Identifying behavioural change among drivers using Long Short-Term Memory recurrent neural networks

Globally, motor vehicle crashes account for over 1.2 million fatalities per year and are the leading cause of death for people aged 15–29 years. The majority of road crashes are caused by human error, with risk heightened among young and novice drivers learning to negotiate the complexities of the road environment. Direct feedback has been shown to have a positive impact on driving behaviour. Methods that could detect behavioural changes and therefore, positively reinforce safer driving during the early stages of driver licensing could have considerable road safety benefit. A new methodology is presented combining in-vehicle telematics technology, providing measurements forming a personalised driver profile, with neural networks to identify changes in driving behaviour. Using Long Short-Term Memory (LSTM) recurrent neural networks, individual drivers are identified based on their pattern of acceleration, deceleration and exceeding the speed limit. After model calibration, new, real-time data of the driver is supplied to the LSTM and, by monitoring prediction performance, one can assess whether a (positive or negative) change in driving behaviour is occurring over time. The paper highlights that the approach is robust to different neural network structures, data selections, calibration settings, and methodologies to select benchmarks for safe and unsafe driving. Presented case studies show additional model applications for investigating changes in driving behaviour among individuals following or during specific events (e.g., receipt of insurance renewal letters) and time periods (e.g., driving during holiday periods). The application of the presented methodology shows potential to form the basis of timely provision of direct feedback to drivers by telematics-based insurers. Such feedback may prevent internalisation of new, risky driving habits contributing to crash risk, potentially reducing deaths and injuries among young drivers as a result.     

An observational study of restraint and helmet wearing behaviour in Malaysia

Unrestrained and unhelmeted occupants are at increased risk of severe injury or death in the event of a crash, and there is evidence that children, particularly in low and middle income countries, have low restraint and helmet wearing rates.Roadside observations of occupants of passing vehicles (7247 cars, vans and taxis and 2897 motorcycles) in nearside lanes were made at seven selected road sites located around the Klang Valley (greater Kuala Lumpur) area representing a variety of demographic locations.The findings revealed an overall low rate of seat-belt wearing in vehicles (front seat drivers: 44.6%; front seat adult passengers: 33.8%; front seat child passengers: 11.8%; rear seat adult passengers: 5.2%; and, rear seat child passengers: 5.8%). For motorcyclists, the majority of adult riders and pillions wore helmets (93.4% and 85.8%, respectively), however a substantial proportion (40%) did not fasten their helmet properly. Of children observed on motorcycles, only 30.5% wore helmets.This study shows low seat-belt and helmet wearing rates, despite enactment of legislation that requires all vehicle passengers to be restrained and for all motorcyclists to wear helmets. This was particularly evident for children (front and rear seating positions) and rear seat adult passengers. It is suggested there is a general lack of awareness of safety and the benefits of restraint/helmet use, and lack of adequate and appropriate enforcement. The implications of the findings are discussed in terms of promoting restraint/helmets use, enforcement of legislation and implementation of technologies to increase appropriate restraint/helmet use.     

The effects of clinical supervision on supervisees and patients in cognitive behavioral therapy: a systematic review

Clinical supervision is a central part of psychotherapist training but the empirical support for specific supervision theories or features is unclear. The aims of this study were to systematically review the empirical research literature regarding the effects of clinical supervision on therapists’ competences and clinical outcomes within Cognitive Behavior Therapy (CBT). A comprehensive database search resulted in 4103 identified publications. Of these, 133 were scrutinized and in the end 5 studies were included in the review for data synthesis. The five studies were heterogeneous in scope and quality and only one provided firm empirical support for the positive effects of clinical supervision on therapists’ competence. The remaining four studies suffered from methodological weaknesses, but provided some preliminary support that clinical supervision may be beneficiary for novice therapists. No study could show benefits from supervision for patients. The research literature suggests that clinical supervision may have some potential effects on novice therapists’ competence compared to no supervision but the effects on clinical outcomes are still unclear. While bug-in-the-eye live supervision may be more effective than standard delayed supervision, the effects of specific supervision models or features are also unclear. There is a continued need for high-quality empirical studies on the effects of clinical supervision in psychotherapy.     

Parent management training for conduct problems in children: Enhancing treatment to improve therapeutic changeCapacitación en entrenamiento de padres para problemas de conducta en niños: mejorar el tratamiento para mejorar el cambio terapéutico

Background/Objective: The study evaluated two variations of Parent Management Training (PMT) for children referred to treatment for oppositional, aggressive, and antisocial behavior. The goal was to evaluate the impact of multiple enhancements to optimize common and placebo factors to augment therapeutic change. Method: The families of all children (N=138, 39 girls and 99 boys, ages 6-13) received PMT. One half of the families were assigned to receive an enhanced version with multiple additions designed to increase bonding of the parent to the therapist, professionalism of treatment and setting, credibility of the intervention, and expectancies for therapeutic change. Assessment included multiple treatment outcome measures of the child (problem behaviors, psychiatric symptoms, social competence, and adaptive functioning) and parents (depression, stress, and family relations) showed marked improvements over the course of treatment, and several process measures (therapeutic alliance, credibility of the procedures, expectancy for change). Results: The results indicated that children and parents made marked improvement in all the treatment outcome measures. The vast majority of children fell within the normative range at posttreatment on problem and prosocial behaviors. The two treatment conditions were no different in outcomes for children or parents. Conclusion: PMT led to marked changes in treatment outcome.     

Validating Sizing Them Up: A parent-proxy weight-related quality-of-life measure,with community-based children

Background/Objective: A weight-related Quality of Life (QoL) questionnaire elicits parents’ perceptions to understand the impacts of weight on children's QoL. Sizing Them Up, a parent-proxy of a weight-related instrument, is a proper measure for the purpose, but its psychometric properties have not been validated on a non-clinical child population. This study aimed to thoroughly examine the psychometric properties of the Sizing Them Up on a non-clinical child population. Methods: Children from Taiwan (M_age = 11; SD = 1.02; N = 236) filled out a weight-related QoL instrument (Sizing Me Up) while their parents completed Sizing Them Up and two generic QoL instruments. Confirmatory factor analysis (CFA), concurrent validity, and known-group validity were conducted. Results: CFA supported the factor structure of Sizing Them Up. The concurrent validity of the Sizing Them Up total score was satisfactory: r = .46 to .53 with other QoL instruments. Known-group validity was supported as obese children had significantly lower Sizing Them Up total score than did normal-weight and underweight children. Conclusions: We extended the use of Sizing Them Up from clinically obese children to a community-based sample of underweight as well as overweight children. Healthcare providers may use Sizing Them Up to measure the weight-related QoL for children.     

Concurrent validity between instruments of assessment of motor development in infants exposed to HIV

BackgroundExposure to HIV during pregnancy is a risks to development. Exposed child should have assessed its development since birth. Alberta Infant Motor Scale is a tool which assess gross motor skills, with easy application and low cost. Up to now, this scale had not proven its validity for the population exposed to HIV. It’s necessary to compare its with a gold standard tool, Bayley scale, which assess gross and fine motor skills, has a high cost and longer application time required. Studies compare results of Alberta with Bayley’s total motor score (gross + fine). However, it’s also necessary to compare Alberta’s result with only Bayley’s gross motor result, because it’s what both evaluate in common.Aimsto verify the concurrent validity of AIMS in infants exposed to HIV; to verify the correlation of AIMS and BSITD III for this population and to compare if these coefficients differ in the central age groups and extremities of the AIMS.Methods82 infants exposed to HIV evaluated in 1st, 2nd, 3rd, 4th, 8th, 12th, 15th, 16th, 17th and 18th months, with Alberta Infant Motor Scale and Bayley Scale (motor subscale). For analysis of concurrent validity, results of raw scores of the scales were compared with the correlation analysis. First analysis: Alberta’s score with Bayley’s total (gross + fine) motor score. Second analysis: Alberta’s score with Bayley’s gross motor score.ResultsIn the first correlation analysis, results were: r = 0.62 in 1 st month, r = 0.64 in 2nd month, r = 0.08 in 3rd month, r = 0.45 in 4th month; r = 0.62 in 8th month, r = 0.60 in the 12th month. In the second correlation analysis, results were: r = 0.69 in 1 st month; r = 0.58 in 2nd month; r = 0.25 in 3rd month; r = 0.45 in the 4th month; r = 0.77 in 8th month; r = 0.73 in 12th month. Analyzes of the 15th, 16th, 17th and 18th months couldn’t be performed because at these ages all the children had already reached the maximum score in the AIMS. Results were significant and indicate correlation between scales. Found results agree with other studies that found high correlations between the scales in premature and risk groups. However, these studies compare results of gross motor skills assessments with gross and fine motor skills assessments. Our results show that correlation only between the gross motor skills have higher coefficient values, and we believe this is the best way to compare the scales, with what both assessed in common.ConclusionsAlberta scale has correlation with Bayley scale in assessing of children exposed to HIV, and can be a substitute to Bayley in assessing of these children. Results are stronger when comparing only what both scales assess in common.     

Parent–Child Conflict Profiles in Chinese American Immigrant Families: Links to Sociocultural Factors and School‐Age Children's Psychological Adjustment

Research suggests that parent–child conflict is a salient family process in Asian immigrant families and often a stressful experience for Asian American youth due to value discrepancies between Asian and Western cultures. The present study examined ratings of parent–child conflict across conflict topics from parents' and children's perspectives in a sample of Chinese American immigrant families with school‐age children (N = 239; age = 7.5–11 years). Latent profile analyses identified three parent‐rated conflict profiles and four child‐rated conflict profiles. Parent and child conflict profiles were unrelated to each other and differentially related to family sociocultural factors and children's psychological adjustment. Parents' moderate conflict profile scored highest on parent‐rated child behavior problems and had the highest household density and lower parent Chinese orientation. Children's moderate‐specific and high conflict profiles scored higher on child‐reported behavior problems than the low conflict profile. These results highlight the need to assess family conflict from both parents' and children's perspectives and target parent–child conflict communication as a pathway to prevent or reduce behavioral problems in Chinese American children of immigrant families.     

Continuous measurement of attachment behavior: A multimodal view of the strange situation procedure

Infant attachment is a critical indicator of healthy infant social-emotional functioning, which is typically measured using the gold-standard Strange Situation Procedure (SSP). However, expert-based attachment classifications from the SSP are time-intensive (with respect both to expert training and rating), and do not provide an objective, continuous record of infant behavior. To continuously quantify predictors of key attachment behaviors and dimensions, multimodal movement and audio data were collected during the SSP. Forty-nine 1-year-olds and their mothers participated in the SSP and were tracked in three-dimensional space using five synchronized Kinect sensors; LENA recordings were used to quantify crying duration. Theoretically-informed multimodal measures of attachment-related behavior (e.g., dyadic contact duration, infant velocity of approach toward the mother, and infant crying) were used to predict expert rating scales and dimensional summaries of attachment outcomes. Stepwise regressions identified sets of multimodal objective measures that were significant predictors of eight of nine of the expert ratings of infant attachment behaviors in the SSP’s two reunions. These multimodal measures predicted approximately half of the variance in the summary approach/avoidance and resistance/disorganization attachment dimensions. Incorporating all objective measures as predictors regardless of significance levels, predicted individual ratings within an average of one point on the original Likert scales. The results indicate that relatively inexpensive Kinect and LENA sensors can be harnessed to quantify attachment behavior in a key assessment protocol, suggesting the promise of objective measurement to understanding infant-parent interaction.     

Is the street-crossing behavior with a head-mounted display different from that behavior in a CAVE? A study among young adults and children

The use of virtual reality (VR) has become increasingly popular in the field of traffic psychology, where realistic traffic situations can be simulated and pedestrians‘ actual crossing behavior can be studied. There are two main kinds of pedestrian simulators: one uses a technology based on rear-projection screens (Cave Automatic Virtual Environment, or CAVE); the other uses a head-mounted display (HMD). Despite their extensive use, it is yet to determine whether they are equally suitable for studying street crossing. The present study was aimed at comparing street-crossing behavior and subjective evaluations in an HMD-based (HTC vive pro) pedestrian simulator and a CAVE-like pedestrian simulator, among young adults and 12-year-old children. Thirty young adults and twenty-six children performed 36 street-crossing trials (combining different speeds, traffic conditions, and gap sizes) on each of the two simulators. The results indicated that, compared to the CAVE-condition participants, HMD-condition participants accepted more crossing trials (hence, shorter gaps), initiated their crossings sooner, crossed at a slower speed, had shorter safety margins, and caused more collisions. The main difference between the two devices was in crossing initiation, which occurred markedly earlier (1.72 s) with the HMD than in the CAVE: the perception–action coupling was less finely tuned in the CAVE, probably because visual information in front of the pedestrian was missing due the absence of ground projection and 3D rendering. A significant effect of vehicle-approach speed was observed for both devices, with more unsafe behaviors at 60 km/h than at 40 km/h. Children displayed riskier crossing behavior than young adults did, with more accepted crossings, slower crossing speeds, shorter safety margins, and higher collision rates, especially in the HMD. Compared to the CAVE, the HMD received higher ratings for level of presence and a preference for VR, but also higher simulator-sickness scores. No adverse effects of exposure to virtual reality was found on stereoacuity or postural balance. The suitability of using CAVE and HMD simulators is discussed, especially for studying child pedestrians.