Machine learning and EEG can classify passive viewing of discrete categories of visual stimuli but not the observation of pain.

Previous studies have demonstrated the potential of machine learning (ML) in classifying physical pain from non-pain states using electroencephalographic (EEG) data. However, the application of ML to EEG data to categorise the observation of pain versus non-pain images of human facial expressions or scenes depicting pain being inflicted has not been explored. The present study aimed to address this by training Random Forest (RF) models on cortical event-related potentials (ERPs) recorded while participants passively viewed faces displaying either pain or neutral expressions, as well as action scenes depicting pain or matched non-pain (neutral) scenarios. Ninety-one participants were recruited across three samples, which included a model development group (n = 40) and a cross-subject validation group (n = 51). Additionally, 25 participants from the model development group completed a second experimental session, providing a within-subject temporal validation sample. The analysis of ERPs revealed an enhanced N170 component in response to faces compared to action scenes. Moreover, an increased late positive potential (LPP) was observed during the viewing of pain scenes compared to neutral scenes. Additionally, an enhanced P3 response was found when participants viewed faces displaying pain expressions compared to neutral expressions. Subsequently, three RF models were developed to classify images into faces and scenes, neutral and pain scenes, and neutral and pain expressions. The RF model achieved classification accuracies of 75%, 64%, and 69% for cross-validation, cross-subject, and within-subject classifications, respectively, along with reasonably calibrated predictions for the classification of face versus scene images. However, the RF model was unable to classify pain versus neutral stimuli above chance levels when presented with subsequent tasks involving images from either category. These results expand upon previous findings by externally validating the use of ML in classifying ERPs related to different categories of visual images, namely faces and scenes. The results also indicate the limitations of ML in distinguishing pain and non-pain connotations using ERP responses to the passive viewing of visually similar images.

The neural correlates of texture perception: A systematic review and activation likelihood estimation meta-analysis of functional magnetic resonance imaging studies.

INTRODUCTION: Humans use discriminative touch to perceive texture through dynamic interactions with surfaces, activating low-threshold mechanoreceptors in the skin. It was largely assumed that texture was processed in primary somatosensory regions in the brain; however, imaging studies indicate heterogeneous patterns of brain activity associated with texture processing. METHODS: To address this, we conducted a coordinate-based activation likelihood estimation meta-analysis of 13 functional magnetic resonance imaging studies (comprising 15 experiments contributing 228 participants and 275 foci) selected by a systematic review. RESULTS: Concordant activations for texture perception occurred in the left primary somatosensory and motor regions, with bilateral activations in the secondary somatosensory, posterior insula, and premotor and supplementary motor cortices. We also evaluated differences between studies that compared touch processing to non-haptic control (e.g., rest or visual control) or those that used haptic control (e.g., shape or orientation perception) to specifically investigate texture encoding. Studies employing a haptic control revealed concordance for texture processing only in the left secondary somatosensory cortex. Contrast analyses demonstrated greater concordance of activations in the left primary somatosensory regions and inferior parietal cortex for studies with a non-haptic control, compared to experiments accounting for other haptic aspects. CONCLUSION: These findings suggest that texture processing may recruit higher order integrative structures, and the secondary somatosensory cortex may play a key role in encoding textural properties. The present study provides unique insight into the neural correlates of texture-related processing by assessing the influence of non-textural haptic elements and identifies opportunities for a future research design to understand the neural processing of texture.

Statistics anxiety and predictions of exam performance in UK psychology students.

BACKGROUND: Statistics anxiety is common among social science students. Despite much evidence examining statistics anxiety and test performance, little research has explored the role of student self-prediction on test performance in a higher education setting. OBJECTIVE: The purpose of this study was to investigate the relationship between statistics anxiety and both students' self-prediction of their future exam performance and actual test performance on a formal statistics assessment at undergraduate level in psychology students in the UK. METHOD: Using a cross-sectional design, two hundred and two students were required to complete Statistics Anxiety Rating Scales, the Mathematical Prerequisites for Psychometrics Scale, and provided self-predicted test performance scores. Test performance data was obtained from a formal statistics assessment. RESULTS: As predicted, we demonstrated statistics test anxiety to be negatively associated with self-predicted performance. Additionally, we found statistics anxiety was positively associated with test performance. CONCLUSION: The findings highlight the complex relationship between statistics anxiety and test performance, suggesting there may be an optimal level of anxiety for performance in statistics assessments. IMPLICATIONS: The results we report have implications for psychology research methods and statistics instructors who may wish to incorporate the findings into statistics instruction modules in order to assuage high levels of statistics anxiety and foster student well-being.

Economic value in the Brain: A meta-analysis of willingness-to-pay using the Becker-DeGroot-Marschak auction.

Forming and comparing subjective values (SVs) of choice options is a critical stage of decision-making. Previous studies have highlighted a complex network of brain regions involved in this process by utilising a diverse range of tasks and stimuli, varying in economic, hedonic and sensory qualities. However, the heterogeneity of tasks and sensory modalities may systematically confound the set of regions mediating the SVs of goods. To identify and delineate the core brain valuation system involved in processing SV, we utilised the Becker-DeGroot-Marschak (BDM) auction, an incentivised demand-revealing mechanism which quantifies SV through the economic metric of willingness-to-pay (WTP). A coordinate-based activation likelihood estimation meta-analysis analysed twenty-four fMRI studies employing a BDM task (731 participants; 190 foci). Using an additional contrast analysis, we also investigated whether this encoding of SV would be invariant to the concurrency of auction task and fMRI recordings. A fail-safe number analysis was conducted to explore potential publication bias. WTP positively correlated with fMRI-BOLD activations in the left ventromedial prefrontal cortex with a sub-cluster extending into anterior cingulate cortex, bilateral ventral striatum, right dorsolateral prefrontal cortex, right inferior frontal gyrus, and right anterior insula. Contrast analysis identified preferential engagement of the mentalizing-related structures in response to concurrent scanning. Together, our findings offer succinct empirical support for the core structures participating in the formation of SV, separate from the hedonic aspects of reward and evaluated in terms of WTP using BDM, and show the selective involvement of inhibition-related brain structures during active valuation.

Tactile estimation of hedonic and sensory properties during active touch: An electroencephalography study.

Perceptual judgements about our physical environment are informed by somatosensory information. In real-world exploration, this often involves dynamic hand movements to contact surfaces, termed active touch. The current study investigated cortical oscillatory changes during active exploration to inform the estimation of surface properties and hedonic preferences of two textured stimuli: smooth silk and rough hessian. A purpose-built touch sensor quantified active touch, and oscillatory brain activity was recorded from 129-channel electroencephalography. By fusing these data streams at a single trial level, oscillatory changes within the brain were examined while controlling for objective touch parameters (i.e., friction). Time-frequency analysis was used to quantify changes in cortical oscillatory activity in alpha (8-12 Hz) and beta (16-24 Hz) frequency bands. Results reproduce findings from our lab, whereby active exploration of rough textures increased alpha-band event-related desynchronisation in contralateral sensorimotor areas. Hedonic processing of less preferred textures resulted in an increase in temporoparietal beta-band and frontal alpha-band event-related desynchronisation relative to most preferred textures, suggesting that higher order brain regions are involved in the hedonic processing of texture. Overall, the current study provides novel insight into the neural mechanisms underlying texture perception during active touch and how this process is influenced by cognitive tasks.

Neural correlates of perceptual texture change during active touch.

Introduction: Texture changes occur frequently during real-world haptic explorations, but the neural processes that encode perceptual texture change remain relatively unknown. The present study examines cortical oscillatory changes during transitions between different surface textures during active touch. Methods: Participants explored two differing textures whilst oscillatory brain activity and finger position data were recorded using 129-channel electroencephalography and a purpose-built touch sensor. These data streams were fused to calculate epochs relative to the time when the moving finger crossed the textural boundary on a 3D-printed sample. Changes in oscillatory band power in alpha (8-12 Hz), beta (16-24 Hz) and theta (4-7 Hz) frequency bands were investigated. Results: Alpha-band power reduced over bilateral sensorimotor areas during the transition period relative to ongoing texture processing, indicating that alpha-band activity is modulated by perceptual texture change during complex ongoing tactile exploration. Further, reduced beta-band power was observed in central sensorimotor areas when participants transitioned from rough to smooth relative to transitioning from smooth to rough textures, supporting previous research that beta-band activity is mediated by high-frequency vibrotactile cues. Discussion: The present findings suggest that perceptual texture change is encoded in the brain in alpha-band oscillatory activity whilst completing continuous naturalistic movements across textures.

External validation of binary machine learning models for pain intensity perception classification from EEG in healthy individuals.

Discrimination of pain intensity using machine learning (ML) and electroencephalography (EEG) has significant potential for clinical applications, especially in scenarios where self-report is unsuitable. However, existing research is limited due to a lack of external validation (assessing performance using novel data). We aimed for the first external validation study for pain intensity classification with EEG. Pneumatic pressure stimuli were delivered to the fingernail bed at high and low pain intensities during two independent EEG experiments with healthy participants. Study one (n = 25) was utilised for training and cross-validation. Study two (n = 15) was used for external validation one (identical stimulation parameters to study one) and external validation two (new stimulation parameters). Time-frequency features of peri-stimulus EEG were computed on a single-trial basis for all electrodes. ML training and analysis were performed on a subset of features, identified through feature selection, which were distributed across scalp electrodes and included frontal, central, and parietal regions. Results demonstrated that ML models outperformed chance. The Random Forest (RF) achieved the greatest accuracies of 73.18, 68.32 and 60.42% for cross-validation, external validation one and two, respectively. Importantly, this research is the first to externally validate ML and EEG for the classification of intensity during experimental pain, demonstrating promising performance which generalises to novel samples and paradigms. These findings offer the most rigorous estimates of ML's clinical potential for pain classification.

Neural correlates of texture perception during active touch.

Previous studies have shown attenuation of cortical oscillations over bilateral sensorimotor cortex areas during passive perception of smooth textures applied to the skin. However, humans typically explore surfaces using dynamic hand movements. As movements may both modulate texture-related cortical activity and induce movement-related cortical activation, data from passive texture perception cannot be extrapolated to active texture perception. In the present study, we used electroencephalography to investigate cortical oscillatory changes during texture perception throughout active touch exploration. Three natural textured stimuli were selected: smooth silk, soft brushed cotton, and rough hessian. Texture samples were mounted on a purpose-built touch sensor which measured the load and position of the index finger, whilst electroencephalography from 129 channels recorded oscillatory brain activity. The data were fused to investigate oscillatory changes relating to active touch. Changes in oscillatory band power, event-related desynchronisation/synchronisation (ERD/ERS), were investigated in alpha (8-12 Hz) and beta (16-24 Hz) frequency bands. Active texture exploration revealed bilateral activation patterns over sensorimotor cortical areas. Beta-band ERD increased over contralateral sensorimotor regions for soft and smooth textures, and over ipsilateral sensorimotor areas for the smoothest texture. Analysis of covariance revealed that individual differences in perception of softness and smoothness were related to variations in cortical oscillatory activity. Differences may be due to increased high frequency vibrations for smooth and soft textures compared to rough. For the first time, active touch was quantified and fused with electroencephalography data streams, contributing to the understanding of the neural correlates of texture perception during active touch.

Systematic Review of the Effectiveness of Machine Learning Algorithms for Classifying Pain Intensity, Phenotype or Treatment Outcomes Using Electroencephalogram Data.

Recent attempts to utilize machine learning (ML) to predict pain-related outcomes from Electroencephalogram (EEG) data demonstrate promising results. The primary aim of this review was to evaluate the effectiveness of ML algorithms for predicting pain intensity, phenotypes or treatment response from EEG. Electronic databases MEDLINE, EMBASE, Web of Science, PsycINFO and The Cochrane Library were searched. A total of 44 eligible studies were identified, with 22 presenting attempts to predict pain intensity, 15 investigating the prediction of pain phenotypes and seven assessing the prediction of treatment response. A meta-analysis was not considered appropriate for this review due to heterogeneous methods and reporting. Consequently, data were narratively synthesized. The results demonstrate that the best performing model of the individual studies allows for the prediction of pain intensity, phenotypes and treatment response with accuracies ranging between 62 to 100%, 57 to 99% and 65 to 95.24%, respectively. The results suggest that ML has the potential to effectively predict pain outcomes, which may eventually be used to assist clinical care. However, inadequate reporting and potential bias reduce confidence in the results. Future research should improve reporting standards and externally validate models to decrease bias, which would increase the feasibility of clinical translation. PERSPECTIVE: This systematic review explores the state-of-the-art machine learning methods for predicting pain intensity, phenotype or treatment response from EEG data. Results suggest that machine learning may demonstrate clinical utility, pending further research and development. Areas for improvement, including standardized processing, reporting and the need for better methodological assessment tools, are discussed.