Height fear might involve dysfunctional, implicit biases in attention and avoidance in the presence of height-related stimuli. The present study used an Approach-Avoidance Task (AAT) for fear of heights to investigate the association between height fear and alterations in attention and approach-avoidance tendencies. The AAT for height-related stimuli assessed individuals' response times when pulling or pushing height-related vs. control images. Self-reported and interview-based measures of height fear were used to measure associations with selective attention and implicit avoidance in a height-related AAT. Self-reported height avoidance was associated with slower responses to height-related images relative to control images, suggesting changes in selective attention for height-related stimuli. A similar pattern of findings was found when using interview-based measures of height fear. We did not find associations between subjective measures of height fear and implicit avoidance bias in the AAT. Our results provide initial support for alterations in selective attention for height-related stimuli in height-fearful individuals. While the AAT may be effective in identifying biased attentional processing, further research is required to draw conclusions about potential avoidance biases in the AAT related to height-fear.

We previously proposed a Vertical Attention Bias (VAB) that directs attention toward object tops and scene bottoms and robustly confirmed this effect in both adults and 4- to 7-year-old children. Our past findings are consistent with progressive ecological theory, and support that our perceptual biases are coupled to informative environmental regularities. This leads observers to generally favor a downward gaze to facilitate attending more to functionally and behaviorally relevant locations. Here, we examine orientation effects using upright or inverted images presented in triptych sets to further test the overall VAB pattern. Participants made similarity judgments between a central target image of an object or scene and flanking images containing either the same top-half or the same bottom-half as the target image. Experiment 1 presented upright triptych images and replicated past VAB findings. Experiment 2 presented the same triptychs in an inverted orientation. In this context, the environmental regularity of interactive feature placement is incongruent with conventional spatial location in the presented image. Here object and scene tops are positioned in the lower image portion, and bottoms in the upper image portion. Results extend previous findings and confirm that VAB effects favoring object tops and scene bottoms flip along with the inverted image, though statistically weaker. Taken together, the findings support that the typical vertical interactive feature imbalance in real-world stimuli drives a generic downward vantage tendency. This directs attention toward the locations of meaningful, behaviorally relevant environmental aspects, which helps focus attention on personal action space and body-level affordances.

Under high-pressure situations, such as crucial games, some athletes often underperform. This is the case even for exceptional athletes in critical moments of competition. Athletes often experience performance anxiety, which creates attentional errors and underperformance. Attentional bias, where emotional stimuli influence decision-making, may also result in selective attention to negative information. Attentional bias correction training (ABMT) aims to modify these attention patterns with the aim of alleviating anxiety symptoms. In this study, we investigated the pre-competition attention patterns among high-level athletes to enhance performance. Attentional bias correction training was employed to train 32 athletes for 4 weeks, every 2 days (emotional pictures for the experimental group and star pictures for the control group). A point detection paradigm was also utilized to test the athletes' attentional bias behavior in a pre-competition anxiety situation before and after training. The results revealed a significant main effect of test time on self-rated anxiety level of high-level athletes (F(1, 32) = 204.072, p < 0.001, η2p = 0.919), while the main effect of group was not significant (F(1, 32) = 0.505, p > 0.05, η2p = 0.025). Moreover, significant interaction was recorded between the test time and group (F(1, 32) = 124.895, p < 0.001, η2p = 0.874). In the simple effects analysis, high-level athletes in the experimental group had significantly lower pre-competition anxiety scores at the post-test than at the pre-test (43.90 ± 2.57 vs. 57.00 ± 2.77, p < 0.001). However, for the control group, there was no significant difference between the two groups (52.20 ± 2.57 vs. 53.80 ± 2.77, p = 0.133). In addition, the attentional bias scores were significantly different before and after the training (5.985 ± 1.045 vs. -0.613 ± 0.60, p < 0.05) in the experimental group, whereas there were no significant differences in the control group (7.813 ± 1.045 vs. 5.773 ± 0.613, p > 0.05). The present findings demonstrate that attentional bias correction training can effectively reduce pre-match anxiety and attentional bias toward negative information in high-level athletes. These results provide an important foundation for enhancing pre-competition attention training and mood regulation in athletes. Future research should explore the underlying mechanisms and practical applications of these findings to facilitate the development of strategies to improve pre-match attention training.

Using the dot-probe paradigm, previous research has demonstrated that women on average show attentional biases towards underweight bodies. However, little research has used these paradigms to examine the malleability of such biases. Here, we examined whether a single session of attention bias modification training, in which participants were trained to attend to healthy-weight bodies, reduced attentional orientation towards underweight bodies and improved body satisfaction. One hundred and twenty-one female participants were randomly assigned to either an experimental group in which they were trained to attend to healthy weight bodies or a control group (with no manipulation). Participants' body satisfaction was measured at two phases, before and following attentional training. We found no changes to attentional biases or body satisfaction across both groups. Dot-probe attention bias modification tasks may not be able to modify body satisfaction and attention biases towards healthy-weight bodies following a single training session. Future research is encouraged to consider alternative attentional modification paradigms to modify pathological body image.

Aberrant cognition is common among individuals at familial risk for mood disorders (MD) and those already affected. However, long-term prospective studies are needed to determine whether specific cognitive features predict illness onset and relapse; and whether cognitive impairments reflect neurodevelopmental traits or neuroprogressive decline. This seven-year prospective study examined the relationship between cognition and illness progression in monozygotic twins with mood disorders, their healthy high-risk monozygotic co-twins, and low-risk twins without a family history. Emotional and non-emotional cognition was assessed at baseline (n = 204) and follow-up (n = 124). Cox regression models tested whether baseline cognition predicted future illness onset in unaffected individuals (n = 89) or relapse in affected ones (n = 112). Longitudinal cognitive changes were analyzed using mixed models. Greater attentional vigilance toward consciously processed happy faces at baseline was associated with a reduced risk of both illness onset (Exp(B) = 0.995, CI [0.990; 1.000], p = 0.03) and relapse (Exp(B) = 0.997, CI [0.995; 0.999], p = 0.003). Paradoxically, better verbal fluency at baseline was linked to an increased risk of illness onset (Exp(B) = 1.589, CI [1.204; 2.097], p < 0.001). Over time, onset was associated with increasing avoidance of subliminal fearful faces (group-by-time interaction, p < 0.001), whereas avoidance decreased in those who remained well. Verbal fluency declined in twins who developed a mood disorder (p = 0.02) but remained stable in those who stayed unaffected. No significant longitudinal cognitive differences emerged between affected twins with and without relapse. Positive attentional biases may protect against illness onset and relapse, while greater baseline verbal fluency may unexpectedly signal vulnerability. Verbal fluency decline after illness onset likely reflects scar effects. The findings underscore the importance of early identification of cognitive-emotional vulnerabilities and suggest targets for preventive interventions.

Numerical magnitude can bias spatial attention, typically facilitating faster responses to the left for small numbers and to the right for large numbers—an effect traditionally attributed to egocentric spatial mappings. However, in everyday environments, individuals often share space with others, raising the question of whether such spatial–numerical associations can spontaneously reorganize based on another person’s visual perspective. To investigate this, we employed a digit-primed visual detection paradigm in which participants judged the location (left, right, up, or down) of a briefly presented peripheral probe following centrally displayed digits. If numerical magnitude implicitly guides attention, probe detection should be faster when its location is congruent with the digit-induced spatial bias. Critically, in the avatar condition, a task-irrelevant avatar was positioned on the participant’s left side, such that the avatar’s horizontal (left–right) axis corresponded to the participant’s vertical (up–down) axis—an axis along which egocentric numerical biases are typically absent. If participants spontaneously adopted the avatar’s perspective, numerical cues might induce attentional biases along this axis. Results revealed two simultaneous effects: a canonical egocentric SNARC-like effect (small–left, large–right) and a novel allocentric effect (small–up, large–down) emerged along the vertical axis, implicitly aligned with the avatar’s left–right spatial orientation. Numerical extremity enhanced the egocentric SNARC-like effect but had no effect in the allocentric case, pointing to a distinct mechanism rooted in embodied spatial perspective. These findings suggest that numerical magnitude can implicitly map onto both egocentric and allocentric spatial frames, reflecting a implicit and embodied mechanism of social understanding.

Predicting anxiety symptoms through gaze-directed attention: A mobile eye-tracking study of adolescents during a real-world speech task.

Introduction: Stuttering encompasses complex cognitive, behavioral, and emotional mechanisms that interact dynamically. Clarifying the multidimensional nature of stuttering is fundamental to both its conceptual understanding and the advancement of evidence-based treatment. This study aimed to investigate cognitive and emotional interference in adults who stutter (AWS) and explore its relationship with anxiety levels, stuttering frequency, and the psychosocial impact of stuttering. Methods: Fifty AWS and fifty age- and sex-matched fluent controls participated in this study. Data were collected using the Stroop TBAG test, the Emotional Stroop test, the Wright and Ayre Stuttering Self-Rating Profile (WASSP-TR), the State-Trait Anxiety Inventory (STAI), and stuttering frequency measures. Descriptive, comparative, and correlational analyses were conducted using SPSS version 25. Results: The AWS demonstrated significantly longer completion times on the Stroop task than fluent speakers. In the Emotional Stroop task, they also exhibited prolonged reaction times to stuttering-related and threat-related words, while reaction times to neutral words did not differ significantly between the groups. State anxiety was significantly associated with both attention measures, whereas stuttering frequency was correlated specifically with selective attention. Conclusions: The findings indicate that AWS differ significantly from fluent controls in their performance on tasks requiring selective attention and exhibit an attentional bias toward emotionally salient stimuli. These results underscore the relevance of cognitive and emotional processes in stuttering and support the integration of anxiety- and attention-related dimensions into clinical assessment and therapy planning.

The remarkable success of autoregressive Large Language Models (LLMs) is predicated on the causal attention mechanism, which enforces a static and rigid form of informational asymmetry by permitting each token to attend only to its predecessors. While effective for sequential generation, this hard-coded unidirectional constraint fails to capture the more complex, dynamic, and nonlinear dependencies inherent in sophisticated reasoning, logical inference, and discourse. In this paper, we challenge this paradigm by introducing Dynamic Asymmetric Attention (DAA), a novel mechanism that replaces the static causal mask with a learnable context-aware guidance module. DAA dynamically generates a continuous-valued attention bias for each query–key pair, effectively learning a “soft” information flow policy that guides rather than merely restricts the model’s focus. Trained end-to-end, our DAA-augmented models demonstrate significant performance gains on a suite of benchmarks, including improvements in perplexity on language modeling and notable accuracy boosts on complex reasoning tasks such as code generation (HumanEval) and mathematical problem-solving (GSM8k). Crucially, DAA provides a new lens for model interpretability. By visualizing the learned asymmetric attention patterns, it is possible to uncover the implicit information flow graphs that the model constructs during inference. These visualizations reveal how the model dynamically prioritizes evidence and forges directed logical links in chain-of-thought reasoning, making its decision-making process more transparent. Our work demonstrates that transitioning from a static hard-wired asymmetry to a learned and dynamic one not only enhances model performance but also paves the way for a new class of more capable and profoundly more explainable LLMs.

ABSTRACT Objective Methamphetamine (MA) addiction is associated with significant cognitive impairments, particularly in attentional bias and inhibitory control. These deficits complicate recovery and increase the likelihood of relapsing. This study aims to evaluate the test-retest validity and reliability of a modified Go/No-Go task designed to measure these cognitive deficits in MA-dependent individuals. Methods Forty MA-dependent individuals and fifty healthy controls participated in a two-session experiment. Both groups completed a MA-related Go/No-Go task, comprising blocks of neutral, negative, and MA-related cues. Efficiency Scores (ES) were calculated based on omission and commission errors, normalized by reaction time. Test-retest reliability was assessed through Pearson correlation coefficients, and differences between groups were analyzed using independent and paired t-tests. Results Significant differences in ES were observed between MA-dependent individuals and controls for MA-related and neutral cues, but not for negative cues. Strong correlations in test-retest ES were noted in both groups, particularly for MA-related cues in MA addicts. However, the task showed limited sensitivity in distinguishing between cue types within the addicted group. Conclusion The MA related Go/No-Go task shows moderate to strong test-retest reliability but limited sensitivity in distinguishing cognitive responses across cue types in MA addicts.

The integration of artificial intelligence (AI) into healthcare marks a paradigmatic shift, raising not only technical and organizational challenges but also profound ethical questions. This theoretical-reflective paper explores the ethical dimensions of AI from a nursing perspective, focusing on the role of nurse leaders as ethical agents in the digital transformation of care. Drawing on philosophical and ethical frameworks, such as Heidegger's notion of humans as "sensors of machines" and the concept of algorethics, the article argues that AI should not replace the relational, interpretive, and moral dimensions of care, but rather support them, provided its implementation is guided by ethical vigilance. The World Health Organization's principles-autonomy, beneficence, transparency, accountability, justice, and sustainability-serve as the ethical foundation for AI in healthcare, highlighting the need to preserve human dignity and prevent discrimination. Yet algorithmic opacity, automation bias, and data-driven inequities demand renewed attention to responsibility, explainability, and the preservation of human agency. Nurse leaders are called to act along three ethical pillars: fostering transparency in algorithmic processes, ensuring shared responsibility for AI-supported decisions, and sustaining human relationships as the core of care. By reclaiming relational time (kairos) and resisting the delegation of moral judgment to machines, nurses safeguard the human face of care. In doing so, they embody a leadership rooted in presence, justice, and ethical discernment. Ultimately, this paper calls for a proactive governance of AI that anticipates its impacts and ensures its alignment with the telos of nursing: to care for persons with competence, conscience, and compassion.

Trait rumination is a repetitive and often maladaptive attentional focus on the consequences of depression. Rumination independently contributes to cognitive control dysfunction associated with depression. However, it is not clear how the effects of rumination on cognitive control may contribute to negative attention biases as well, or whether it is specific to brooding or reflective rumination. To address these questions, the current study examined the link between trait rumination, cognitive control, and attentional biases. Participants were given a task to remember three neutral faces across a delay period with a single irrelevant sad, happy, or scrambled face distractor. Memory accuracy was also collected. Additionally, the amplitude of the emotion processing late positive potential (LPP) component was recorded by electroencephalograph (EEG) in response to distractors. Brooding and reflection were not associated with memory accuracy. Brooding was not significantly related to LPP amplitudes. A significant emotion by reflection interaction on LPP amplitudes was observed. As the reflection levels increased, the LPP amplitudes for sad faces decreased relative to amplitudes for scrambled faces. The effects were maintained while controlling for brooding and depression. The results suggest that reflection may bias attention toward control over negative distraction but not improve accuracy and, thus, may contribute to cognitive inefficiency associated with depression.

Somatosensory amplification (SSA) is a cognitive-emotional construct defined as the tendency to perceive normal physiological sensations as unusually intense, disturbing, and threatening. Originally introduced by Barsky et al., SSA involves heightened attention to bodily sensations, catastrophic interpretations, and exaggerated emotional responses. This tendency is closely linked with various psychiatric disorders, including anxiety disorders, somatic symptom disorder, panic disorder, and health anxiety. SSA is also conceptually related to constructs such as interoception, somatization, and bodily awareness, although it remains distinct in its emphasis on cognitive-affective distortion. Neuroimaging studies suggest that SSA is associated with hyperactivation in brain regions responsible for somatic awareness and emotion regulation, such as the insula and anterior cingulate cortex. The SSA scale is the most commonly used tool to assess SSA, although complementary methods like heartbeat detection tasks and attentional bias paradigms are also employed. Therapeutically, cognitive-behavioral therapy, mindfulness-based interventions, and attention training programs have shown promise in reducing SSA levels. This review emphasizes the importance of SSA as a transdiagnostic mechanism that bridges bodily, emotional, and cognitive domains, with implications for both research and clinical practice. Future studies should further investigate SSA&amp;rsquo;s neurobiological basis and develop targeted interventions to address this amplifying tendency in diverse clinical populations.

The impact of perceived social support on children’s problem behaviors: the parallel mediating roles of attentional bias

Longitudinal Relationships Between Childhood Emotional Maltreatment and Depression in Chinese University Students: The Roles of Attention Bias and Expressive Suppression

Psychological distress is understood to be maintained by attention. We performed two experiments examining the impact of attention training (AT) on psychological distress symptoms. Experiment one (N = 336) investigated what effects might be detected in a simple experimental design with longitudinal measurements, while experiment two (N = 214) examined whether using a different emotional stimulus could induce an immediate anxiolytic effect in response to AT. Attentional biases were operationalized as the target search latency correlated with mood and psychological distress scores. While limited evidence of attentional biases was found in participants with higher mood distress, correlations emerged in the experimental conditions at day thirty, indicating a relationship between task latency, stress, and changes in depression (experimental one). We found no immediate between–within-group differences in outcome when including different emotional stimuli (experiment two). Despite attentional biases being less apparent in community samples, attentional training for bias modification was effective in eliciting positive biases, leading to improved mood. Notably, participants in the control condition reported the greatest mood and psychological distress improvements, whereas changes in the experimental condition primarily pertained to attentional biases. Taken together, these findings suggest that AT tasks can improve distress, but not through changes in attentional biases.

Investigating attentional bias in low-dependence smokers with steady-state visual evoked potentials (SSVEPs): Linking wanting, liking, and neural responses.

This study examines the impact of dialectal migration on cognitive processing and aesthetic tendencies across three major Shandong dialect groups: Heze, Jinan, and Jiaodong. The research highlights the interrelation between language evolution, ethnic identity, and sustainable cultural heritage in shaping memory, attention, and aesthetic preferences. Using three quasi-experiments, the study explores memory recall, attentional biases, and aesthetic evaluations. Results show enhanced memory and attentional biases toward one’s ethnic group across all dialect groups, with Heze speakers displaying a significant affinity toward Henan people, influencing their aesthetic preferences. These findings emphasize the importance of language evolution and cultural identity in fostering sustainable development through the preservation of cognitive and aesthetic diversity. Future research should extend to other regions and ethnic groups to validate and expand these findings, thereby contributing to the discourse on cultural sustainability.

Attention mechanisms of social anxiety in daily life: Unique effects of negative self-focused attention on post-event processing.

Research suggests that presenting an action via multimodal stimulation (verbal and visual) enhances its perception. To highlight this, in most studies, assertive instructions are generally presented before the occurrence of the visual subevent(s). However, verbal instructions need not always be assertive; they can also include negation to contrast the present event with a prior one, thereby facilitating processing—a phenomenon known as contextual facilitation. In our study, we investigated whether using negation to guide an action sequence facilitates action perception, particularly when two consecutive subactions contrast with each other. Stimuli from previous studies on action demonstration were used to create (non)contrastive actions, that is, a ball following noncontrastive and identical (Over–Over or Under–Under) versus contrastive and opposite paths (Over–Under or Under–Over) before terminating at a goal location. In Experiment 1, either an assertive or a negative instruction was provided as verbal guidance before onset of each path. Analyzing data from 35 participants, we found that, whereas assertive instructions facilitate overall action recall, negating the later path for contrastive actions is equally facilitative. Given that action goal is the most salient aspect in event memory due to goal‐path bias in attention, a second experiment was conducted to test the effect of multimodal synchrony on goal attention and action memory. Experiment 2 revealed that when instructions overlap with actions, they become more tailored—assertive instructions effectively guide noncontrastive actions, while assertive–negative instruction particularly guides contrastive actions. Both studies suggest that increased attention to the goal leads to coarser perception of midevents, with action‐instruction synchrony modulating goal bias in real‐time event apprehension to serve distinct purposes for action conceptualization. Whereas presenting instructions before subactions attenuates goal attention, overlapping instructions increase goal attention and reveal the selective roles of assertive and negative instructions in guiding contrastive and noncontrastive actions.