility of impossible, improbable, and ordinary events. Across cultures and ages, children judged ordinary events as possible and impossible events as impossible; cultural differences emerged in developing conceptions of improbable events. Whereas U.S. children became more likely to judge these events possible with age, Chinese children's judgments remained consistent with age: Chinese 4- to 8-year-olds judged these events to be possible ∼25% of the time. In Study 2, to test whether this difference was attributable to differential prioritization of epistemic versus deontic constraints, children also judged whether each event was an epistemic violation (i.e., required magic to happen) and a deontic violation (i.e., would result in someone getting in trouble). With age, epistemic judgments were increasingly predictive of possibility judgments for improbable events for U.S. children, and decreasingly so for Chinese children. Contrary to our predictions, deontic judgments were not predictive. We propose that cultural valuation of norms might shape children's developing intuitions about possibility. We discuss our findings in light of three accounts of possibility conceptions, suggesting ways to integrate cultural context into each.