Cognition最新文献

Lineups are considered a superior method of identification to showups, but why is contested. There are two main theories: diagnostic feature detection theory, which holds that surrounding the suspect with fillers causes the eyewitness to focus on the features that are most diagnostic, and differential filler siphoning theory that claims that the fillers draw incorrect choices away from the suspect. Colloff and Wixted (2020) created a novel identification task, called a simultaneous showup, designed to prevent filler siphoning, while still allowing comparison to occur between members of the array. However, even in the simultaneous showup, it is possible that covert filler siphoning occurs. In Experiment 1, we replicated the simultaneous showup condition and also asked participants if the other photos affected their decision making; we found evidence that participants self-reported both diagnostic feature detection and covert filler siphoning. In Experiment 2, we replicated Colloff and Wixted (2020, Experiment 3) main findings. Additionally, we found that participants self-reported both diagnostic feature detection and covert filler siphoning. This led us to conclude that the simultaneous showup procedure could not fully exclude covert filler siphoning from occurring.

Crossmodal correspondences, the tendency for a sensory feature / attribute in one sensory modality (either physically present or merely imagined), to be associated with a sensory feature in another sensory modality, have been studied extensively, revealing consistent patterns, such as sweet tastes being associated with pink colours and round shapes across languages. The present research explores whether such correspondences are captured by ChatGPT, a large language model developed by OpenAI. Across twelve studies, this research investigates colour/shapes-taste crossmodal correspondences in ChatGPT-3.5 and -4o, focusing on associations between shapes/colours and the five basic tastes across three languages (English, Japanese, and Spanish). Studies 1A-F examined taste-shape associations, using prompts in three languages to assess ChatGPT's association of round and angular shapes with the five basic tastes. The results indicated significant, consistent, associations between shape and taste, with, for example, round shapes strongly associated with sweet/umami tastes and angular shapes with bitter/salty/sour tastes. The magnitude of shape-taste matching appears to be greater in ChatGPT-4o than in ChatGPT-3.5, and ChatGPT prompted in English and Spanish than ChatGPT prompted in Japanese. Studies 2A-F focused on colour-taste correspondences, using ChatGPT to assess associations between eleven colours and the five basic tastes. The results indicated that ChatGPT-4o, but not ChatGPT-3.5, generally replicates the patterns of colour-taste correspondences that have previously been observed in human participants. Specifically, ChatGPT-4o associates sweet tastes with pink, sour with yellow, salty with white/blue, bitter with black, and umami with red across languages. However, the magnitude/similarity of shape/colour-taste matching observed in ChatGPT-4o appears to be more pronounced (i.e., having little variance, large mean difference), which does not adequately reflect the subtle nuances typically seen in human shape/colour-taste correspondences. These findings suggest that ChatGPT captures colour/shapes-taste correspondences, with language- and GPT version-specific variations, albeit with some differences when compared to previous studies involving human participants. These findings contribute valuable knowledge to the field of crossmodal correspondences, explore the possibility of generative AI that resembles human perceptual systems and cognition across languages, and provide insight into the development and evolution of generative AI systems that capture human crossmodal correspondences.

The word-frequency distributions children hear during language learning are highly skewed (Zipfian). Previous studies suggest that such skewed environments confer a learnability advantage in tasks that require the learner to discover the units that have to be learned, as in word-segmentation or cross-situational learning. This facilitative effect has been attributed to contextual facilitation from high frequency items in learning lower frequency items, and to better learning under the increased predictability (lower entropy) of skewed distributions. Here, we ask whether Zipfian distributions facilitate learning beyond the discovery of units, as expected under the predictability account. We tested children's learning of novel word-referent mappings in a learning task where each mapping was presented in isolation during training, and did not need to be dicovered. We compared learning in a uniform environment to two skewed environments with different entropy levels. Children's learning was overall better in the two skewed environments, even for low frequency items. These results extend the facilitative effect of Zipfian distributions to additional learning tasks and show they can facilitate language learning beyond the discovery of units.

Perception has long been envisioned to use an internal model of the world to explain the causes of sensory signals. However, such accounts have historically not been testable, typically requiring intractable search through the space of possible explanations. Using auditory scenes as a case study, we leveraged contemporary computational tools to infer explanations of sounds in a candidate internal generative model of the auditory world (ecologically inspired audio synthesizers). Model inferences accounted for many classic illusions. Unlike traditional accounts of auditory illusions, the model is applicable to any sound, and exhibited human-like perceptual organization for real-world sound mixtures. The combination of stimulus-computability and interpretable model structure enabled ‘rich falsification’, revealing additional assumptions about sound generation needed to account for perception. The results show how generative models can account for the perception of both classic illusions and everyday sensory signals, and illustrate the opportunities and challenges involved in incorporating them into theories of perception.

People appear to prefer explanations that minimize unobserved effects, a pattern known as the latent scope bias in explanatory reasoning. A recent set of studies published in Cognition argues that the bias can be elicited only in certain narrow conditions and with certain tasks, such as a forced-choice task (Stephan, 2023). This commentary assesses the robustness of the bias in two ways: it weighs the most recent discoveries against previous research, and it presents two new studies using the most general possible elicitation task, i.e., spontaneous written responses to problems designed to test for a latent scope bias. Across 35 previous studies, 7 studies published in Stephan (2023), and 2 new studies described herein, the overwhelming majority of studies showed that people preferred narrow latent scope explanations over broad ones. This analysis led us to conclude that the bias is both robust and replicable. Taken together, Stephan's (2023) contribution and our new analyses advance our understanding of explanatory reasoning behavior.

People are offloading many tasks to artificial intelligence (AI)—including driving, investing decisions, and medical choices—but it is human nature to want to maintain ultimate control. So even when using autonomous machines, people want a “manual mode”, an option that shifts control back to themselves. Unfortunately, the mere existence of manual mode leads to more human blame when AI makes mistakes. When observers know that a human agent theoretically had the option to take control, the humans are assigned more responsibility, even when agents lack the time or ability to actually exert control, as with self-driving car crashes. Four experiments reveal that though people prefer having a manual mode, even if the AI mode is more efficient and adding the manual mode is more expensive (Study 1), the existence of a manual mode increases human blame (Studies 2a-3c). We examine two mediators for this effect: increased perceptions of causation and counterfactual cognition (Study 4). The results suggest that the human thirst for illusory control comes with real costs. Implications of AI decision-making are discussed.

Inhibition control is an essential executive function during children's development, underpinning self-regulation and the acquisition of social and language abilities. This executive function is intensely engaged in music training while learning an instrument, a complex multisensory task requiring monitoring motor performance and auditory stream prioritization. This novel meta-analysis examined music-based training on inhibition control in children. Records from 1980 to 2023 yielded 22 longitudinal studies with controls (N = 1734), including 8 RCTs and 14 others. A random-effects meta-analysis showed that music training improved inhibition control (moderate-to-large effect size) in the RCTs and the superset of twenty-two longitudinal studies (small-to-moderate effect size). Music training plays a privileged role compared to other activities (sports, visual arts, drama) in improving children's executive functioning, with a particular effect on inhibition control. We recommend music training for complementing education and as a clinical tool focusing on inhibition control remediation (e.g., in autism and ADHD).

Loopholes offer an opening. Rather than comply or directly refuse, people can subvert an intended request by an intentional misunderstanding. Such behaviors exploit ambiguity and under-specification in language. Using loopholes is commonplace and intuitive in everyday social interaction, both familiar and consequential. Loopholes are also of concern in the law, and increasingly in artificial intelligence. However, the computational and cognitive underpinnings of loopholes are not well understood. Here, we propose a utility-theoretic recursive social reasoning model that formalizes and accounts for loophole behavior. The model captures the decision process of a loophole-aware listener, who trades off their own utility with that of the speaker, and considers an expected social penalty for non-cooperative behavior. The social penalty is computed through the listener’s recursive reasoning about a virtual naive observer’s inference of a naive listener’s social intent. Our model captures qualitative patterns in previous data, and also generates new quantitative predictions consistent with novel studies (N $=$ 265). We consider the broader implications of our model for other aspects of social reasoning, including plausible deniability and humor.

In this research, we examine whether moral judgments sometimes violate the normative principle of procedure invariance – that is, whether normatively equivalent elicitation tasks can result in different judgment patterns. Specifically, we show that the relative morality of two actions can reverse across evaluation modes and elicitation tasks, mirroring preference reversals in consumer behavior. Across six studies (five preregistered, total N = 719), we provide evidence of three reversals of moral judgments of sacrificial dilemmas. First, directly killing one person to save many others was rated as morally worse than indirectly killing one person via an intervening mechanism in order to save a few others in separate evaluation, but this difference reversed in joint evaluation, in both between-subjects (Studies 1a and 1b) and within-subjects (Study 2) designs. Next, directly killing one person to save many others was judged as morally better than indirectly killing one person to save a few others more often in matching than in choice (Study 3) and rating (Study 4), between-subjects. Lastly, we replicate the results of Studies 3 and 4 within-subjects and show that susceptibility to these moral preference reversals is correlated with Faith in Intuition (Study 5). The present research introduces a new methodological approach to moral psychology, demonstrates that moral judgments can fully reverse across tasks, and supports an emerging view that moral judgments, like consumer preferences, are at least sometimes constructed in the moment, relative to the context and task at hand.

We explore how DNNs can be used to develop a computational understanding of individual differences in high-level visual cognition given their ability to generate rich meaningful object representations informed by their architecture, experience, and training protocols. As a first step to quantifying individual differences in DNN representations, we systematically explored the robustness of a variety of representational similarity measures: Representational Similarity Analysis (RSA), Centered Kernel Alignment (CKA), and Projection-Weighted Canonical Correlation Analysis (PWCCA), with an eye to how these measures are used in cognitive science, cognitive neuroscience, and vision science. To manipulate object representations, we next created a large set of models varying in random initial weights and random training image order, training image frequencies, training category frequencies, and model size and architecture and measured the representational variation caused by each manipulation. We examined both small (All-CNN-C) and commonly-used large (VGG and ResNet) DNN architectures. To provide a comparison for the magnitude of representational differences, we established a baseline based on the representational variation caused by image-augmentation techniques used to train those DNNs. We found that variation in model randomization and model size never exceeded baseline. By contrast, differences in training image frequency and training category frequencies caused representational variation that exceeded baseline, with training category frequency manipulations exceeding baseline earlier in the networks. These findings provide insights into the magnitude of representational variations that can be expected with a range of manipulations and provide a springboard for further exploration of systematic model variations aimed at modeling individual differences in high-level visual cognition.