Communications Psychology

Habits in humans are commonly studied through outcome devaluation paradigms, but most existing tasks fail to capture the robustness of habitual behavior seen in animal models. I introduce two novel behavioral tasks designed to overcome these limitations. In the first task, ("shooting aliens task", n = 45), I simplified an existing instrumental learning task and implemented a novel intra-block reversal method in which stimulus positions changed unexpectedly within blocks while maintaining the same stimulus-action mappings. Participants also completed a classical devaluation phase with explicit reward changes. In the second task ("hands-attack task", n = 44), which relied on real-life avoidance behavior, devaluation was achieved by reversing reward contingencies and allowing participants to inhibit the dominant avoidance response in favor of a more effortful counterattack. Across both tasks, overtrained conditions led to more errors and longer response times after devaluation, confirming increased insensitivity to outcome change. Intra-block reversals in the shooting aliens task produced stronger habitual signatures than standard whole-block devaluation, revealing a greater cost of overriding automatic responses. In the hands-attack task, even without prior training, participants showed clear markers of habitual behavior, suggesting that real-world action patterns can replicate key features of laboratory habits. Interestingly, participants were more accurate in overriding overtrained responses when attacks were highly familiar, possibly due to enhanced perceptual processing, although this came at the cost of longer response times. These findings introduce two complementary tools that address key limitations in current paradigms: the intra-block reversal increases habit sensitivity without inflating working memory demands, while the hands-attack task captures naturalistic habit expression without artificial training, using a single, ecologically valid session. Both are suited for clinical applications, particularly where time constraints or cognitive load limit the feasibility of traditional approaches.

We propose a computational instantiation of three cognitive stages from the Dot-Linear- Network (DLN) framework, grounded in a compression-efficiency thesis. DLN stages are characterized as graph-structured belief-dependency representations used to evaluate options: Dot as no persistent belief graph (reactive policies with negligible internal state), Linear as a null graph over option beliefs (K independent option estimates with no information sharing), and Network as shared latent structure (a bipartite factor graph in which F latent factors connect to K options), augmented by a temporal exposure state and an explicit structural learning cycle (hypothesis [->] test [->] update/expand). We distinguish two compression targets--option-factor structure (shared components in expected outcomes) and stakes-factor structure (shared drivers of consequence-bearing exposures)-- whose intersection yields jointly efficient actions that simultaneously improve expected outcomes and marginal exposure impact. In a bandit-like simulation (100 seeds, K [isin] { 20, 50, 100, 200 }, F =5), Network policies dominate Linear policies in cost-adjusted utility at large K, with the empirical crossover occurring much earlier than an analytic cost-only prediction (K* = F + cmeta/cparam), revealing that the advantage is primarily statistical (shrinkage-like estimation gains from factor pooling) rather than purely computational. Under stakes, all non-DLN agents--including Linear-Plus agents with identical factor structure and Network-standard agents with hierarchical Bayesian learning--collapse due to unmodeled cumulative exposure, while Network-DLN maintains positive utility. Within-stage consistency tests (two algorithmically distinct agents per stage) confirm that the collapse pattern is determined by representational topology, not algorithmic choice. These results evaluate internal consistency of a DLN-to-computation mapping under explicit assumptions; they do not validate a developmental theory in humans.

Suddenly finding the solution to a problem after a period of impasse often comes with a feeling of insight. This subjective experience is proposed to arise as a consequence of prediction errors. Accordingly, previous studies have revealed that more incorrect initial predictions result in more intense insights. Crucially however, prominent models of Bayesian inference suggest levels of computationally-defined surprise are not a simple feature of distance between predictions and inputs, but also their precision or certainty. Yet, how these two factors interact to give rise to insight experiences remains unknown. In this pre-registered study, participants were exposed to ambiguous images while they tried to guess the correct label of the image (to derive prediction accuracy) and rated their confidence in that label (for prediction uncertainty). We then measured the intensity of their insight when a solution was given. As predicted, we found that the intensity of insight was a result of both the prediction accuracy and the uncertainty awarded to it. More specifically, when initial predictions were far from the true label, those made with lower confidence induced weaker insights, while the opposite pattern was observed when predictions were closer to the reality. Trial-by-trial estimations of prediction errors from participants responses closely mirrored insight ratings. Finally, we analysed data from two additional independent datasets with different modalities and setups and replicated the interaction between prediction accuracy and uncertainty on the intensity of insight. Altogether, these findings suggest that insight experiences are read out from prediction errors and highlight the key role of uncertainty in characterising this relationship.

Humans recognize everyday actions without conscious effort despite challenges such as poor viewing conditions and visual similarity between actions. Yet the visual features contributing to action recognition remain unclear. To address this, we combined semantic modelling and feature reduction methods to identify critical features for recognizing actions from challenging egocentric perspectives. We first identified egocentric action videos from home environments that a motion-focused action classification network could correctly classify (Easy videos) or not (Hard videos). In Experiment 1, participants (N=136) labelled the action and object in the videos. Using a language model framework, we derived human ground truth labels for each video and quantified its recognition consistency based on semantic similarity. Participants recognized actions and objects in Easy videos more consistently than in Hard videos. In Experiment 2, we recursively reduced the Easy and Hard videos with high recognition consistency to extract minimal recognizable configurations (MIRCs), in which any further spatial or temporal reductions disrupted recognition. The data was collected using a large-scale online study (N=4360). We extracted information related to the hand, objects, scene background and visual features (e.g., orientation or motion signals) from the 474 MIRCs. Binary classification showed that recognition was disrupted when regions containing the manipulated object and strong orientation signals were removed, while temporal reduction by frame-scrambling disrupted recognition in 73% of MIRCs. The active hand had some marginal contribution. Our results highlight the importance of both mid- and high-level information for egocentric action recognition and link hierarchical feature theories with naturalistic human perception.

Older adults rely increasingly on prior knowledge to make sense of their deteriorating representation of the visual world, but how this shapes scene perception and spatial reorientation remains unclear. To address this issue, 28 young and 25 older adults viewed artificially generated rooms either before or after learning the position of a goal hidden in an adjacent room. We manipulated both the number and the eccentricity of navigational affordances (i.e., open doors) to investigate the interaction between bottom-up scene features and top-down spatial knowledge. Consistent with previous findings, younger adults showed decreased performance as the number of open doors increased, but only after learning the goals position, indicating a top-down interaction with the automatic processing of affordances. Door eccentricity did not affect this interaction, suggesting our findings were not due to a distractor effect. In older adults, this interaction between prior spatial information and navigational affordances was markedly amplified: reaction times increased at twice the rate observed in younger adults. These findings show that prior spatial knowledge interacts with the automatic extraction of navigational affordances, and that this influence is markedly amplified with age. While prior knowledge helps stabilize perception when sensory processing becomes less reliable, it can also increase the processing time for complex scenes, particularly when multiple action possibilities are present. By revealing how aging shifts the balance between top-down and bottom-up mechanisms, these results refine models of age-related spatial navigation decline and highlight a trade-off whereby increased reliance on prior knowledge supports perception but can also slow interaction with complex environments.

The ability to learn broad concepts from individual instances is relevant throughout our lifespans as new concepts enter the world, and we seek to acquire new skills and hobbies that can enrich our lives. While older age has been associated with declines in the ability to remember individual instances, less is known about how these declines impact concept learning and generalization or the neural systems that older adults recruit to support abstraction. In the present study, we used prototype-based category learning as a domain to test age differences in concept learning. Young and older adults completed a category-learning task while undergoing fMRI. We fit formal prototype and exemplar models to behavioral and brain data to index concept learning based on abstraction versus memory for individual category members. We found that the fit of both models to behavior was poorer in older adults, but older adults were more likely than young adults to be best fit by the prototype model and less likely to be best fit by the exemplar model. While only young adults showed significant prototype-tracking in the hippocampus, both young and older adults recruited the ventromedial prefrontal cortex (VMPFC) to support prototype-based generalization. Although evidence for age differences in prototype representations emerged in a whole-brain analysis, evidence for age differences were weak in the VMPFC and hippocampus. Thus, engagement of the VMPFC prototype-learning system may help maintain concept generalization in older adults.

Memory generalization allows individuals to extract and apply information from prior experiences to novel situations, supporting flexible learning and efficient decision-making. Theoretical models suggest that sleep should facilitate generalization, yet the literature examining its role in promoting generalization is mixed. We recruited 137 participants via Prolific to complete an image-location memory task over two sessions spaced 12 hours apart. Participants were randomly assigned to the Wake group (learning in the morning) or the Sleep group (learning in the evening). In Session 1, participants learned the location of stimuli on the screen and were tested on their memory five minutes later. Twelve hours later, in Session 2, they were tested on their memory again. Stimuli consisted of 160 images from eight semantic categories and were strategically positioned on-screen to test the effects of generalization on retrieval (i.e., category-based memory distortions and biases). After the delay, retrieval was less accurate and demonstrated more generalization. However, these effects were mostly independent of Group, with some evidence for enhanced generalization following a period of wakefulness over sleep. Generalization was also driven by time of day, with more generalization in the evening relative to the morning. Taken together, our results, based on a large online sample, do not support a role for sleep in promoting memory generalization. Significance StatementBehavior is often guided by memories of previous experiences. However, for behavior to be adaptive and flexible (e.g., when encountering never-before-seen stimuli), regularities about the world must be extracted from these memories. This process, termed memory generalization, has been hypothesized to rely on sleep. We used a large online sample to test sleeps role in generalization and found no support for this hypothesis. Our results suggest that sleep and wakefulness contribute to generalization equally, with the latter potentially having a larger contribution.

The benefits of routines for daily functioning are widely acknowledged, yet, despite their apparent importance, methods for quantifying routine maintenance and the causes of their disruption remain lacking. Here, we propose a novel means of defining and quantifying routines (transition entropy). Using the transition entropy, we show that routines can be robustly elicited on tasks that require searching through a grid of squares for a hidden target. Over two experiments (N=100 each), we show that use of routines--as quantified by transition entropy--is robustly perturbed by frequent switches between search grids, as locations specific to the currently irrelevant grid become competitive for selection. Using a normative model that tracks task dynamics, we show that disruption to routines can be attributed to reduced sensitivity to the odds of success for completing a task. This suggests that routine maintenance may be disrupted by over-sensitivity to a lack of reward early in routine performance, or increased expectations regarding the utility of pursuing other tasks.

Predictive coding has influenced many conceptual accounts of delusions, the bizarre and distressing beliefs that accompany a range of neuropsychiatric conditions. However, these explanations remain incomplete and have rarely been tested directly using formal modelling. Here, we present a formal account of delusional beliefs based on hybrid predictive coding, which sheds light on the computational mechanisms underpinning the core features of delusions: thematic recurrence and imperviousness to contradictory evidence. In simulation experiments, we demonstrate that a combination of contextually inadequate initialisation of beliefs and excessive certainty (a hallmark of psychosis), triggers a reorganisation of the generative model relating observed events to hidden causes. This reorganisation enables the maintenance of delusional beliefs that are thematically stable, internally consistent with external inputs, and impervious to contradictory evidence, all without an increase in prediction error. Overall, our results suggest that delusions may arise not from faulty inference, as previously argued, but as an adaptive consequence of generative models learned under atypical conditions. These findings provide mechanistic insights into the computations underpinning delusions and have important implications for a novel therapeutic strategy in terms of re-training generative models.

Despite several age-related processes impacting motor performance, older adults often retain the ability to implicitly adapt to sensory prediction errors. Here, we leverage the fact that implicit adaptation is not attenuated by aging to study the impact of aging on responses to motor errors. In other domains, such as reinforcement learning, aging has been shown to influence how task outcomes or rewards are processed and used to guide subsequent actions, with some studies emphasizing that older adults react more strongly to a miss than to a hit. We aimed to extend these reinforcement learning findings to the motor domain with two preregistered experiments testing whether missing the target leads to larger implicit adaptation in young and older adults to the same extent. In addition, we compared these results to one reinforcement learning task in the motor domain (Boolean feedback after reaching in the absence of visual feedback) and one in the cognitive domain (reward-based decision making). While we found age-related effects in the cognitive domain, we did not observe a consistent effect of age on the modulation of reaching direction or motor adaptation by task outcomes. These results suggest a domain-specific nature of age-related changes in sensitivity to task outcomes.

Semantic knowledge about concepts and their relationships is central to human cognition. Taxonomic relationships link concepts belonging to the same category (e.g. dog and bear), while thematic relationships connect concepts co-occurring in the same events (e.g. dog and leash). The dual-hub theory proposes that taxonomic relations rely on the anterior temporal lobe (ATL), whereas thematic relations rely on the temporo-parietal cortex (TPC). However, it remains unclear whether taxonomic and thematic representations depend on the concurrent task and how they change with aging. The present fMRI study addressed these gaps by jointly investigating the effects of semantic relationship, task, and age on semantic processing. Young and older adults performed taxonomic and thematic judgments on picture pairs that were taxonomically related, thematically related, or unrelated. Our results do not support the dual-hub theory: TPC and ATL responded to both taxonomic and thematic relationships, with a consistent thematic bias. Moreover, their activity was task-dependent. Semantic control regions flexibly responded to task-relevant semantic relations. Finally, aging was associated with a decline in domain-general and domain-specific semantic control, more bilateral ATL engagement, as well as behavioral and neural shifts from taxonomic towards thematic processing. Increased thematic activity was associated with higher accuracy but slower responses. These findings support accounts of age-related neural dedifferentiation and semanticization: Older adults require increased cognitive resources to maintain accuracy at a high level, but this comes at the cost of efficiency.

Are there individuals who resist episodic memory decline into older age? Analyzing 728,000 memory tests from 80,000 participants with at least 4 assessments, we introduce a simulation-calibrated framework to identify genuine memory stability. Across cohorts and models, [~]10% of adults [≥]70 years showed stable performance over a decade. In an MRI subgroup (n{approx}2,000), stable performers exhibited lower rates of brain atrophy across widespread regions, anchoring cognitive stability in structural brain maintenance. However, stability was often transient rather than trait-like: many individuals followed trajectories with extended plateaus of stable performance punctuated by episodes of accelerated decline. Accordingly, 54% showed at least one period of observed stability, averaging 10 years, whereas only 0.4% upheld stable performance over 24 years under the strictest definition. These findings are consistent with a complex-systems model of cognitive aging in which decline often reflects critical transitions rather than continuous erosion.

Cognitive operations require recently encountered information to remain available beyond the moment of sensory input. However, how such transient representations are accessed, and how they differ from sensory processing and long-term memory, remains unclear. Here, we combine hierarchical drift-diffusion modelling with behavioural manipulations to dissociate the decision processes underlying feature prioritisation in visual working memory. We first reanalysed a previously collected working memory dataset to characterise semantic and perceptual judgements at the level of latent decision processes. Semantic judgements were associated with reduced non-decision time across conditions, indicating faster access to task-relevant information, while advantages in evidence accumulation emerged selectively under higher cognitive demands. Two further experiments manipulated attentional prioritisation using retro-cues and dissociated the effects of interference from mere maintenance. Across manipulations, semantic prioritisation was selectively expressed in pre-accumulation processes and was amplified when representations fell outside the focus of attention or had to be maintained under interference. Together, these results suggest that semantic representations remain more readily accessible than perceptual details when working memory representations fall outside the focus of attention, consistent with a shift towards more abstract, long-term memory-like formats under conditions of limited attentional support.

ObjectiveDemographic corrections (e.g., sex, education, race, ethnicity) are often applied when assessing cognition in adults; however, these corrections have significant limitations (e.g., using years of education does not capture the quality of, or access to, education). It is therefore critical to develop novel assessment options that are less susceptible to demographic factors. This study compared demographic effects on a verbal memory test and a performance-based test of cognition and daily functioning in older adults. Based on prior work, we hypothesized the performance-based tests would be less susceptible to demographic factors than paired associates learning. MethodData from 1326 participants (mean{+/-}SD age=61.9{+/-}10.9 yrs; Female = 1066, 80%) were collected through the MindCrowd electronic cohort, with 79 (6%) non-White, 109 (8.2%) identifying as Hispanic/Latino ethnicity, and 327 (25%) reporting education as less than a college degree. Paired associates learning is a well-established measure of medial temporal lobe-dependent learning and memory through recall of word-pairs, scored as the number of correct word pairs entered out of 36 possible. The performance-based test involved functional upper-extremity movement, specifically transporting beans to target cups in a repeating sequence (a task also shown to be dependent on the medial temporal lobe), scored as the intraindividual variability (standard deviation) in trial time across four consecutive trials. ResultsAs hypothesized, linear regression analysis showed that PAL was significantly affected by sex, education, race (particularly Black/African American), and ethnicity, whereas the performance-based test was affected only by sex and with a much smaller effect size than that of PAL. ConclusionsPerformance-based assessments may be an equitable approach to evaluating cognition without requiring score corrections, particularly for diverse populations.

Human cognition is capacity-limited, requiring strategies to actively structure information. Eye movements offer a natural mechanism for sequential sampling, but whether such sequences organize mnemonic representations is unknown. We developed a working-memory task where color-frequency pairings created a consistent latent ordinal structure to optimally reduce memory load. Across two experiments, gaze patterns spontaneously aligned with this structure. Participants sampled items following this sequence during encoding and covertly replayed them during maintenance. Critically, the expression of this structure depended on cognitive demand. In a 3-item task, high performers showed robust sequential sampling during encoding, whereas lower performers compensated with replay-like revisitation during maintenance. Under higher demand (4 items), encoding-based organization was disrupted, and structured replay emerged primarily during maintenance to support memory. These findings show that eye movements do more than reflect memory; they actively organize it, revealing a flexible, behavioral analogue to neural replay when encoding resources are strained.

Spatial memory is invaluable for most mobile organisms, yet nature of the underlying representations that we employ for spatial memory has been fiercely contested. On the one hand, the presence of place cells in the hippocampus and grid cells in the medial entorhinal cortex appear to support the argument spatial representations may follow Euclidean axioms, termed "the cognitive map hypothesis." On the other hand, decades of behavioral research in humans reveals that spatial memory often shows characteristic distortions, leading to the alternative, cognitive graph hypothesis, to account for this aspect of spatial memory. Importantly, the majority of laboratory studies tend to occur within novel environments in which participants often have only limited exposure and no personal relevance. We were interested in studying large-scale memory across multiple time scales: from a virtual environment (e.g., learned over several minutes) to a college campus (e.g., months to a few years) to a hometown environment (e.g., many years). Across several tasks, we found that participants exhibited systematic distortions in their memory for all of these environments. Likewise, we found significant correlations between performance on several spatial memory tasks (both between participants and within-participant analyses of patterns of errors), thus suggesting that these tasks tap into partially overlapping cognitive representations and supporting their construct validity. Altogether, our findings provide clear evidence for cognitive graph hypothesis and support the construct validity of several spatial memory tasks within large-scale, real-world environments that are learned over the course of several months to years. Public Significance StatementSpatial memory is key for our ability to live independent lives (e.g., patients with Alzheimers disease lose independence, partially due to disorientation in familiar environments). Typical laboratory-based measures of spatial memory use novel environments that may differ in complexity vs. real-world environments (e.g., size, layout, number of landmarks, duration of exploration, personal relevance). We leveraged breakthroughs in technology to study spatial memory across several tasks and temporal scales, from a novel environment navigated over the course of several minutes to a university campus (e.g., months to years) to hometowns (e.g., years to decades). We observed consistent evidence for systematic distortions in spatial memory, which supports the hypothesis that spatial memory is supported by a cognitive graph, thus posing an important challenge to the extremely influential Euclidean, "cognitive map" hypothesis that was awarded a Nobel Prize in 2014.

Inferring the true cause of noise--distinguishing between volatility (environmental change) and stochasticity (outcome randomness)--is essential for learning in noisy environments. While most studies rely on binary outcomes, previous models are designed for continuous outcome and use ad hoc approximations to handle binary data, introducing theoretical inconsistencies and interpretational issues. Here, we develop a normative framework for inferring the causes of noise from binary feedback that remains faithful to the discrete nature of the generative process and underlying statistical structure. First, we establish a generative model using a state space approach tailored for binary outcomes and derive the corresponding hidden Markov model inference procedure. Second, we introduce a computational model combining the hidden Markov model with particle filtering to simultaneously infer volatility and stochasticity from binary outcomes. Third, we validate predictions through a 2x2 probabilistic reversal learning task with human participants, systematically manipulating both noise parameters. Results show that participants adjust their learning rates consistent with model predictions, increasing learning rates under volatile conditions and decreasing them under high stochasticity. Our theoretical and experimental results offer a principled approach for dissociating volatility and stochasticity from binary outcomes, providing insights into learning processes relevant to typical cognition and psychiatric conditions.

Some images are consistently remembered better than others, suggesting that memorability reflects intrinsic image properties. We tested whether within-category distinctiveness underlies this effect. Across three experiments (N = 477), participants categorized indoor scenes previously rated for subjective typicality and then completed recognition memory tests. Typical scenes were categorized faster and more accurately, but were remembered worse and showed a more liberal response bias than atypical scenes. These opposing effects were robust across categories. To link subjective typicality to visual representations, we quantified image distinctiveness using a convolutional neural network (CNN). Across layers, CNN-derived distinctiveness closely tracked human typicality judgments and predicted both categorization speed and memorability, with strongest effects in higher, semantic layers. Critically, the memory advantage for atypical scenes persisted even when most images were atypical, ruling out rarity within the experimental context. Together, the results show that intrinsic scene memorability reflects an images position within a category-specific representational space.

Previous studies suggest that visual mental imagery (VMI) acts as a weaker form of top-down visual perception (VP), with the two becoming more similar as VMI vividness increases. However, this relationship remains ill-defined, and it is unclear precisely how much weaker VMI is relative to VP. Here, we introduce an original probabilistic deep learning approach to quantify vividness at the neural level. Thirty-four participants either imagined or perceived stimuli presented at varying levels of vividness and provided trial-by-trial, picture-based vividness ratings. EEG activity recorded during VP was used to train a convolutional neural network (EEGNet) to predict perceived vividness from eight posterior electrodes located around early visual areas. A leave-one-subject-out cross-validation procedure showed that the model generalised across participants with above-chance accuracy during VP. On VP trials, predictions tracked vividness labels, with reliable interpolation to new vivid labels not included during training. Applied to VMI trials, mean expected VMI vividness remained substantially lower than expected vividness for seen stimuli but slightly higher than baseline, supporting a barely rather than quasi depictive imagery. For 91% of participants, mean expected VMI vividness was also lower than, yet scaled with, mean reported VMI vividness. This framework provides a principled way to quantify and compare VMI and VP on a shared neural-behavioural scale, with implications for studying individual differences and aphantasia.

Digital mental health applications enable high-frequency behavioral monitoring and scalable interventions. Journaling provides a therapeutically grounded and intrinsically engaging activity for many users. AI-based text analysis enables privacy-preserving phenotyping of clinically relevant patterns in naturalistic writing, including emotional distress and behavioral risk (e.g., indicators of intent, planning, or preparatory actions for harm to self or others). We evaluated a mobile journaling platform in an 8-week randomized controlled trial (N = 507) of young adults with mild-to-moderate anxiety and depression symptoms. Journaling produced modest reductions in anxiety relative to controls at the 8-week endpoint and 1-month follow-up (d = 0.16-0.19). Effects were small and did not remain significant after correction for multiple comparisons; complementary Bayesian models nonetheless provided moderate-to-strong directional evidence (90-97%) supporting a modest anxiety reduction. In parallel, behavioral phenotyping analyses showed that high-risk journal entries were more common among younger users (OR = 0.77 per year of age, p = 0.007). Text-based risk signals and self-reported energy exhibited significant circadian variation (e.g., risk probability was highest during late-night and overnight hours). Within-person analyses demonstrated strong short-term persistence in mood and risk states, with calm/relaxed showing the highest persistence and anxious/agitated exhibiting the lowest persistence. High-risk journal entries clustered temporally and were preceded by sustained low valence and energy. Although affective volatility was associated with acute declines within the same affective dimension (pleasantness or energy), it was not associated with escalation to high-risk states. Key behavioral dynamics observed in the trial were replicated in an independent general population dataset (N = 16,630). Collectively, these findings demonstrate that privacy-preserving digital journaling can support scalable longitudinal behavioral phenotyping and real-time risk monitoring while providing modest clinical benefit for anxiety symptoms.