Mark Robert Wilson

Attentional Control Theory: Sport (ACTS) proposes that competitive anxiety arises from appraisals of both the cost and the probability of failure, predicting that an error should be most disruptive when the stakes are already high. We report the first large-scale test of this account in association football, drawing on 1,190,821 StatsBomb open-data event records from 330 professional matches across ten competitions. Three studies examined unforced passing errors. In Study 1 (272,381 passes, 2,202 players), situational pressure and a player’s own immediately preceding error each independently raised error probability: the odds of an error were approximately 21% higher per additional pressure component, and 57% higher following an error than following a success (OR = 1.57). The interaction, however, ran opposite to the ACTS prediction, with pressure narrowing rather than widening the post-error gap (OR = 0.91). Study 2 (54,409 post-error sequences) showed that this elevation was not confined to the player who erred: teammates taking the team’s next pass erred 3.9 percentage points above their own baseline rate, an interpersonal effect roughly half the magnitude of the within-player effect. Study 3 followed chains rather than pairs of consecutive errors. Error probability escalated monotonically with each additional consecutive prior error, from 17.7% following a success to 51.6% at a chain depth of six, and mediation analysis indicated that most of this escalation (91.7%) operated through riskier pass selection rather than degraded execution. Between-player variation in susceptibility was substantially greater for error chains than for success chains, and within every playing position the players with the lowest baseline error rates showed the steepest escalation, an association only partly accounted for by pass selection. Errors in elite football are therefore sequentially dependent, partially contagious, and unevenly damaging across players, suggesting that the cost of an error lies less in the error itself than in what it becomes if allowed to compound.

Freezing of gait (FOG) is one of the most disabling symptoms of Parkinson’s. However, we understand little about lived experiences of FOG. The current study aimed to develop a new conceptual model designed to reflect the broad lived experience of FOG. In Phase 1 we carried out semi-structured interviews with 35 people with Parkinson’s and FOG. Participants described memorable freezes along with factors that they considered to influence, or be influenced by, FOG. We used inductive thematic analysis to identify themes that served as foundation for an initial conceptual model. Phase 2 was a 5-year long iterative process of Patient and Public Involvement and Engagement. Through discussion with over 250 people living with Parkinson’s, the conceptual model generated from Phase 1 was scrutinised and adapted to: i) provide additional context that supports the interpretation of findings from Phase 1 results; and ii) maximise the resonance of interpretations. Collectively, these processes highlighted the critical importance of one’s perceived sense of rhythm. Several themes describe ways in which perceived rhythm can be compromised either through requirements for adapting movement (Changers), distracting attention away from the intended movement (Distractors) or causing problems with gait initiation (Blockers). Other themes describe factors that promote the maintenance (Maintainers) or regeneration (Initiators) of rhythm to avoid FOG or break a freeze and start walking. Results also highlight the importance of Accumulated Experiences of FOG (e.g. altered perceptions of one’s sense of self and identity), and how these can amplify Rhythm Changers, Distractors and Blockers. Though Phase 2, a clear focus emerged on one’s perceived uncertainty about how planned movements will feel and associated shifts in attention towards vigilant monitoring and control of movements. We provide a detailed interpretation of study findings (relating to cognitive, emotional, sensory and motor processes) using a Predictive Processing Framework, followed by descriptions of practical implications for managing FOG and associated Accumulated Experiences.

An X-ray baggage screener plays a critical role in aviation security. However, evidence about which cognitive and perceptual factors are important for effective threat detection during X-ray baggage screening tasks remains fragmented. This review aimed to (1) scope the existing literature examining cognitive and perceptual factors associated with X-ray baggage screening performance, and (2) systematically synthesise evidence linking these factors to X-ray baggage screening performance outcomes. Following PRISMA 2020 guidelines and a preregistered protocol, searches of six databases (2000–2025) yielded 24 eligible studies (n = 11184 participants) including manipulation and observational designs. Cognitive and perceptual processes appear central to effective X-ray baggage screening, particularly under conditions of uncertainty, time pressure, and low target prevalence. Evidence broadly supports theoretical accounts such as Feature Integration, Schema, and Signal Detection Theory, suggesting that performance relies on attentional feature binding, experience-based knowledge structures, and the ability to discriminate signal from noise. Training and experience were consistently associated with improved detection, including generalisation to novel threats, indicating development of flexible perceptual schemas rather than simple memorisation. However, this relationship is not uniform, with some cognitive assessments (e.g. mental rotation, search speed) showing limited or no improvement with experience, highlighting a tension between learned and potentially stable traits. Furthermore, existing “talent identification” tools testing specific cognitive skills are undermined by circularity issues. Therefore, despite intuitive links between cognitive skills, human factors, and X-ray screening performance, the fragmented and heterogeneous evidence base currently limits definitive conclusions and practical application within the aviation security sector.

Background: Autistic people can face remediable barriers and challenges with driving, yet there is no consensus around effective support methods. This review therefore sought to evaluate the implementation and effectiveness of driving support interventions for autistic people.Methods: We conducted a multimethod, participatory synthesis of evidence. Studies were obtained from Medline, Social Policy and Practice, ASSIA, PsycINFO, CINAHL, ERIC, Scopus and Epistemonikos. Data were extracted from those that implemented a driving support intervention with autistic people. No specifications on study design, outcomes, or comparators were made. Risk of bias was appraised using the Cochrane ROB and ROBINS tools (for controlled quantitative studies), as well as the EPPI-Centre (for qualitative studies) and NHLBI (for other uncontrolled studies) quality tools. The application and effectiveness of support interventions were narratively synthesised, while focal delivery components and mechanisms of action were examined using realist synthesis. All analyses were co-produced and developed by autistic adults. Results: Only 17 relevant studies (involving 402 autistic participants) were identified, which utilised varying designs and were generally appraised to be low in quality. The most well-researched form of support was simulation training, however e-learning applications, anti-anxiety medication, listening to music, and cognitive-behavioural programmes were also studied. Although positive results were documented for all interventions, at least in some form, assessments typically consisted of simulated driving tests and various contextual factors were overlooked. Specific delivery components were identified from the effective support programmes, but evidence was lacking with regards to intervention mechanisms. Conclusion: Our participatory syntheses of data highlight that research is urgently required to address the driving barriers and challenges faced by autistic people. Despite promising but inconclusive evidence for simulation training and holistic cognitive-behavioural programmes, widespread limitations are evident in existing studies and rigorous, generalisable intervention assessments are lacking.

Recent advancements in extended reality (XR) and data modelling present new opportunities for adaptive simulation solutions, which can measure and respond to individual neuropsychological states. However, questions remain about the optimal metrics for real-time data capture and the applicability of these solutions for enhancing user experiences. The present research examined a novel form of adaptive XR, called “prediction-based attention computing” (PbAC), which tailors simulations based on computational models of the brain and, thus, the dynamic sensorimotor processes theorised to underpin human perception and learning. Specifically, this study aimed to demonstrate whether PbAC can adaptively capture users’ internal state predictions and modulate associated neuropsychological responses. To test this, we used an XR-based racquetball paradigm, in which participants were tasked with intercepting virtual balls that emerged from different starting locations. For PbAC conditions, in-situ eye tracking data assessments were utilised to index participant’s prior beliefs and manipulate levels of expectedness (i.e., prediction error) on each trial. Various measures of predictive sensorimotor behaviour were then extracted and compared with data from probability-controlled and matched-order control conditions. Results showed that sensorimotor responses were affected by the expectedness of XR stimuli, and that clear, prediction-related biases emerged within PbAC conditions. The novel computing software also provoked marked surprisal responses on trials designed to elicit high levels of prediction error, and these surprisal effects were similar, or even greater than, those in our comparison conditions. Together, the findings provide proof of concept for PbAC and support its development within future research and technology innovations.

Organisations are increasingly implementing extended reality (XR) in education, organisational training, and sports practice. However, the effectiveness of XR compared to conventional training methods, and the potential improvements that additional sensory information, such as haptic feedback, may offer are unclear. This systematic review and meta-analysis compared XR with non-XR training programmes in improving psychomotor skill performance and assessed whether haptic feedback improved programmes’ effectiveness. Eight databases were searched on June 30th, 2025, for randomised controlled trials quantitatively measuring the effectiveness of a psychomotor skill after an XR-based training programme. Sensitivity analyses, publication bias and risk of bias were assessed. Due to non-independence of effect sizes, robust variance meta-regression models were run and presented using forest plots. Seventy-eight reports were assessed, totalling 3546 participants. Both the post-training and pre-post-training change scores between XR and non-XR groups were assessed. Two RVEMR models were created by splitting the control groups into two types: passive (no training) and active (non-XR training). All models found statistically significant positive effects of XR training over non-XR training on task performance. No variables were found to be consistently associated with task performance, although haptic feedback showed a consistent, non-significant negative effect on task performance. Evidence certainty was limited by high heterogeneity, low-moderate risk of bias, and poor methods reporting. Overall, XR-based training generally appears to be beneficial for training psychomotor skills over and above conventional training. Further research into the role of haptic feedback and other potential predictive variables in performance outcomes is warranted. Researchers are also advised to improve methodological reporting to aid future evidence syntheses in this research area.

This study assessed the efficacy of an extended reality device to augment human performance during a dual-task paradigm. Dual-tasking is critical for successful high-risk military operations. However, current practices to improve mission success involving dual-task elements often require additional personnel which can increase risk and strain resources. Therefore, head-mounted extended reality devices which augment an individual's sensory and cognitive capacities could circumvent these issues whilst enhancing/maintaining task success. Utilising a repeated measures, counterbalanced, randomised design, 18 participants (8 females) simultaneously completed a primary ball-searching task and a secondary semantic verbal fluency task. Participants completed three conditions: using extended reality human augmentation; receiving memory prompts; or no support, each comprising three trials. Participants reported their perceived effort (post-trial), environmental presence (post-condition), and system usefulness/usability/sickness (post human augmentation condition only). Friedman’s tests identified that participants retrieved significantly more correct balls, significantly faster during the human augmentation versus memory prompt or no support conditions. Semantic verbal fluency and aggregated performance were also significantly better within the human augmentation condition than the no support condition. Perceived effort was lowest within the human augmentation condition. Within-condition, trial-by-trial differences emerged for semantic verbal fluency and aggregated performance within the human augmentation condition only. An extended reality, human augmentation device elicited significantly better dual-task performance compared to conditions reflecting current best practice (memory prompts) or no support which could be attributed to an alleviation of cognitive load. Within-condition effects suggest that even short-term exposure to extended reality devices can lead to immediate dual-task performance improvements.

This study examined the role of predictive processing mechanisms in naturalistic visuomotor skill learning. We investigated whether the acquisition of an interceptive skill was characterised by encoding and using learned probabilistic associations. Forty-five participants were randomly assigned to either an intervention or control group and practiced intercepting bouncing balls in virtual reality across two sessions. Only the intervention group received probabilistic visual cues – specifically, ball and room colours – that signalled likely bounce trajectories, offering an opportunity to learn and exploit these associations. We assessed task performance (interception rate, return accuracy), anticipatory gaze behaviour (predictive fixations and gaze tracking), and pupillary responses to unexpected outcomes. Participants in the intervention group learned the cue-outcome contingencies and demonstrated more predictive gaze strategies, finer post-bounce tracking, and physiological surprisal responses to probabilistically unlikely bounce outcomes. These findings suggest that encoding probabilistic environmental cues is a key component of effective visuomotor control. The results offer novel evidence for the role of top-down prediction and contextual modulation in motor learning, with implications for training and expertise development in sport. More broadly, this work contributes to an alternate account of expertise as the capacity to encode and exploit environmental regularities through predictive mechanisms.

This study assessed the efficacy of an extended reality device to augment human performance during a dual-task paradigm. Dual-tasking is critical for successful high-risk military operations. However, current practices to improve mission success involving dual-task elements, often require additional personnel, which can increase risk and strain resources. Therefore, head-mounted extended reality devices which augment an individual's sensory and cognitive capacities could circumvent these issues whilst enhancing/maintaining task success. Utilising a repeated measures, counterbalanced, randomised design, 18 participants (8 females) completed a primary ball-searching task and a secondary semantic verbal fluency task simultaneously. Participants completed three conditions: using extended reality human augmentation; receiving memory prompts; or no support, each comprising three trials. Participants reported their perceived effort (post-trial), environmental presence (post-condition), and system usefulness/usability/sickness (post human augmentation condition only). Friedman’s tests identified that participants retrieved significantly more correct balls (χ2 = 34.40, p < .001), significantly faster (χ2 = 21.70, p <.001) during the human augmentation versus memory prompt and no support conditions. Semantic verbal fluency (t = 3.465, p < .001) and aggregated performance (t = 4.130, p <.001) were also significantly better within the human augmentation condition than the no support condition. Perceived effort was lowest within the human augmentation condition. Trial-by-trial differences emerged for semantic verbal fluency and aggregated performance within the human augmentation condition only. An extended reality, human augmentation device elicited significantly better dual-task performance compared to conditions reflecting current best practice (memory prompts) or no support which could be attributed to an alleviation of cognitive load. Learning effects suggest that short-term exposure to extended reality devices can lead to immediate dual-task performance improvements.