The ability to understand and share others' emotional states (e.g., feeling of pain) plays a fundamental role in survival and prosocial behavior. The current study utilized pupillometry to assess automatic psychophysiological responses to others' painful facial expressions in both adults and children (N = 72). Results revealed that pupil size significantly increased when perceiving painful versus neutral expressions, independent of low-level visual features. Notably, both adults and children exhibited a racial in-group bias, with pupil dilation effects observed only for same-race painful faces. Furthermore, individuals' Autism Spectrum Quotient scores were negatively correlated with pupil dilation effects toward painful expressions of same-race faces. These findings suggest that pupillary responses might reflect automatic empathic arousal to others' pain and are modulated by racial group membership and autistic traits, providing a potential physiological indicator, at least at the group level, for probing affective resonance in children or individuals with socio-cognitive disorders (e.g., autism spectrum disorder).

Episodic memory allows us to remember when an event occurred by situating it within a coherent temporal context. Pavlovian fear conditioning, a widely studied form of associative emotional learning, creates implicit memories for neutral stimuli paired with aversive outcomes. However, conditioning's influence on the temporal organization of episodic memory remains poorly understood. We addressed this by analyzing data from 17 multi-session hybrid conditioning-memory experiments (N = 474). Participants encoded non-repeating category items, with items from one category (CS+) being aversively reinforced (shocks) during threat conditioning but presented without shock before and after conditioning. The next day, recognition memory ('did you see this image yesterday?') and temporal source memory ('when did you see this image?') were tested for each category item. We had two aims; (1) examine the robustness of temporal memory distortion across different experiment groups, and (2) test whether these temporal effects were associated with recognition memory performance. CS+ category exemplars were disproportionately (mis)attributed to the conditioning phase, even if they were encoded before or afterwards, and this effect strongly predicted selective recognition memory (CS+ > CS-). Overall temporal source bias effects and source-item memory associations were largely resistant to between-experiment variations, including month-long encoding-retrieval intervals, varying shock intensities, and enhanced extinction. Paradoxically, salient emotional experiences may enhance memory for neutral events by distorting their perceived position in time. This mechanism may safeguard potentially relevant information by anchoring otherwise forgettable experiences to salient contexts, supporting their preservation in long-term memory.

Wishful thinking or desirability bias refers to instances where the desire for an outcome inflates the expectation that it will occur. Although studies have demonstrated influences of outcome desirability on people's predictions, the cognitive mechanisms behind such an effect have remained unclear. Both biased criteria for evidence judgment and biased evidence search/accumulation have been suggested as possible mechanisms. In the present work, we used drift-diffusion modeling to examine on which levels of processing desirability has its impact. Participants (N = 147) made predictions about the color of a randomly selected square from 2-color grids. Crucially, certain color outcomes were made more desirable than others, and the strength of evidence was manipulated by varying the proportion of desired-color squares in the grid. We found that both manipulations-and their interaction-significantly affected predictions. More importantly, drift-diffusion model analyses showed that outcome desirability resulted in a judgment-level bias, where participants required less evidence to predict a desired outcome. Notably, we also found that desirability impacted the evidence accumulation process itself. Participants more readily construed evidence as supporting the desired outcome, indicating that desirability had a top-down influence on how prediction-relevant evidence was accumulated. The present results have implications for existing accounts of how desire impacts expectations and highlight the utility of drift diffusion modeling as a tool for assessing the mechanisms underlying motivated biases.

For billions of bilinguals, many communicative acts involve a choice between languages. Here, we evaluate the theory that bilinguals choose a language to regulate their emotional reactions. We present four experiments demonstrating that language choice could be guided by anticipated emotional impact. Across several languages (Chinese, English, and Spanish), 1083 bilinguals from China, the United States, Canada, Mexico, and Spain preferred a foreign language to speak about embarrassing topics, and this preference was associated with expecting fewer serious emotional and interpersonal consequences. Language preference was a function of native-ness rather than its associated culture, as the effect was evident even when languages were crossed (English native/Spanish foreign, Spanish native/English foreign). Foreign language use increases emotional distance, and bilinguals prefer using a foreign language over a native language to avoid feeling the embarrassment of discussing aversive topics. Hence, language choice could be an emotional regulation tool for bilinguals.

Rhythm, such as in music, contains structure in the form of rhythmic patterns: the more or less predictable successions of longer and shorter intervals (i.e., the "morse code" of the rhythm). Listeners can use rhythmic patterns to predict the timing of sounds and guide their perception and action. It is still unclear how rhythmic patterns are represented in the human mind. Here, we used a probabilistic model of auditory expectations to simulate the perception and production of rhythmic patterns. We modelled expectations in rhythmic sequences at three different levels of abstraction: as the predictability of absolute inter-onset intervals (IOI), ratios between successive intervals (ratio), and the direction of change of successive intervals (contour). Subsequently, we selected rhythms that varied maximally in their modelled predictability across the three levels of abstraction for three behavioral tasks: a target detection task in which the rhythm was not task-relevant (implicit task), a complexity rating task (explicit task), and a tapping task (motor task). We found that both ratio and contour affected behavioral responses across all tasks, with the largest effects in the explicit rating task. IOI only affected responses for the explicit and motor tasks, where the rhythm was task-relevant, and to a greater extent when an imprecise, categorical representation of IOI was assumed. These findings suggest that humans rely mostly on imprecise representations of rhythmic patterns, but may flexibly adapt their representation based on task demands.

How sophisticated is young children's comprehension of geometric lines, curves and patterns, and how can we probe it? We investigated early proto-mathematical intuitions by asking kindergarteners (N = 39, 25 girls, 66 months) and first-graders (N = 42, 20 girls, 76 months) to draw the prolongation of mathematical patterns. Children's drawings revealed an early yet partial understanding of key mathematical properties such as linearity, curvature, periodicity, and compositionality. These abilities were confirmed in a second task, where participants were asked to select the correct prolongation among six options, ruling out motor entrainment as an explanation for the drawing task. These findings highlight children's emerging intuitions of proto-mathematical concepts and underscore the potential of drawing as a powerful and concrete tool for assessing early mathematical reasoning.

It remains unclear how episodic memories, our memories for past events, are generated from the rich sensory details encompassing everyday experience. Here, 40 healthy young participants navigated in a large-scale virtual environment under two conditions: an immersive-ambulatory condition, in which participants walked on a treadmill with full-body motion and head turns, or a restricted-movement condition, in which participants used a hand-held joystick to control gaze and position in virtual-reality while standing still. Participants then freely recalled their experience navigating in the virtual town. Analysis of verbal narratives revealed a double dissociation such that ambulation resulted in more detailed episodic elaborations, particularly those grounded in a first-person perspective, whereas restricted movement promoted more generalized spatial elaborations, particularly those not anchored to participants' point of view. These findings suggest that bodily cues contribute to the construction of episodic memory, while their absence may redirect attention toward external features of the environment. Participants also judged the relative direction of landmarks (space) and reproduced the amount of time they had spent navigating (time); however, objective indicators of spatiotemporal memory did not vary based on condition and, with one exception, did not correlate with subjective measures of memory quality. Together, these results suggest that the construction of vivid episodes is not necessarily tied to the precision of spatial and temporal encoding.

Cross-linguistically, lexicons tend to be more phonetically clustered than required by the phonotactics of the language; that is, words within a language are more similar to each other than they need to be. In this study, we investigate how this property evolves under the influence of competing communicative pressures: a production-side pressure to re-use more easily articulated sounds, and a comprehension-side pressure for distinctiveness of wordforms. In an exemplar-based computational model and a communication experiment using a miniature artificial language, we show that natural-language-like levels of clustering emerge from a trade-off between these pressures. With only one pressure at work, the resulting lexicons tend to inhabit an extreme region of the possible design space: production pressures alone give rise to maximally clustered lexicons, while comprehension pressures alone give rise to maximally disperse lexicons. We also test whether clustering emerges more strongly for high-frequency items, but our results lend support only to a weak relationship between frequency and clustering. Overall, this study adds to a growing body of evidence showing that mechanisms operating at the level of individual language users and individual episodes of communication can give rise to emergent structural properties of language.

A drive towards efficiency seems to regulate communicative processes and ultimately language change. In line with efficiency principles, signed, spoken, and/or gestural utterances tend to reduce in overall effort over repeated referrals in referential tasks. Although theories generally assume multimodality and interaction, this process has mostly been operationalized as individual effort in a single communicative modality. Here we seek to understand reduction of communicative effort in its natural environment, i.e. during multimodal and collaborative face-to-face dialogues about displaced referents. We ascertain that the reduction in joint effort (y) over repeated referrals (x) follows a negative power relationship, y = a*x^c, where a and c are constants. This reduction in communicative effort is multimodal, occurring across gesture, speech, prosody, and turn taking, and it is interactive, based on joint effort. The pattern is robust, being confirmed through reanalyses of published datasets about (individual) effort reduction. Crucially, the pattern is communicatively relevant. The coefficient of the power relationship predicts change and convergence in interlocutors' conceptualizations of the communicative referents over the interaction. The negative power relationship reflects therefore how effort translates into mutual understanding - a process we call communicative work. We suggest that the power function captures an exploration-exploitation trade-off during human dialogue which emerges from multiscale processes. Joint conceptualization of novel referents benefits from early conceptual exploration followed by later exploitation of selected signals. The current report proposes a novel 'power law of joint communicative work' that is relevant for linguistic theory, agent-based modeling, and experimental psychology.

Moore's Paradox-e.g., "It's raining but I don't think it's raining"-is widely considered infelicitous despite being logically consistent. In this paper, we extend Moore's Paradox to moral discourse and test whether moral statements like "Murder is wrong but I don't disapprove of it" elicit similar intuitions. Rooted in moral expressivism, the Parity Thesis predicts that moral assertions express non-cognitive attitudes (e.g., approval/disapproval) in a manner analogous to how descriptive statements express beliefs. In a pre-registered study with 1200 participants, we empirically test this thesis using a mixed design that manipulates moral term type (thick vs thin), evaluative polarity (positive vs negative), perspective (first vs third person), and attitude (belief vs disapproval). The results of our main study and one qualitative follow-up study suggest that while moral statements resemble Moorean Paradoxes in important ways, participants find it largely acceptable to call an action wrong without disapproving of it. As the infelicity of such statements is a core ingredient of Moorean Paradoxes and, as we suggest, the Parity Thesis, we conclude that moral language does not express approval and disapproval like declarative language expresses beliefs.

Autobiographical memory entails the reconstructing of the visual features of past events. While eye movements are associated with vivid autobiographical recollection, this research has yet to capitalize on the high temporal resolution of eye-tracking data. We aligned eye movement data with participants' extemporaneous free recall of a verified real-life event, allowing us to assess the temporal correspondence of saccades to production of episodic and non-episodic narrative content at the millisecond level. Episodic autobiographical details were preceded by an increase in saccade frequency and followed by a reduction in saccades prior to the next detail. There was no such effect observed for non-episodic details. Oculomotor responses in the temporal window preceding freely-recalled details may facilitate recollection by reinstating spatiotemporal context, or they may reflect post-retrieval processes-or a combination of both-in cyclical sensory-motor-mnemonic interactions that promote vivid recall.

Individuating a single item presented within a continuous sequence of items requires segregating its signal from that of the other items. In contrast, representing a global aspect of the sequence, such as its average orientation, involves integration of information across time. Individuation and integration allow us to focus on individual events while maintaining an overall perception of our environment. To examine the relations between temporal averaging and individuation, we measured orientation averaging over short and long timescales using the same stimuli and orientation-estimation procedure previously used to measure individuation. Participants reported the average orientation of a sequence of three oriented items separated by either short (SOAs<150 ms) or long intervals (SOAs>150 ms). Analysis of the error distribution and mixture-modeling revealed distinct patterns of results for the different tasks and timescales, but also some similarities, particularly for the short timescale. In this timescale, the relative contribution of each individual item to the final response was similar across tasks, indicating the involvement of low-level factors operating regardless of the task. With the long timescale, the two tasks showed dissociable pattern across all performance aspects, except guessing rate, indicating that long-scale individuation and averaging engage mainly higher-level, task-related processes. Importantly, regardless of timescale, estimation errors in these tasks were best described by different models: in integration they primarily reflected unequal weighting of the averaged items, whereas in individuation they reflected imprecise target encoding with occasional misreports of distractors. Together, the findings reveal dissociable dynamics for integration and individuation.

Learning science often appears to involve replacement of naïve, intuitive ideas with correct, counterintuitive ones. Recent studies indicate that the old naïve, intuitive ideas are not actually replaced but exist alongside the correct but often counterintuitive ones. On this account, newer knowledge for scientific thinking might involve inhibition of the old idea. However, instead of merely inhibiting old ideas, it is possible that switching is necessary to select between new and old scientific ideas. In this study, we explored the direct and indirect contributions of behavioural inhibition, cognitive inhibition and switching to intuitive and counterintuitive science reasoning in adults (N = 167). After replicating the commonly observed processing costs of counterintuitive items relative to intuitive ones, we find that individual differences in switching rather than in inhibition are most strongly associated with variation in the accuracy and speed of adult intuitive and counterintuitive science reasoning. These results suggest that adults switch between older and newer ideas when reasoning about science rather than suppressing one in favour of the other.

Recent work has proposed that there may be two kinds of beliefs: Symbolic beliefs which express the believer's identity and epistemic beliefs which represent facts. On this proposal, several disparate features of belief - from whether a belief is important to identity to whether it is sensitive to evidence - would be related to a single dimension. In five studies, participants rated beliefs on features that were related to symbolicness and epistemicness. Study 1 found that beliefs which were important to participants were consistently rated higher on the symbolic features, but not consistently lower on the epistemic feature. Study 2 found that the symbolicness features loaded onto a single factor, while the epistemic feature did not. Study 3 expanded the assessment of epistemicness to a series of features related to objectivity, but found that symbolicness and objectivity features loaded on separate, uncorrelated factors. Study 4 recalibrated the assessment of epistemicness using a series of features related to the evidence-drivenness, and again found symbolicness and epistemicness loaded on separate factors. In the final study, third-party participants rated the beliefs from Study 4 on symbolicness and evidence-drivenness, and largely replicated the findings from Study 4. These results suggest that symbolicness and epistemicness are coherent dimensions of belief, but that these dimensions are largely orthogonal.

The present study examined how bilinguals switch languages under conditions of varying contextual support for each language. Bilinguals were cued to name pictures in two contextually biased blocks, one biasing the dominant-language by cuing its use on 88 % of trials, another biasing the nondominant-language, and a third no-bias block (in which each language was used 50 % of the time). Experiment 1 tested 70 young Spanish-English bilinguals, while Experiment 2 compared a proficiency-matched subset of the young bilinguals to 40 older bilinguals tested on the same tasks. In the no-bias block, young but not older bilinguals exhibited reversed language dominance, replicating a previously reported aging deficit in global inhibition of the dominant language. In biased-language contexts, young, but not older, bilinguals switched-back to the biased language faster than they switched-out, and switching-back was not easier than switching-out, if anything, switching back was more costly (for young bilinguals in Experiment 1, and for older bilinguals in Experiment 2). Surprisingly, older bilinguals exhibited larger switch-back, but not switch-out, costs than younger bilinguals. To explain these results, we hypothesize that young bilinguals engage multiple forms of proactive control to switch languages, including inhibition of the dominant language, and proactive activation of a selected language, which they maintain even while temporarily switching out of it to facilitate imminent switches back. By contrast, older bilinguals rely primarily on reactive control to switch languages, and without proactive selection, bottom-up activation of the contextually supported language collapses upon switching out of it.

Language acquisition involves drawing systematic generalizations from messy data. On one hypothesis, this is facilitated by a domain-general bias for children to "regularize" their input, sharpening the statistical distributions in their input towards more systematic extremes. We introduce a general computational framework for modeling a different explanation: on this view, children expect that their data are a noisy realization of a restrictive underlying grammatical system. We implement a learner that evaluates a choice among composite context-free grammars, in which a restricted set of "core" rules, comprising the particular grammatical processes that the learner is currently trying to acquire, operate alongside a less restricted set of "noise" rules, representing other independent processes that have yet to be learned, and conspire to introduce distortions into the data. Our Noisy Grammar Learner partitions its data into portions that serve as evidence for one of the possible core grammars in its hypothesis space, and portions generated by these noise processes. It does so without knowing in advance how much noise occurs or what its properties are. We compare our learner to a common implementation of the general regularization bias approach, and show that both can account for children's behavior in a representative artificial language learning experiment. However, we find that only our approach succeeds on two naturalistic case studies in early syntax acquisition: learning the rules governing canonical word-order and case-marking, given natural language data with "noise" from non-canonical sentence types. We show that our learner succeeds because its architecture allows a natural way to express linguistically-motivated expectations about the character of those rules. This suggests that, in certain domains, successful learning from messy data may be enabled by a hypothesis space comprising restrictive grammatical options.

Speaking a language that obliges the future tense for linguistic Future Time Reference (FTR) may cause speakers to devalue future outcomes. Evidence suggests such grammars make speakers less "future-oriented": less likely, for example, to invest, eat healthily, or support costly climate change mitigation efforts. This has been explained using the notion that the future tense (e.g., will) encodes temporal notions of distance and/or precision; its obligatory use is therefore hypothesized to cause speakers to perceive delayed outcomes as less valuable. We argue that this causal account is not supported by extant evidence. Rather, we hypothesize the obligation to use low-certainty modal verbs (e.g., may) causes speakers to construe delayed outcomes as risky and therefore less valuable. We tested this in speakers of Dutch (which does not oblige FTR marking) and English (which does). English speakers used more low-certainty modal verbs, which in turn caused them to place a relatively lower value on future outcomes; at the same time, future tense had no effect, in terms of either distance or precision, on reward value construals (Study 1). When bilinguals were tested in English and Dutch, increased relative use of low-certainty modals again caused English speakers to devalue future outcomes, addressing possible cultural confounds (Study 2). English and Dutch speakers were tested on a non-linguistic probability estimation task; higher modal verb use in English caused lower probability estimates relative to Dutch speakers on matched visual stimuli-supporting the modal account that the obligation to use low-certainty language impacts judgments about probability (Study 3). Relative to matched US nationals, corporate executives from countries which speak languages that, like Dutch, do not oblige future statements to be grammatically marked, used fewer low-certainty modal verbs and more present tense FTR statements, while there was no difference in future tense use (Study 4)-broadly supporting the modal account by suggesting the modal differences characteristic of English and Dutch are widespread. Together, these results indicate that, relative to Dutch, English FTR requires speakers to use more low-certainty modals, and that this negatively biases construals of probability, which in turn leads to increased discounting (Studies 1-3), and that this cross-linguistic contrast may be general (Study 4). The studies provide evidence for linguistic relativity by identifying cross-linguistic effects of FTR grammar on discounting via low-certainty modals. However, the hypothesis that obligatory tenses impacted discounting via temporal notions was not supported, suggesting numerous reported results should be re-evaluated using the causal framework we propose.

There are many factors that can influence a voter's decision in the ballot booth but not all of them are policy related. One non-policy factor that may influence voters is the tendency to choose options that attract attention. Here, we investigate this possibility in two proof-of-concept laboratory studies with people choosing between proposed laws. We find that people are slower to vote when their party is split over an issue, and that they tend to vote for laws that they look at more. Moreover, this gaze effect is stronger for more important issues. We also find that we can increase the probability that someone will vote for one of two laws by getting them to look at that option first. Our work harnesses the power of sequential sampling models to explain the relationship between gaze and vote choice. We find support for a goal-based model where overt attention amplifies information supporting a particular law. This model explains why gaze has a stronger effect on choice for more important issues. Our findings indicate that some voting decisions are not predetermined and instead rely on an on-the-spot evaluation. As a result, these decisions can be swayed by attentional manipulations. Thus, visual attention may serve as a unifying framework for understanding different biases that occur in the voting booth, such as ballot-order and candidate-name-familiarity effects.

The directive function of episodic memory - using past experiences to guide current behavior - plays a crucial role in human decision-making. Research suggests that children who recall a past good deed are more likely to act prosocially. However, the emergence of this relation remains unexplored in early preschool years. To address this question, 134 French-speaking preschoolers (24-47 months) were recruited and assigned to either a "good deed" condition, in which they actively helped a female accomplice hide a birthday gift, or a "neutral" condition, in which they were asked to watch the accomplice hide the gift. One week later, after their memory of the previously experienced event was assessed, they were put in a situation requiring them to provide help (i.e., the experimenter dropped some cards and had to pick them up). Their prosocial response was recorded. Generalized Linear Analyses revealed that, in the good deed condition, children with more detailed memories were more likely to help the experimenter. This effect was not found in the neutral condition. Interestingly, children also appeared to be less likely and slower to provide help as they grew older, suggesting a developmental shift in their memory-based decisions: younger children rely on fast, automatic decisional processes while older children engage in a slower, more deliberate decisional balance.

Language acquisition and processing rely on a dynamic network of cognitive abilities, where various mechanisms interact to support the recognition, integration, and application of linguistic patterns. Previous research has largely focused on the dual relationships between statistical learning and language abilities, or between core cognitive functions (perceptual speed, working memory, cognitive control) and linguistic abilities, leaving their combined interaction underexplored. To address this gap, this study investigates how statistical learning-a process that enables individuals to detect patterns in language-relates to linguistic abilities and the extent to which core cognitive functions contribute to this relationship. We assessed a large sample of 608 Hungarian speakers (ages 14 to 92) on multiple tasks measuring statistical learning (speech segmentation, artificial grammar learning), linguistic performance (grammatical sensitivity, pragmatic comprehension, semantic prediction, violation processing, and reading efficiency), and core cognitive abilities (perceptual speed, working memory, cognitive control). Structural equation modelling revealed significant small to moderate relationships between statistical learning and language abilities, with offline statistical learning tasks predicting linguistic performance better than online measures (which assess statistical learning in real time). Importantly, core cognitive abilities, especially perceptual speed and working memory, consistently mediated the relationship between statistical learning and language processing, revealing the interconnected dynamics between these functions. These results support the notion that while statistical learning contributes to individual differences in language abilities, its effect is partially explained by core cognitive mechanisms implicated in both statistical learning and language processing. The findings highlight the complexity of language acquisition and processing, and underscore the need for further investigation into the mediating role of other cognitive factors.

In order to maximize reward, humans need to balance engaging with currently available sources of reward and searching for better ones. Optimal foraging theory provides a formal but simple mathematical choice rule to make such stay/leave decisions, contrasting expected and experienced rewards. However, this rule (given by the Marginal Value Theorem; MVT) describes a strategy that does not consider the structure of the environment. In other words, it does not leave room for planning during foraging. Yet, the real world is replete with such opportunities. Therefore, we developed a new structured foraging task to study how people employ goal-directed planning during foraging. Specifically, we explore the extent to which participants incorporate an internal model of the task structure during stay/leave decisions. We find that behavior in this task follows the basic principles of the MVT, but that its structure invites people to also consider the value of alternative reward options when deciding to leave their current one. Importantly, this behavior is pronounced in more goal-directed participants. Computational modeling suggests that incorporating this alternative information is beneficial, but to an extent dictated by choice stochasticity. This study provides a novel method for studying decision making in structured environments, and has implications for understanding how foraging and planning interact.

Humans often struggle to incorporate chance information into performance evaluations. Across diverse samples in China and the United States (total N = 1387), we show that people systematically misperceive or ignore chance-level success rates when judging the efficacy of technological practices. Using scenarios where chance performance is objectively known (e.g., ∼50 % success rate for fetal sex prediction), we find that (1) many participants underestimate the success achievable by random guessing, (2) even when they accurately recognize chance-level information, they often fail to use it as a baseline for evaluating expert predictions, and (3) this "chance neglect" is especially pronounced in performance-related judgments. These findings highlight a cognitive bias that may contribute to the persistence of ineffective technologies across societies.

There is robust evidence that people associate certain sounds with meanings, yet the prevalence and importance of sound symbolism in natural language remains debated. This paper proposes that some failures to detect sound symbolic patterns stem from mischaracterizing how sound symbolism operates. Many studies examine whether words with similar meanings have similar sounds. Yet if words shared sounds with lexical competitors, sound symbolism could hinder processing and not persist. Instead, this paper proposes that sound symbolism highlights features that distinguish referents from their competitors. To test this, the study revisits size sound symbolism in English. Front vowels are associated with small size and back vowels with large size. It has also been suggested that voiceless stops are associated with small size whereas voiced stops with large size. Prior studies found that size adjectives in English are sound symbolic, but the general vocabulary is not. This paper instead analyzes words whose definitions suggest relative size in context-e.g., classifying terrella ("a little Earth") as small, despite its absolute size. Words were extracted from the Oxford English Dictionary if their definitions included a size adjective (e.g., small, large), and retained if language models judged them as referring to objects relatively small or large in context. Results show that words for relatively small referents contain more front over back vowels and more voiceless over voiced stops than words for relatively large referents. These findings suggest that sound symbolism highlights referents' distinctiveness, and that prior conceptual and methodological approaches might have obscured sound symbolic patterns and benefits. This reframing has implications for language evolution, acquisition, and processing, and suggests new directions for future research on iconicity in language.

Even though memory plays a pervasive role in perception, the nature of the memory traces left by past sounds is still largely mysterious. Here, we probed the memory for natural auditory textures. For such stochastic sounds, two types of representations have been put forward: a representation based on sets of temporally local features, or a representation based on time-averaged summary statistics. We synthesized naturalistic texture exemplars and used them in an implicit memory paradigm based on repetition, previously shown to induce rapid learning for artificial sounds such as white noise. Results were similar for artificial and natural sounds, exhibiting a general trend for a decrease in repetition detection performance with increasing exemplar duration, although with some variation depending on texture type. This trend could be captured by a summary statistics model, but also by a new model based on the random sampling of temporally local features. Moreover, repeated exposure to a same natural texture or artificial noise exemplar systematically induced a performance gain, which was comparable across all sound types and exemplar durations. Thus, natural texture exemplars were amenable to learning when repeated exposure was available. The findings are consistent with two interpretations: the existence of a special processing mode when acoustic repetition is involved, to which natural textures are not immune, or a convergence of the local features versus summary statistics descriptions if a continuum of time scales is considered for auditory representations.

People often favour information aligned with their ideological motives. Can our tendency for directional motivated reasoning be overcome with cognitive control? It remains contested whether cognitive control processes, such as cognitive reflection and inhibitory control, are linked to a greater tendency to engage in politically motivated reasoning, as proposed by the "motivated reflection" hypothesis, or can help people overcome it, as suggested by cognitive science research. In this pre-registered study (N = 504 UK participants rating n = 4963 news messages), we first provide evidence for motivated reasoning on multiple political and non-political topics. We then investigated the associations of the two cognitive control variables cognitive reflection and inhibitory control with motivated reasoning. We find that associations between cognitive control processes and motivated reasoning are likely small. On political topics specifically, we find that a negative association with cognitive reflection is more likely than a positive association. This finding is contrary to predictions from the popular motivated reflection hypothesis. Results for inhibitory control are inconclusive. We discuss how these findings relate to interdisciplinary literature from cognitive and political psychology.

A Sense of Agency (SoA) is the feeling of being in control over own actions and their outcomes. However, people can also experience a "vicarious" SoA over the actions performed by other agents, including artificial agents. The present study aimed to understand the minimal conditions for vicarious SoA toward artificial agents. Specifically, we addressed whether vicarious SoA emerges when people have access only to the action effect (proximal and distal), i.e., when no motor action is executed. In addition, we manipulated the expectancy of the content of the distal effect of the action to check whether the proximal action effect is sufficient for the emergence of the vicarious SoA, or if this effect is due to the learned association between proximal and distal effects. In two experiments, participants performed an Intentional Binding (IB) task, where the IB effect was the behavioural measure of SoA. In the first experiment (Solo), participants judged the onset of self-generated tones, whereas in the second experiment, a new sample of participants judged the onset of tones produced by a computer via an automatically pressed button, i.e., a customized device designed to generate a keypress (proximal action effect) in the absence of an effector executing a keypress (no motor action). In both experiments, participants' neural activity was recorded via electroencephalography (EEG), to examine the N1 and P2 components as neural measures of SoA. Behavioural results across experiments showed that the IB effect always emerged, suggesting that the vicarious IB effect toward an artificial agent emerges when access to the proximal action effect is provided, even in the absence of the action itself. The neural results suggested that while individual (self) SoA seemed to partially rely on motor predictions indexed by the N1, vicarious SoA relies on later, more cognitive (although still predictive) processes indexed by the P2. Overall, these results suggest that individual and vicarious SoA, although behaviourally manifested through a similar IB effect, might - to some extent - rely on different neural mechanisms.

Effective attentional selection requires filtering task-irrelevant stimuli. We examined the cognitive cost of such filtering using an object-based attention paradigm across four experiments (N = 320). Participants discriminated the orientation of a Gabor patch presented either alone or overlapped with an irrelevant object or scrambled stimulus. The filtering cost was measured as an increase in response times on distractor-absent trials embedded in 'mixed' blocks (with interleaved distractor-present trials) compared to 'pure' distractor-absent blocks. The filtering cost was robust and scaled with distractor probability and with the presence of one, two or four possible distractors occurring within the session. The cost disappeared when eight distractors were interleaved randomly, but re-emerged when the same eight distractors were presented orderly, one in each sequential mini-block, indicating a strategy shift once working-memory capacity is exceeded. The cost correlated negatively with interference on distractor-present trials and was unaffected by distractor semantic content, consistent with the active maintenance in working memory of distractor templates.

Our experience of time is often distorted in striking ways. Although prior work has shown that boundaries between events can shape temporal perception and memory, less is known about whether, or how, the structure within an event can affect our experience of time as it unfolds. In this study, we asked whether an event's internal structure (i.e., its beginning, middle, and end) systematically biases representations of duration. Across four experiments, participants listened to sequences of tones and either reproduced or judged their durations. We consistently found robust and systematic biases in subjective time: Beginnings were compressed, endings were expanded, and subjective time lengthened progressively over the course of the sequence. These results reveal a distortion in temporal experience that arises not from the transitions between events, but from the way we parse and organize time within them.

Perceptual confidence describes the degree to which we believe our internal perceptual representations reflect the external stimuli that caused them. How confidence is derived from internal representations is currently debated, but answering this question is made difficult because the nature of internal representations is also an ongoing debate. Here we leverage recent findings regarding how global motion is internally represented to test the prominent hypothesis that confidence is derived from internal representations. Participants estimated the global motion direction of two types of random-dot stimuli. In Experiment 1 a stimulus was used which induces discrete, all-or-none segmentation of signal and noise dots. Confidence nearly perfectly tracked whether or not these discrete representations existed, but had no relationship with their precision despite some representations being more precise than others. In Experiment 2 a motion stimulus was used which induces continuous local motion pooling characterized by large differences in response precision. Although confidence tracked the precision of these continuous representations, the relationship between confidence and precision was much weaker, suggesting that meta-knowledge of precision is highly error prone. This dissociation was also present when both discrete and continuous motion stimulus were presented intermixed in Experiment 3, suggesting that it is not merely driven by task set. Instead, the way in which confidence is derived from motion representations depends on whether they arose from a discrete or continuous encoding process. These results support theories which posit that confidence is generally derived from internal perceptual representations, but has only limited access to the precision of those representations.

Prediction in language processing is a fundamental cognitive function that supports efficient comprehension and interaction by anticipating upcoming linguistic input from multiple cues, including semantic (meaning-based) and phonological (sound-based) information. This study examined the interaction between semantic and phonological pre-activation during predictive language processing using the visual world paradigm. Across seven experiments with Spanish speakers, we assessed predictive eye movements toward semantic and phonological competitors, both when these competitors were presented independently and when they directly competed, with two experiments specifically targeting bottom-up integration to evaluate the phono-semantic competition effects. Our results showed that predictive looks toward semantic competitors preceded the availability of phonological information. Moreover, when both cues were presented simultaneously, semantic predictions were prioritized during the anticipatory period, whereas phonological activation emerged primarily after the onset of the target word. Importantly, the bottom-up phonological effect occurred earlier in highly constraining sentence contexts than in neutral or absent contexts, indicating that semantic contexts facilitated phonological processing. Finally, the last two experiments demonstrated predictive fixations toward competitors phonologically related to the semantic competitor, suggesting that participants generated parallel predictions encompassing both semantic and phonological forms. Taken together, these findings support the framework of hierarchical predictive processing, in which semantic predictions guide and shape phonological predictions. The observed flexibility and context sensitivity of this system underscore the complexity of language comprehension, revealing that predictions are dynamically integrated and prioritized across multiple levels of linguistic representation.