Traffic Psychology Notes PDF

lOMoARcPSD|7065412 Traffic Psychology - Samenvatting colleges + literatuur Traffic Psychology and Sustained Mobility (Rijksuniversiteit Groningen) Scannen om te openen op Studeersnel Studeersnel wordt niet gesponsord of ondersteund door een hogeschool of universiteit Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 Traffic Psychology Lecture 1: Models, behavioral adaptation Psychologist make conceptual models, a good theory / model should be descriptive and it should predict. Driving is just another human behavior, specifically a transport behavior. A lot of people die because of traffic. Driving requires sustained attention, hazard identification and motor coordination. There are a lot of theories, none of them are widely accepted. Below are stated a few models / theories about traffic psychology Skills models → A drivers perceptual and motor skills are what makes them save o Reaction time, vision, level of driver training. Crashes occur when Task Demands exceed Driver Skills ○ Critique → too simplistic, early attempts using visual attributes and reaction time found weak/no relationship to accidents, there are other properties that are more important (like biographical and exposure factors) Attitude theories → Theory of planned behavior (TPB) ○ Attitude = what I say I think about something ○ Subjective norm = what I think others think about something ○ Perceived control = how much control I think I have over my behavior ○ Intention = what I say I will do ○ Behavior = what I actually do ○ Also applicable to drunk driving. ○ Critique: Intention-/attitude behavior gap. TPB is very good in predicting the intention, but we want to know if it will actually happen. Predicts 90% of intentions, but only 10-20% of observed behavior. Very weak link between attitude and behavior. ○ Rothengatter: drivers find it easier to adapt their behavior than their attitudes. Strong habits cause a weak relation between intentions and behavior, when there are weak habits there is a correlation between intention and behavior. Utility theories ○ Maximise gain and minimize loss, assumes rationality → homo economicus ○ Or at least trade these variables off, rational decision models; Utility maximization Subjective expected utility Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 Peltzman’s Driving Intensity theory Do you say yes/no to a certain person on the street. = too slow, for most of the decisions to make in traffic it simply takes too long. But for example if you are deciding if you’re going by bus or bicycle it may work. Rational approach of US paying for blood: did not work, unpayed (like in NL) gave more blood and of better quality. People are not calculators Risk / motivational Theories → takes motives into account. ○ Risk Homeostasis Theory (Wilde) → The Risk Homeostasis Theory in traffic psychology suggests that individuals have a fixed level of acceptable risk. When safety measures are introduced, such as seat belts or improved road conditions, people may respond by taking more risks, consciously or subconsciously, to maintain their desired level of risk. In essence, they offset the safety benefits of these measures by engaging in riskier behaviors, like speeding or aggressive driving. Individuals have a target level of risk; Target level is set at a societal level as well. If experienced risk does not equal target risk then action is taken Does not make full testable predictions, relies on constant risk perception. Therefore very difficult (impossible) to falsify. Predicts safety measures. Behavioural adaptation: Winter tyres caused more accidents instead of less because people feel more save, they think they have better control over a vehicle. Similar effect with ABS. Also with blind spot monitoring, people pay too much attention to the systems and accidents happen. ○ Risk Allostasis Theory / Task-Difficulty Homeostasis Theory (Fuller) → proposes that individuals adapt their risk-taking behavior to external conditions. Instead of a fixed risk threshold, this theory suggests that people adjust their risk tolerance based on factors like traffic conditions, weather, or the presence of safety measures. They may drive more cautiously in adverse conditions and be more risk-prone when conditions are favorable, indicating a flexible approach to risk. Feeling of Risk the central motivator Feeling of Risk is an indicator of perceived task difficulty Preferred range of Feeling of Risk compared to perceived level Stimuli are ‘marked’ with emotions, emotions are body states Cognition arises from the body People who lack emotion should be irrational then, but can perform well in tests Task demand and task difficulty, Perceived task difficulty; finally moving away from risk! o Risk threshold === zero risk theory (at a certain moment there will be a response) = Allostasis is more dynamic, Homeostasis posits a fixed target level of task difficulty of risk. Therefore Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 allostasis: better representation of changing motivational influences on the driver. Example: ambulance, depending of the cruciality of the incident different risks are taken to arrive in as little time as possible A threshold can be seen at a speed op to 50km/h in perceived risk and task difficult Het grootste verschil tussen de twee modellen is dat de Risk Homeostasis Theory suggereert dat mensen een vaste acceptabele risicogrens hebben en hun gedrag aanpassen om deze grens te behouden, terwijl de Risk Allostasis Theory stelt dat mensen hun risicotolerantie aanpassen aan veranderende omstandigheden zoals verkeersomstandigheden en veiligheidsmaatregelen. Het eerste model impliceert een vaste risicodrempel, terwijl het tweede model meer flexibiliteit in risicoaanpassing suggereert. Zero-Risk Theory (Näätänen & Summala) ○ Argues that risk is hardly ever experienced ○ When it is it acts as a warning ○ Risk Threshold Risk/Threat Avoidance Theory (Fuller) Risk is aversive and avoided. Behavioural contingencies are learnt What is risky depends on experience (Learnt Riskiness) RISK = The probability of an outcome x the consequence of that outcome The Safety Margin Model / multiple comfort zone model (Summala) People are motivated to be comfortable. In the context of traffic psychology, Safety Margin Mode refers to strategies and behaviors that drivers employ to ensure an additional safety buffer on the road. This mode involves consciously maintaining extra distance, time, and awareness to minimize the risk of accidents and respond effectively to unexpected situations. Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 Constant perception models → RHT & RAT Threshold perception models → Zero-Risk, Risk Avoidance Theory, SMM Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 Conclusions models: 1. Skills are necessary, but only to a point 2. Change behaviour, not attitude’s a. It’s easier, and more effective 3. Do not expect rationally a. People are not calculators + Emotion plays a role 4. A lot of behaviour is habitual/well learnt a. Change/shape behaviour early, if possible 5. Critical decision points a. When thresholds are crossed 6. The best way to change behaviour is to change the environment in which people operate, not to try and change the person directly a. e.g. median barriers, increased enforcement, electronic aids, and regulation Readings 1 - Testing models of traffic behavior This text discusses the state of traffic psychology and reviews major models and theories related to driving behavior. It mentions that there have been concerns about the atheoretical and fragmented nature of traffic psychology, but few comprehensive reviews have attempted to unify the field. Some notable reviews by Michon, Ranney, Huguenin and Rumar, and Rothengatter are highlighted. Michon's 1985 and 1989 reviews categorized driver behavior models into two groups: those based on curve fitting (magical) and those driven by explicitly stated concepts and rules (rational or functional). Michon found promise in rational models, particularly rule-based theories, which emphasize applying rules in an "IF THEN" fashion. He mentioned SOAR cognitive architecture as a potential way forward. Ranney's 1994 review traced the historical evolution of driver behavior models, from accident proneness theories to examining individual traits and their impact on behavior. He discussed motivational models and information-processing theories, highlighting the use of hierarchies in understanding driver behavior. Huguenin and Rumar's 2001 review expressed skepticism about unifying driver behavior models due to their limited scope and the complexity of the driving task. They classified models as driver task related, functional control, or motivational but found these classifications overly simplistic. They discussed risk, behavioral adaptation, and motivational models, critiquing them for various reasons. Rothengatter's 2002 review advocated for more inclusion of social psychology in driver behavior models. He criticized motivational models for lacking testable hypotheses and being overly reliant on risk. Rothengatter also discussed attitude theories and mentioned Task Difficulty Homeostasis Theory (TDH) as promising but noted its circular reasoning. He concluded that the Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 intersection of performance theories and social psychology should be the focus of traffic psychology. The text acknowledges that these reviews highlighted the lack of consensus in the field and criticized older trait-based models for their vagueness and unfalsifiability. It mentions the continued development of new models, especially those explaining driver adaptation to driver assistance features, and identifies four dominant models: TDH, Risk Allostasis Theory (RAT), Risk Monitor Model (RMM), and the multiple comfort zone model. 2.1 How Task Difficulty became a RAT (Risk Allostasis Theory) The text discusses two interconnected models in the field of driver behavior analysis: Task- Capability Interface (TCI) and Task Difficulty Homeostasis theory (TDH). Here's a summary: Criticism of Driver Behavior Models: The text begins by mentioning the common criticism of driver behavior models, which is their inability to generate testable hypotheses and their descriptive rather than predictive nature. Task-Capability Interface (TCI): ○ Capability: TCI posits that drivers have a hierarchical capability structure. It starts with constitutional features like reaction time and mental processing speed. Training, education, and experience enhance this capability, and this combination is termed "driver's competence." Human factors, like motivations, attitudes, and state conditions (e.g., intoxication, fatigue, emotions), are subtracted from competence, resulting in the driver's situational ability. ○ Task Demands: On the other side of TCI are task demands, influenced by factors like road conditions, weather, vehicle characteristics, and driver behavior. Task Difficulty and the Loss of Control: Task difficulty is determined by the interaction between task demands and capability. If task demands exceed capability, there is a risk of losing control. As task demand approaches capability, performance may deteriorate. Limitations of TCI: TCI is criticized for being descriptive and not generating testable hypotheses. It also doesn't explain how drivers assess their choices or adapt their behavior. Task Difficulty Homeostasis Theory (TDH): ○ TDH proposes that drivers have a preferred range of experienced task difficulty, influenced by perceived capability, effort motivation, and trip goals. ○ Perceived task difficulty is continuously compared to the preferred range, and behavior adapts to maintain task difficulty within this range. Proximal and Distal Determinants: TDH incorporates distal and proximal determinants that affect the comparison between perceived task difficulty and the preferred range. These determinants aim to make the model more comprehensive. Risk Threshold: TDH introduces the concept of a risk threshold, which triggers warnings when drivers approach the limits of their preferred task difficulty range. This threshold relates to a "feeling of risk" rather than crash risk. Risk Allostasis Theory (RAT): RAT expands upon TDH by emphasizing the role of feelings of risk and their continuous monitoring in driver decision-making. Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 Somatic Marker Hypothesis: RAT connects with the Somatic Marker Hypothesis, suggesting that emotions and feelings can influence decision-making processes, including those related to risk. Research and Criticisms: The text highlights research supporting TDH and RAT, particularly a study showing a correlation between task difficulty and feelings of risk. However, it also mentions criticisms, including conflicting findings and the constant monitoring of feelings of risk. Other Criticisms: The models are criticized for focusing on speed and vehicle trajectory as the primary means of changing task demand. The concept of a target range for task difficulty is also questioned, as it lacks clarity. Comparison to Risk Homeostasis Theory (RHT): TDH and RAT are compared to RHT, with a distinction made in terms of the role of objective accident risk and utility decisions in the models. Evidence and Predictions: The text concludes by discussing a study supporting TDH and generating specific predictions but notes that real-world speed data doesn't always align with these predictions. In essence, the text presents an overview of two driver behavior models (TCI and TDH/RAT), their components, criticisms, and evidence, highlighting the ongoing debate and challenges in this field of research. Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 2.2 Monitoring the Risk Monitor Model This paragraph discusses the Risk Monitor Model (RMM) and the Multiple Comfort Zone Model, two models related to driver behavior in the context of risk and comfort. Risk Monitor Model (RMM): Originally called the Monitor Model, it was developed to understand driver behavior. It combines elements from various older models and theories, including the Somatic Marker Hypothesis, zero-risk theory, and operant conditioning. According to RMM, a driver's personality traits, interactions with other drivers, and other factors are filtered through the somatic marking system before influencing driver decisions. It suggests that drivers are motivated to maintain their body's functional balance and strive for a "best feeling," which can vary in type and intensity. Feelings of risk or monitoring risky situations play a key role in achieving this "best Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 feeling." Some critiques challenge the assumption that the body primarily acts as a risk detector and question the constant monitoring of risk. Multiple Comfort Zone Model: An evolution of zero-risk theory and other models, it's a hierarchical and motivational model of driver behavior. It considers driver motives, personality, and driving goals as key factors influencing behavior. Safety margins are crucial in this model, with drivers maintaining personal space around their vehicles to avoid feeling uncomfortable. Safety margins are based on factors like time to collision and time to lane crossing, affecting driver decisions. Operating within safety margins results in a general mood of comfort. The model emphasizes satisficing (doing enough to meet goals) rather than optimizing. Critiques and discussions: Some studies question the threshold account of feelings of risk and task difficulty proposed by this model. Critics argue that the model's ideas could be integrated into other existing models like the Risk Awareness Theory. Objections include concerns about the need for drivers to constantly monitor risk and the learning complexities involved in creating all safety margins. There are discussions about the model's hierarchical nature and whether it truly reflects the complexity of real-world driving environments. Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 In summary, the paragraph provides insights into two models of driver behavior, the Risk Monitor Model (RMM) and the Multiple Comfort Zone Model, along with critiques and discussions about their concepts and applicability. Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 Lecture 2: Automation & Drivers support People used to think that machines work beter than people, so machines should take over control. However; Some tasks are better performed by man (overview, improvise) and others by machines (calculations). Examples of automation in traffic Support systems (navigation, speed limit info, warningssystems) Automatic driving (vehicle control) Automation means the driver is not in control anymore, but only needs to monitor the situations. SAE levels: Human driver monitors the driving environment 1. Non-automated: full-time performance by the human driver of all aspects of the dynamic driving task (NHTSA: 0) 2. Assisted: the driving mode-specific execution by a driver assistance system of either steering or acceleration/deceleration using information about the driving environment and with the expectation that the human driver performs all the remaining aspects of dynamic driving task. (NHTSA: 1) 3. Partial automation: the driving mode-specific execution by one or more driver assistance systems of both steering or acceleration/deceleration using information about the driving environment and with the expectation that the human driver performs all the remaining aspects of dynamic driving task (NHTSA: 2) Automated driving system monitors the driving environment Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 4. Conditional automation: the driving mode-specific performance by an automated driving system of all aspects of the dynamic driving task with the expectation that the human driver will respond appropriately to a request to intervene (NHTSA: 3) 5. High automation: the driving mode-specific performance by an automated driving system of all aspects of the dynamic driving task, even if a human driver does not respond appropriately to a request to intervene (NHTSA: 4) 6. Full automation: the full-time performance by an automated driving system of all aspects of the dynamic driving task under all roadway and environmental conditions that can be managed by a human driver (NHTSA: 4) The categorization of SAE is the most used. Level 3 is one that is a lot of going right now. Most of the tasks are automatic, but we are still responsible. We must do the boring tasks. Humans and Automation (Parasuraman & Riley 1997) Use : Voluntarily activating or disengaging automation Abuse : task is taken over by machine without considering the consequences for the operator (e.g. “old fashioned take-over”) (clumsy automation) Disuse: neglect automation/underutilisation ○ distrust precedes (gaat vooraf aan)disuse (e.g. high rate false alarms/ bad communication intentions -> automation induced surprises) Misuse: unjustified overreliance (complacency) ○ over-trust, out-of-the-loop* behaviour, reduced situation awareness, skill loss Trust calibration (Franseco Walker) involves the ability to accurately gauge and adjust one's level of trust in the vehicle's capabilities and the driver's skills, helping to make safer driving decisions. SLIDE 18 ™ 23 TERUG KIJKEN Study on automated highway system → 50% did not reclaim control on an automated highway system when something urgent happened Why use automatization? It makes driving more easier and you need less asphalt. “tram on tyres” → three modes of operation Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 - Bus: lateral (steering the wheels for path tracking) & longitudinal (regulation the vehicle’s cruise velocity) control by hand - Tram: lateral control automated - Underground: full automated The driver is still under control in all of these modes. Phileas: - Use: automatic driving or not - Disuse: switch off frustrated by false alarms - Misuse: reclaim control in a dangerous situation Adaptable automation: human determines how automation is applied Adaptive automation: system assigns automation level. Support adequately and adapt to needs: Monitor the operator Keep him/her in the loop with an active role Tasks depend dynamically on workload (avoiding both underload and overload) [top of inverted U] Support pilots in aviation: High workload situations: complex taxiing, take-off, landing Low workload situations: cross-Atlantic flights Trends in aviations: operating cost reductions, reduced crew operations. We have adapted automation. You let the computer do the work which it's good at and you let the human do what he/she is good at. Fully automated driving may be (overall) safer, less accidents may happen, but there will still be accidents. These will be larger, concentrated, and will have a major impact (also on acceptance). “The combination of responsibility without authority is a formula for extreme stress (and will not be welcomed by the purchasing public)” -Hancock & Warm 1989 There are still some technological limitations to full automatization. In the most difficult situations the system doesn’t work. For example, when it has snowed extremely the camera can’t see the lane. Also, dealing with contradictory information can lead to problems (e.g. normal speed limit traffic signs and electronic speed limit traffic sign). Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 Remember Rasmussen Skill level (milliseconds) → errors because of failing sensors (leading to wrong information) Rule level (seconds to minutes) → pattern recognition. machines can get smarter also by self learning, Rule based AI (If-Then) to Machine Learning (Data driven AI ). But which errors can be afforded to learn from. Knowledge (minutes to hours) → Ethics, act on instinct vs. think through in advance. VANAF SLIDE 62 TERUGKIJKEN Some ethics about automated vehicles Hacking, terrorism, smuggling, robojacking Speeding? Can we get drunk? Braking for animals? Beliefs that human are still better in many tasks (People might not agree with the speed or performance of the automated vehicle) In general: when doing research, be aware of ethics!! Respect for [vulnerable] participants Demands made are reasonable and safe [no exploitation, balance benefits and harms] Participants are free to participate and are informed Respect for privacy and confidentiality Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 Readings 2: Automated driving: Safety blind spots Automated technologies are evolving at an accelerated pace. This paper critically examines the promises for improved safety that are espoused by proponents of AD through a variety of lenses and identifies key considerations in the safety of AD. The underlying goal is not to debunk the safety benefits that can be realized, but to shed light on the challenges that we must overcome if we are to maximize the benefits that are possible. Because not enough attention is paid to addressing the challenges discussed in this paper within the scientific, technical and popular media, we refer to them as blind spots. Some of the challenges discussed in the paper have arisen in other contexts such as aviation, but they have not been brought forward or fully articulated in the conceptualization of AD. This paper draws attention to the need for vehicle and technology manufacturers to apply what has been learned in high-reliability organizations, such as aviation, to avoid automation surprises and to make interfaces user-friendly and intuitive. In addition, governments will need to address the myriad of jurisdictional conflicts that currently exist and develop the regulatory framework for ensuring that the technologies being implemented are truly as safe as they can be. Industry and government will need to jointly develop an underlying strategy to ensure consistency, reliability and functional interoperability of automated vehicles. And academic research will need to play a more compelling key role in helping to guide solutions to safer progress in each of these areas. Three guiding principles emerge from these considerations. 1. It’s not just about the driver, it’s about all the people in the system. 2. It’s not just about designing the human interface, it’s about creating value and the right level of trust. 3.It’s not just about driving, it’s about mobility within the Economy of Things. Until such time as the driver is taken out of the loop completely, road safety may well continue to be a major public health problem, though some improvement is likely. It is posited that current advances in AD technologies are suboptimal in that they fail to address critical blind spots and will likely lead to unnecessary losses and injuries because insufficient consideration is given to integrating the human element into the overall sociotechnical road transportation system. To achieve the promise that digital connectivity and distributed intelligence offers, a more human- centric and coherent approach to mobility design is indicated. Lecture 3: Mental Workload, fatigue Mental workload is central in driving & important in involvement of accidents: distraction & overload. Fatigue: capacity of the driver Mental workload: the difference between the processing resources available to the operator & the resource demands of the multiple tasks => can have consequences for performance Demand lower than capacity: workload low/okay Demand higher than capacity: workload high & performance decreases Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 Availability of resources depend on: individual differences (experience), arousal (state) & motivation (drive) => difference between & within people: hard to assess: show how well performance can be Resource demands depend on: task structure (information processing) & task mixtures (number & type) => the task itself You cannot measure resources. That is what makes it difficult. The availability of resources depends on: Individual differences (incl. experience) → tired, Arousal (‘state’) & Motivation (‘drive’) Resource demands depend on: task structure (information processing), task combinations (number & type) Concepts Complexity of a task increases with an increase in the number of stages of processing that are required to perform a task Difficulty of a task is related to the processing effort (amount of resources) that is required by the individual for task performance Context is important think of walking on a 10cm wide plank (on the ground or at a huge height) On the one (1) hand- Task demands › The task one has to complete › Complexity, stages of processing On the other (2) hand- The Operator › Capacity, resources: what you have available, “able to use...” › Barrel metaphor Interaction = Task(demand) X Operator (ability) + resources are often not fixed. For the operator there are long term factor (suc as expert vs. Novice or young/old), but also temporary factors (fatigue, ilness, alcohol) Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 You can’t measure mental workload that easily. Maybe for the tasks, but it differs too much for the operator. High Mental Workload - › Mental workload can be high, both as a result of o high task demands, a complex task o a reduced state, reduced capability to deal with the task demands - › in car driving: Many factors have an influence A. Driver State affecting factors - monotony leading to boredom - fatigue - sedation by drugs, alcohol - stimulant drugs - time pressure B. Driver Trait factors - experience - age - strategy C. Environmental factors - road environment and traffic demands - vehicle ergonomics - automation - feedback Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 Reality is more complex. The reasons for this is that we have to take Mental Effort into account: - › “trying hard” - › keep performance level steady (protect) - › voluntary process - › performance is achieved/continued at costs Two types of effort: - › COMPENSATORY effort: o counteract reduced driver state (e.g. I really should try to stay awake) - › COMPUTATIONAL effort: o deal with increased task demands (e.g. try to deal with all the information flow that’s comes to you) (controlled mode processing) Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 x-as = demands Arrows: manage the demands by trying hard. Assessing Mental Workload (MWL) Measures have to be: 1. Sensitive to changes in MWL 2. Selective, i.e. only sensitive to MWL 3. Stable & reliable (transferable) 4. Non -intrusive to the primary task 5. Accepted by the operator Measures can be: 6. Diagnostic or generally sensitive → guide you to the problems that are occurring 7. Implementation requirements → how difficult is it to manage a vehicle Types of Measures A. PERFORMANCE measures A1. Primary task Longitudinal (speed and speed control, like how constant is speed & headway control) Lateral (lane position, steering wheel movements)) A2. Secondary task —> “Fill-up capacity” one of the tasks will suffer A21 Added → artificial tasks, e.g. calcutations next to keeping the car in the lane A22 Embedded (mirror looking behaviour; maybe this task suffers when you are Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 distracted by the added task) From research about the driving telephone mirror behaviour suffered when people could call in the car. Also; the busier the road is the more the glances decrease. All performance measures are always speed/accuracy measures B. SELF -REPORT measures; fo example multidimensional ratings on mental/physical workload, time pressure, performance, effort & frustration C. PHYSIOLOGICAL measures ECG (heart rate) —> increase in mental workload = increase in workrate Respiration (frequency & amplitude) EMG (energy) Behaviour (eye movement (pupil dilation, eye lid closure, blink rate), gaze duration and frequency of scans) Divided attention is a risk. Some studies Study behavior eye movements: people spend some time at irrelevant stuff, but most of the time they spend time on relevant behavior. This increases if effort increases. Study ECG: over the 3 weeks, you get lower levels of beats per minute. People get used to operate the phone in driving a car. CYCLING Dual task performance, results: ○ Speed: The more demanding the task on the phone, the more people slow down, people will compensate in speed. ○ Music no effect in speed ○ Position on the cyclepath: you take more distance from the kerb when the tasks on the phone ○ Swerving: the higher the demands of using your phone, the more you swerve ○ Periphereal detection: the moment when they were using their phone, they detect less objects. ○ Honk a horn and listening to music: in ear buds, many of them heard nothing, only a small amount of participants stopped. Conclusion for cycling while listening to music, calling, texting 1. Lower speed for calling & texting 2. stay away from the kerb (centre of the path) 3. swerve more 4. miss more stimuli in the periphery Texting is worse than calling Even larger effect for texting on a touchscreen phone Making use of higher workload → adapt road layout to reduce speeding Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 Study on speed: speed went down when layout was adapted to reduce speed. Assessing Mental Workload - › Need to interpret and integrate information from different measures above. But not all measures are equally sensitive (not sensitive in the same bandwith) Fatigue: Hypovigilance / Vigilance decrement / Sleepiness/ Tiredness/ Drowsiness/Lack of energy Subjective experience of tiredness and unwillingness to continue working” Lower attention level Desmond & Matthews (1997): “disrupts matching of effort to task demands, such that the fatigued driver fails to regulate effort effectively when the task appears easy” Three types (sources) of fatigue [May and Baldwin 2009] › Sleep-related fatigue (circadian rhythm) › Active task related fatigue (exhaustion after high demand) › Passive task related fatigue (monotony, boredom) Fatigue Role in 5 – 35% accidents (Norway: 8%, Australia 35%) + serious accidents (no avoiding action) Young male drivers (20-25) Shift workers People with sleep problems % of crashes because of fatigue when people have been longer , also there are more crashes during night time. This can be also because of darkness (visual illusions), monotonous road, quiet, alcohol etc. Also it’s normal to have an increase in sleep urge during th afternoon (post-lunch dip), this is called circadian rhythm. So reaction time will be high and alertness will be higher during that time and during the night. Fatigue can be measured by physiological measures (EEG, ocular/eye measures (eye-lid drooping, aka PERCLOS measures), permanent measures (steering wheel, weaving, speed) and subjective measures such as the karolinska sleepiness cable Results: perclos is the best measure to use. Fatigue detection in cars often fail. Signal detection theory: Goal: discriminate signals from noise TERUGKIJKEN SLIDE 97 to 100 Fatigue related crashes are more likely during light time, on dry pavement & near truck stops Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 In Europe there are maximum driving times (56 hours a week, 90 hour per 2 weeks), maximum uninterrupted driving time = 4-5 hours with a break time of a minimum of 45 minutes. During night time there are different rules as well. When you wake up early the sleep need during the day also increases. So night work / early morning shifts / to many shifts without time off all cause a shift in sleep quality and sleepurge/need. Free time between shifts of at least 11 hours is recommended, also: napping, avoid night/early shifts, rapid rotation and direction of this rotation clockwise. If above 45 years transfer to day work and avoid working during prime social time (during evening/weekend) When you start travel time fatigued there is a strong decline in attention after 20-35 minutes, after shortage of sleep 65% of the accidents happen within the first hour after starting the ride. Subjective perception is another factor —> optimism about begin able to continue, people deny that they are tired. Countermeasures: Opening a window has no effect Going for a walk and listening to music had an effect for 10 minutes Drinking coffee has an effect for approximately 1 hour Taking a power nap has effect as well. Dit terwijl het grootste gedeelte moeheid probeert t voorkomen door het raam te open of de radio aan te zetten. En maar een klein gedeelte door een powernap te nemen. Other countermeasures; rumble strips, more variation in environment, education (what is effective, plan journeys, rest in advance, dangers), vehicle technology Readings 3: Measurement of mental workload A research thesis by Dick de Waard titled "The measurement of drivers' mental workload," which was published in 1996 at the University of Groningen. Here's a summary of the key points in the text: Background: The introduction highlights the importance of studying mental workload, especially in tasks where operators, such as aircraft pilots and drivers, are required to perform complex tasks. It mentions that high workload can lead to errors and safety issues, and even economic interests can raise questions about workload. Objective of the Thesis: The main focus of the thesis is to explore how to measure the mental workload of drivers. It emphasizes that driving is a dynamic task influenced by both external factors and the driver's behavior. Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 Theories Relevant for Mental Workload: The text introduces various theories related to mental workload, such as capacity theories, resource allocation, and effort-based theories. These theories help explain how individuals allocate mental resources to tasks and how they adapt to different levels of workload. Driver Workload: The text discusses the complexity of the driver's task, which includes strategic decisions (route planning), maneuvering (reacting to traffic situations), and control (vehicle handling). It highlights that demands on drivers can come from both inside and outside the vehicle, and factors like monotony, fatigue, and in-car technology can affect workload. Factors Affecting Workload: The text provides a list of factors that can either increase or decrease driver workload. These factors encompass driver states (e.g., fatigue), driver traits (e.g., experience), and environmental factors (e.g., road conditions and automation). Chapter 2 discusses a model of mental workload, task performance, and task demands. Here's a summary of the key points: Introduction: Chapter 1 introduced concepts related to driver mental workload. It distinguishes between objective task descriptions and subjective task interpretation, which affects task demand. Mental Workload: Mental workload, a central concept in the text, is directly influenced by task difficulty. It reflects the allocation of processing resources based on task difficulty. Directly influence by → Task Complexity: Task demand is divided into different operating stages that determine task complexity. Performance is an objective measure of how well a task is done, while task difficulty is subjective and depends on various factors. Regions Model: The text discusses a model with three regions (A, B, and C) that relate task demand to task performance. In region A, performance remains stable with increased demand. In region B, performance decreases with increased workload. Region C represents extreme overload. Workload Sensitivity: The text suggests that primary-task workload measures are sensitive to variations mainly in region B, where performance decreases with increasing demand. Self- Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 report measures may be sensitive in region B and reveal overload in region C. Inverted U Hypothesis: The text references the inverted U hypothesis, which relates arousal levels to performance. This hypothesis helps complete the region model. Deactivation Region: The model is expanded to include a D-region on the left end, representing low-demand tasks that can increase workload due to monotony or boredom. Workload Redline: The concept of determining a workload redline is discussed, representing the point at which workload becomes excessive. It's suggested that this redline should be placed at the transition from region A2 to A3, rather than A3 to B. Resource Allocation: The model accounts for the allocation of resources for various types of demands (e.g., auditory, visual, central), following Wickens' multiple-resource theory. Measuring Driver Workload: The thesis aims to evaluate different techniques for measuring driver workload, their sensitivity, and their application in traffic research. In summary, the text presents a model that relates task demand, task performance, and mental workload, with a focus on driver workload measurement techniques and their sensitivity in traffic research. The concept of a workload redline is discussed as a critical point in determining excessive workload. Chapter 3: Here's a summary of the key points: Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 General and Specific Properties: Measures for assessing mental workload have various properties, ranging from general aspects like equipment requirements to specific ones like validity. Sensitivity: Sensitivity refers to a measure's ability to reflect changes in workload. It should be defined within the context of the region of performance. For instance, a primary task performance measure can't be sensitive to mental workload in regions where performance doesn't change (e.g., region C or A). Diagnosticity: Diagnosticity relates to a measure's ability to identify the specific resources or causes of workload. Highly diagnostic measures can pinpoint the sources of workload, while less diagnostic ones may only reflect general demands. Primary-Task Intrusion: Primary-task intrusion measures the degree to which a technique disrupts primary task performance. Techniques like secondary tasks can have a significant degrading effect on primary task performance. Implementation Requirements: This property considers practical constraints, such as the need for specific equipment or operator training. In field studies, these requirements can become important factors. Operator Acceptance: Operator acceptance measures the degree to which the technique is accepted by the operator. Techniques that are less intrusive or have face validity tend to be more accepted by operators. Selectivity: Selectivity refers to a measure's sensitivity specifically to mental workload and not to other factors like physical load. A measure can be selective or sensitive to multiple factors, depending on the context. Bandwidth and Reliability: Bandwidth and reliability concern the measure's ability to provide reliable workload estimates both within and across tests. It's essential for a measure to remain stable in different test environments. Interdependence Characteristics: These characteristics are not independent of each other. For example, diagnosticity restricts sensitivity, and high diagnosticity presupposes selectivity. Interdependence is particularly relevant in the case of secondary tasks. Desirable Characteristics: The most desirable characteristics for workload measures are high sensitivity, a wide bandwidth, high reliability, and low primary task intrusion. Diagnosticity can also be important, especially when identifying specific stages of information processing. Text about self-report skills Mental vs. Physical Workload: Mental workload cannot be quantified in the same way as physical workload, which involves measuring forces required for tasks. Mental workload depends on task demands and the operator's capacity. / Capacity and Task Demands: Mental workload results from the interaction between task Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 demands and the operator's capacity. Mental capacity cannot be measured directly, but the proportion of capacity used for a task reflects workload. Operator Differences: People vary in their ability to perform mental tasks, influenced by innate differences and experience. Tasks can be skill-based, rule-based, or knowledge-based, affecting the demands on mental capacity. / State-Related Factors: Short-term factors like lack of sleep, illness, or emotional state can impact mental workload. Operators can invest extra effort to maintain performance, even when affected by these factors. Strategy and Mental Workload: The choice of strategy can influence mental workload. Efficiency-thoroughness trade-offs (ETTO) occur when thoroughness trades off with efficiency, affecting mental workload. One Measure Limitations: Relying on a single measure, such as a subjective questionnaire like NASA-TLX, to assess mental workload is simplistic. Mental workload is a dynamic concept influenced by both performance and subjective experience. Performance Protection: Performance can be protected even in high workload conditions through increased effort, which may not be visible externally. One measure alone may not capture this complexity. Self-Report Scale Limitations: Self-report scales have limitations, including issues with reliability, retrospective reporting, and cultural influences on responses. / Conceptual Usefulness: Mental workload, though not a tangible entity, is a concept that can be indirectly assessed. It is an operational tool useful for understanding and discussing cognitive demands. Recommendations: Multiple measures, including performance, self-reports, and physiology, should be considered to assess mental workload comprehensively. Dissociation of measures provides insights into task strategies and performance protection efforts. In summary, mental workload is a complex concept influenced by various factors, and its assessment requires a multifaceted approach beyond a single subjective measures/. Text about fatigue: This text discusses research on fatigue measurement methods in the context of driving. Here's a summary of the key points: / Research Overview: The study identified 53 research studies that used 25 Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 different fatigue measurement methods. The analysis focuses on the 14 most frequently used methods. / Challenges in Fatigue Measurement: Many studies combine various causes of fatigue, making it challenging to specifically determine monotony-related fatigue. To study monotony-related fatigue, it's crucial not to disrupt the monotonous driving scenario with a stimulus-intensive test setup. / Objective Condition Data: Parameters that are not significantly influenced by test subjects, such as physiological measures (e.g., EEG and heartbeat activity), provide reliable and accurate results for fatigue assessment. EEG, while effective, can be challenging due to data volume and device discomfort. / Skin Conductance: Skin conductance measurements can be inconsistent due to intraindividual variations and should not be relied upon as a sole parameter. Proper electrode placement is essential. / Eyelid Closure: Methods like PERCLOS and blinking parameters may not detect daydreaming with open eyes, common during monotonous drives. They are more suitable for sleep-related fatigue. / Gaze and Head Movements: Gaze and head movement parameters are suitable but only detect fatigue in late and pronounced states, making them less sensitive for shorter journeys. / Subjective Condition Data: Subjective methods that quantify current fatigue levels are often used but should be measured regularly during the test drive to provide meaningful results. Single comparisons before and after the test drive may not be reliable. / Sleepiness Scales: Among sleepiness scales like KSS and SSS, KSS is slightly preferred due to more frequent use and reliable results. ESS is not suitable for fatigue measurement. Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 / Performance Data: Performance parameters related to vehicle guidance (e.g., lane keeping, speed, SWM) are widely used. The standard deviation of these variables provides meaningful results and can be used effectively to study monotony-related fatigue. / Secondary Tasks: Using secondary tasks as indicators of fatigue can be problematic as they introduce new stimuli, breaking monotony and affecting the results. / Conclusion: The paper provides a comprehensive overview of the state of the art in fatigue measurement methods for driving research. While no single method consistently yields positive results, combining multiple methods, including objective, subjective, and performance data, is recommended for robust fatigue assessment. In summary, this research highlights the challenges and recommendations for studying monotony-related fatigue in driving and provides guidance on selecting appropriate measurement methods for future research. Lecture 4: 3 E’s (Education, Enforcement, Engineering) 1. Education a. Licencing & driving training Skill/conrol level → makes people more skilled drivers, not necessarily safer Strategic/manoeuvre level → danger recognitions tests Hoe ouder mensen worden, hoe meer rijervaring en hoe minder ongelukken. Dit is ook wel de experience paradox (mensen met meer ervaring gaan roekelozer rijden) → This may seem counterintuitive because you might expect experienced drivers to be safer and more cautious. However, several factors contribute to this paradox: familiarity, risk compensation & overconfidence. ○ Parental supervision → solution for getting more experience in a safe way, because young drivers have poor hazard perception ○ Younger drivers often drive older cars Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 b. Campaigns Fun? Fear? (kill a kid, kill a family) Confronting? (... people died), understandable Message can be informative/emotional/rational/shocking Fear appeals can motivate but can also lead to defence response (denial, neutralising, ridicule the message) → has most effect on cognition not on behaviour. ○ Better: threat + coping → show that there is something you can do ○ Like: bag a phone, not a body(=response efficacy) ○ Differs within cultural tradition (NL:humor, Aus: casualties0 The message relevant effect ○ Short-term: Negative Emotional Appeals Short-term messaging often utilizes negative emotional appeals to grab the audience's attention or evoke a quick response. These appeals might use fear, anxiety, or other negative emotions to persuade or motivate people to take immediate action. For example, an anti-drunk driving campaign might use graphic images of accidents and the potential consequences of drunk driving to elicit fear and discourage immediate risky behavior. ○ Long-term: Positive Appeals In the long term, messages often transition to more positive emotional appeals. These appeals aim to establish a more lasting connection with the audience and promote behavior change over time. For example, instead of using fear to deter drunk driving in the long term, a campaign might shift to positive appeals, focusing on the benefits of responsible and sober driving, such as saving lives and avoiding legal troubles. Social norms campaigns can be effective in changing attitudes towards road safety behaviour, but: focus on detection appears more effective than focus on a risk/injury. Personal communication and road side delivery are associated with greater accident reductions + be innovative 2. Enforcement Handhaving: snelheid controle, verkeerslichten, gordel, helm, flitsers, camera’s & alcohol controle Deterrence theory → activity happens in case of positive utility, dus mensen gaan bijvoorbeeld harder rijden als ze weten dat ze niet gepakt kunnen worden (denk aan gebruik van flitsmeister) → mensen ondernemen actie wanneer ze verwachten positief nut ter ervaren ○ Benefit form criminal activity is bigger than the sanction of criminal activity ○ Probability of detection → Probability of detection (chance of apprehension) and the perceived certainty of punishment are conditional ○ detected= gevangen Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 Drinking and driving grote oorzaak van ongelukken + deze personen worden vaak ook opgepakt, te hoge snelheid is ook risico factor maar mensen worden hiervoor minder vaak opgepakt. Een kleine snelheidsverhoging kan al groot verschil maken tussen ernstig ongeluk of fataal ongeluk Halo effects → ○ Distance Halo: The Distance Halo effect relates to drivers' perceptions of distance. It refers to a cognitive bias where drivers tend to underestimate the actual distance between objects, vehicles, or obstacles on the road. This bias can lead to problems such as misjudging the space needed for passing or merging, making decisions based on inaccurate distance estimations, and potentially causing unsafe driving maneuvers. ○ Time Halo: The Time Halo effect, on the other hand, pertains to drivers' perceptions of time. It refers to the cognitive bias where drivers may underestimate or overestimate the time it takes for certain events or actions to occur while driving. This bias can impact decisions like when to brake, accelerate, or make lane changes. Drivers may respond too late or too soon, which can lead to unsafe situations or traffic conflicts. ○ Visible speed policy appears to be the most effective for reducing violations and accidents. Stationary enforcement in unmarked vehicle is also effective ○ Average speed enforcement (traject controle) → homogenised flow and accepted by the public ○ Traffic light camera’s result in less light running but increase in rear-end crashes. ○ Waarschuwings borden van: rood licht controle → proportion of drivers that stop increases ○ Steeds meer controles die controleren op telefoon gebruik in de auto 3. Engineering → most effective + expensive of the tree E’s Also in-car technology: ADAS, active safety measures such al abs and electric power steering (preventation of accidents) or passive safety measures such as safety glass, seatbelts (after accidents) Self-explaining roads → the design of the road evokes correct expectation and driving behaviour from road users ○ Bicycle street, groene 100 streep Road lighting Horizontal curves Pedestrian safety Shared space (woonerf / 30 zone) → removing guidance = uncertainty = slowing down = accident and injury reduction Innovation: divergent diamang interchange → reduces number of passages at junction to motorways and saves space Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 Dynamic speed bump → to reduce speed Readings 5 Education The text discusses the evaluation of pre- and post-license driver training and its impact on subsequent crash rates. It highlights that many studies have failed to find a direct link between driver training and reduced crash rates. Some have taken this as evidence that driver training may not improve safe driving behavior. However, the text points out two key considerations: Driver training can be evaluated from two perspectives: improving driving skills (e.g., vehicle control) and improving on-road safety. Some training may enhance skills but not safety, while others may improve safety without affecting skill. Ideally, driver training should focus on both aspects, emphasizing that safe driving involves more than just vehicle control. / Different types of training may be more effective for novice drivers at different stages of the licensure process. Pre-license training should emphasize safety over skill development, and short-term training programs should not replace substantial on-road experience. Post-license training should be carefully designed, considering the risk of overconfidence and the need for cognitive training to improve hazard perception. The text also points out that previous evaluations of driver training programs have limitations, which hinder the reliability and validity of their findings. To produce more valid and generalizable results, future evaluations should overcome these limitations, employ randomized controlled designs, control for relevant variables, record training content and extent, and use more sensitive outcome measures. In conclusion, the text suggests that certain forms of pre- and post-license driver training can be beneficial for skill development and may potentially improve novice drivers' safety. However, safety benefits have not been adequately evaluated for some training types. The main issues in young novice drivers' unsafe behavior may result from factors like overconfidence, ignorance, and poor hazard perception, which could potentially be addressed through better driver training. The text emphasizes the need to make driver training more effective at promoting safe driving behaviors and to ensure that it is appropriately evaluated to determine its impact on the safety of young novice drivers. TEXT 2: The discussion on road safety campaigns has primarily focused on current practices, their effectiveness, and potential improvements. However, the article shifts its attention to the future, exploring insights from social psychology and economics. These insights challenge the assumption that humans are always rational decision-makers. The article argues that many behaviors are automatic, driven by habits, feelings, biases, and circumstances. Automatic behaviors play a role in various aspects of life, including driving. To influence Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 behavior effectively, campaigns should consider the power of priming, where certain stimuli can trigger specific behaviors. This approach doesn't require active processing of the campaign message. For traditional campaigns with information and reasoned arguments, it's crucial to choose images and words carefully, as they can have unintended effects. → In the context of road safety campaigns mentioned in the article you provided earlier, the power of priming suggests that carefully selecting images, words, or other stimuli that activate desired behaviors or attitudes can be a powerful tool for influencing people's actions, often without them realizing it. Additionally, the article discusses the role of modeling and social influence. People tend to mimic behaviors they see in others. Therefore, campaigns should showcase desired behaviors as normal and appropriate. The article also mentions framing the message, highlighting the prospect theory, which suggests that people react differently to information framed as potential losses or gains. The framing of the message can impact how people respond to it. Lastly, the article suggests that informal education is essential. Beyond the primary target audience, campaigns should consider how their message might influence others who, in turn, can influence the main audience. For example, encouraging experienced drivers to set an example for novice drivers or involving parents and children in promoting safety behaviors. In conclusion, the article emphasizes the need to move beyond the assumption of rational decision-making and embrace the automatic nature of many behaviors. Understanding these principles can enhance the effectiveness of road safety campaigns by carefully selecting stimuli, framing messages, and considering the broader impact on various audiences. Enforcement Section control (point-to-point speed cameras, sport traject controle) was found to have a more substantial effect on reducing crashes compared to traditional speed cameras. It resulted in a 30% reduction in total crashes and a 56% reduction in crashes causing killed or seriously injured (KSI) casualties. While some "kangaroo driving" behavior (braking and accelerating) near camera locations was noted, it did not lead to adverse effects on safety. Overall, both speed cameras and section control achieved significant speed reductions, and the crash effects were in line with or greater than what would be expected based on the speed reduction effects alone. Engineering While the study's short evaluation period and location-specific nature raised some concerns, it concluded that the project effectively achieved the goal of creating a multi-level road hierarchy with distinct road categories and speed profiles. The SER (self explaining roads) treatments enhanced road aesthetics, reduced traffic speeds, and aligned residents' speed perceptions, while remaining cost-effective and consistent with the area's character. The research investigates a phenomenon called behavioral adaptation, where drivers adjust Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 their speed and lateral position on the road in response to changes in road width. Using a driving simulator, the study explored whether drivers adapt their behavior due to conscious decisions or implicit perceptual processes. The results showed that drivers reduced their speeds on narrower roads, indicating behavioral adaptation, but they were often unaware of the change in road width. Interestingly, even when the road width increased, drivers did not consistently increase their speeds. This suggests that risk perception and its influence on driving behavior are primarily implicit and preconscious. The findings have implications for road safety and countermeasures designed to influence drivers' behavior. The paper provides examples of poor road system design that makes it difficult for road users to behave safely. These issues include inappropriate speed limits, inadequate road design, poor signage, confusing lane markings, and problems with driver-assistive technologies. The text emphasizes the need to expand the road safety focus to better accommodate the needs of road users. Instead of merely addressing the consequences of road user errors, there should be a greater emphasis on reducing or preventing errors in the first place. The text suggests that engineering, education, and enforcement are necessary but insufficient strategies to improve road safety. They must incorporate human-centered design principles to consider how humans naturally behave and expect the road system to function. Education can raise awareness, but it may not be enough to change behavior permanently. Rules and enforcement can lead to temporary compliance and may not gain user support. The text concludes that the road system should be designed to accommodate human behavior and usability, as this can significantly reduce road user errors. It emphasizes the missed opportunity to improve road safety by not addressing the interaction of human road users with other elements in the road system. In a time when progress in reducing road injuries and fatalities is slow, this approach offers an opportunity for improvement. Lecture 5, Driving with mild cognitive impairments Cognition = Brain activities that lead to knowledge, including all means and mechanisms of acquiring information. (Reasoning, attention, memory & language) Cognitive impairment means experiencing impairment in daily life. It can be assessed by: Psychometric neuropsychological tests Clinical interviews Questionnaires Observations The aging society Older population is growing There is a normal cerebral and cognitive decline because of aging, but also a pathological decline, think of Alzheimer’s/Parkinson’s/huntington’s. Risk of pathological decline doubles with every 5 years of age. Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 ○ Often there’s a progressive decline in cognitive functions because of this. In many cases resulting in severe reduction of abilities to perform everyday. Think of travelling to desired places, feeling of independence, social involvement in local community and potential travel for example in emergencies. ○ Out of home activiteit are associated with personal well-being, physical health and lower mortality risk. Driving assessment Behind the wheel examination —> measurements are direct and quantifiable, but it’s not objective and sometimes you’re unable to measure challenging scenarios Instrumented vehicle —> instruments (e.g. camera’s) linked to vehicle inputs. Measurements are objective and quantifiable, but it’s expensive and also unable to measure challenging scenario’s Virtual reality driving simulation —> objective & quantifiable & a safe presentation of challenging scenario’s. But how to translate form simulated driving to the real world? Crash statistics —> strong clinical relevance, but collected after the fact and only if reported to authorities Self-report —> based on driving history. Easy to assess but self-report bias Also -> neuropsychological assessment Speed of processing —> slow processing may lead to fatigue, exhaustion, feelings of depression etc. It gets assessed by ○ Behavioral observations —> ratting scale of attention and behaviour (RSAB); a 5- point report scale. ○ Self-report measures —> Mental slowness questionnaires (MSQ), for example; I have trouble following a conversation. ○ Neuropsychological tests —> like trail making test where speed is a measure of processing speed. Or accuracy within a limited period of time. Cognitive impairment does not always means driving impairment. Therefore it’s important to look both at history of illness, off-road assessment and on-road assessment. Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 Do individuals with ADHD have an increased driving risk? Because of inattention, impulsivity, reduced flexibility and increased processing speed. Self reports show that people with ADHD are more often involved in traffic accidents. But those reports are often unreliable because of underreporting of symptoms and overestimation of driving performance. Driving simulations show also more crashes and slower reactions times. These simulations are valid, safer of risks in comparisons with on-road testing and there is an exact replication of conditions across participants. ○ But there are also some disadvantages, such as measuring short-term driving skills rather than habits/practices, abilities assessed may not correspond to on- road driving and there are no clear standards yet. On road driving also showed more crashes, speeding and more driving errors. Sensitivity (True Positive Rate) —> Sensitivity measures the ability of a test or model to correctly identify individuals or cases with a particular condition or characteristic (positives) among those who actually have it (true positives). Classified as unsafe to drive : actually unfit to drive. How many of older people unfit to drive do I classify as unfit? Specificity (True Negative Rate) —> Specificity measures the ability of a test or model to correctly identify individuals or cases without a particular condition or characteristic (negatives) among those who truly do not have it (true negatives). Classified as safe to drive : actually fit to drive How many of older people fit to drive do I classify as fit? Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 Research Groningen, Goal: To develop a strategy for determining fitness to drive of patients with dementia in the clinical setting. A combination of interviews, driving simulation and neuropsychological test seems to have useful information about determining driving fitness. Patients with mild form of dementia may be able to drive a car safely Still 42% of the people keep driving after the recommandation. But from the people that do stop driving because of age, the biggest reason is because of a Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 negative recommendation, Interventions / alternatives: Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 Cognitive interventions: Speed of processing training leads to improving visual attention Educational interventions: Identifying potential problems Developing solutions how to address problems Facilitating family discussions related to driving cessation Leads to: better awareness and self-regulation Clinical intervention —> to prevent negative outcome of less mobilisation after driving cessation, like depression or loss of independence. Mandatory readings driving The text explores the growing phenomenon of older individuals retaining their driver's licenses and the complexities that arise as a result, particularly in light of age-related medical conditions such as dementia. These conditions can have adverse effects on the cognitive, visual, and physical abilities necessary for safe driving. Consequently, clinicians are increasingly tasked with assessing older patients' capacity to continue driving safely. This chapter aims to equip clinicians with a comprehensive Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 understanding of the body of research conducted on driving in older adults. The text is organized into several sections, each serving a specific purpose: 1. Introduction: The chapter opens by underscoring the rising number of older individuals who remain active drivers in society. It highlights the dependence of modern society on automobiles and the fact that people are staying on the road longer into their later years. As a result, the number of drivers over the age of 65 continues to rise. 2. Concepts and Challenges in Driving Research: This section provides a foundational framework for driving research. It introduces key concepts and challenges associated with studying driving in older adults. It acknowledges the complexities involved in assessing an individual's driving capabilities. 3. Effects of Healthy Aging on Driving Performance: The chapter delves into the driving performance of older adults without neurological impairments. It offers insights into common driving errors made by older adults, particularly at intersections, stop signs, and during lane changes. The text also explores the intricate relationship between cognitive abilities and driving performance in healthy aging individuals. 4. Characterization of Older Drivers with Neurological Disease: This section narrows the focus to older drivers who have neurological conditions, such as Alzheimer's disease, Parkinson's disease, and mild cognitive impairment. It acknowledges that these conditions can further complicate driving safety and provides a glimpse into the research conducted within this specific demographic. 5. Clinical Application: The final section aims to bridge the gap between research findings and clinical practice. It discusses how clinicians can use the research outcomes in their work, particularly in the field of clinical neuropsychology. It offers practical guidelines to help clinicians evaluate the driving capacity of older adults, balancing safety concerns with the individual's need for mobility and quality of life. Throughout the text, the authors emphasize that while older drivers as a group do not exhibit a higher overall crash risk compared to younger drivers, they do tend to make specific driving errors that increase their crash risk. These errors often occur at intersections, stop signs, and during lane changes. The role of cognitive factors, including memory, attention, and executive functioning, is emphasized, as these abilities play a crucial role in driving performance, both in healthy older adults and those with neurological conditions. In conclusion, the text addresses the multifaceted challenges and considerations associated with older drivers. It provides valuable insights into the extensive body of research conducted on this topic, with a strong focus on cognitive aspects and practical applications for clinicians tasked with evaluating the driving capacity of older adults. Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 Characterization of Older Drivers with Mild Cognitive Impairment (MCI): The chapter acknowledges that research on the driving performance of older adults with MCI is less developed compared to AD and PD. MCI is defined as an intermediate stage of cognitive decline associated with a higher risk of dementia. One study involving adults with MCI found that they performed comparably to cognitively healthy older adults on cognitive tests but showed some difficulties with specific driving maneuvers. Application to Clinical Neuropsychology: Clinical neuropsychologists often play a crucial role in evaluating the driving abilities of older adults. The chapter emphasizes the need for comprehensive cognitive assessments in the driver evaluation process and suggests using a combination of neuropsychological measures associated with driving ability. It discusses the importance of considering clinical and cognitive risk factors when evaluating fitness to drive in patients with PD, including disease duration and severity, visual acuity, and cognitive performance. Additional Risk Factors and Sensitive Functional Assessment: The text highlights that other medical conditions, vision problems, respiratory diseases, musculoskeletal conditions, and medications can also impact driving performance. It emphasizes the importance of assessing patients' insight into cognitive impairment and their awareness of functional ability. The role of sensitive functional assessment, such as the Behind-the-Wheel (BTW) exam, is discussed in evaluating driving abilities, but it is noted that it may not capture all aspects of driving performance. Interventions and Recommendations: The chapter suggests discussing potential driving cessation with older patients early on, especially those with progressive conditions like dementia. Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 It emphasizes the importance of educating patients and their families about the impact of relinquishing a driver's license and finding alternative transportation solutions. Driving education and interventions for modifying driving habits are mentioned as potential strategies to help older drivers maintain safe driving for longer periods. In conclusion, the chapter underscores the multifaceted nature of assessing driving abilities in older adults, particularly those with PD and cognitive impairments. It highlights the role of clinical neuropsychologists in conducting comprehensive evaluations, considering cognitive and clinical factors, and making informed recommendations regarding driving safety. Additionally, it underscores the importance of involving family members, staying updated on state laws, and addressing the complex relationship between cognition and driving performance. Clinical Pearls Know the law. State laws vary in their require- ments for reporting and assessing drivers. Clinicians are strongly encouraged to be familiar with their individual state require- ments (additional resources listed below). Ask about driving. Clinicians should be aware that driving cessation is often a topic of conflict. Too often, older adults do not raise the issue for fear of complete loss of driving privileges. Family members are also conflicted and in many cases are unsure about how to handle/raise the discussion. Clinicians can help minimize this conflict by including ques- tions about driving performance in their regu- lar checkups or appointments. Know what cognitive domains are most relevant. Although there is not a specific pattern or defined group of tests that 100% predict driv- ing performance, general domains of cognition relevant to driving are identified in the litera- ture. Neuropsychological test selection should be based upon empirical evidence with multiple abilities assessed under the domains of atten- tion, information processing speed, working memory, executive functions, visual-spatial abilities, visual spatial learning, and memory. Be familiar with the clinical driving evalua- tion process. This includes identifying referral procedures and locations offering BTW evalu- ations with Certified Driving Rehabilitation Specialist (CDRS) accreditation. The neurop- sychological evaluation should serve as guide to inform further evaluation of driving ability Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 and should not serve as a substitute for a comprehensive driving evaluation. Be familiar with age-related medical conditions (i.e., dementia, stroke, seizures) that affect driving. Communication with the treat- ing physician (i.e., neurologist, cardiologist) can help educate colleagues of the need to consider driving capacity. Be on the lookout for medication effects. Given the high number of medications commonly used by older adults, clinicians should con- sider the effect (individual or combined) of medications on driving behavior. Medications altering cognition, alertness, increasing fatigue, drowsiness, or altering sleep patterns may warrant consideration. Potential driving cessation should be dis- cussed as early as possible. It is often beneficial to include significant others or additional family members in this dialogue as they may provide additional insight into driv- ing performance. Familiarize yourself with transportation options. Clinicians have a responsibility to counsel and educate the patient and his or her family on the impact of relinquishing a driv- er’s license. Being prepared with appropriate referrals (i.e., medical transportation services) or community information (i.e., transit sched- ules) can help adults begin to explore/plan alternate methods of transportation. Consider interventions. Interventions can benefit individuals who do not need to relin- quish their driving privileges but require modification of their driving habits. These interventions can range from structured approaches (i.e., improving field of view) to more practical recommendations, such as restricting or limiting driving. Lecture 6: Drugs and driving Why do people drink/drive All police checks now include standard checks for drugs Disinhibition model → when we drink we can’t control ourselves so we show our true selves, can’t control inhibitions fall away. Sober an individual inhibits it's true self Myopia model → if we are sober individuals can consider a wide range of values, rules and concerns. Drunk: I need, I feel, I want NOW. People can’t see beyond the now, role of context is important. The effect of alcohol on driving 1. Epidemiological results → high alcohol % is correlated with higher accident risk 2. Driving related tasks → high alcohol percentage = worse speech, reduced ability to brake / maintain road position, slower reaction time, difficulty Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 steering 3. On road studies → expontential higher risk risk of accident after 0,5 promillage 4. Simulator driving → SDLP increases (lateral control), RT increase in the car following task How we study effects of alcohol on traffic safety Epidemiological research ○ Prevalence of drunk driving population → survey research and roadside testing. ○ Prevalence of driving under influence related accidents → survey questions + hospital registrations ○ Accident risk calculation Culpability studies → did this drunk person cause the accident vs. is this drunk person not responsible for accident → odds ratio = accident rate Case-control studies → odds ratio of hospitalized dui drivers, vs. control group of dui in general driving population Borkenstine curve: exponential increase in accident risk: ○ Problem with those studies → causality, but does give an indication. Experimental study Driving related tasks → alertness, attention, memory, risk taking, psychomotor speed ○ Problem: is this a real test for risk taking, because people do not really experience risk, skills are tested in isolation, not realistic ○ pro’s : easy to administer, cost efficient, widely available On road driving → road tracking test or car following tests → lateral position, how much do people swerve, how much variation in speed, how much distance do people keep ○ Pro’s → representative of real world, con’s → not always possible (ethical constraints) + not always possible to assess all risk taking Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 Conclusion Drink driving limits differ between countries. Psychoactive substances Drugs and medicines Determine drug impairment ○ 1. Test if the effects of the drug differ significantly from placebo ○ 2. Test relevance of this effect by comparing it to the effects of a benchmark drug (a drug of which we know the effects → think of alcohol) Because for alcohol we know the related accident risk, such as SDLP But not all drugs are sedative, some are stimulating Although drugs are illegal, use is normalized at festivals, 25% of the drug users at festival indicated to drive home afterwards Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 Driving simulator tasks → also road tracking and car following test. Interaction with other traffic participants is possible. Gap acceptance test (risk assessments) ○ Pro’s → easy adminsitrable, low costs, interaction with other traffic, repeat situations, con’s → simulator sickness, how representative? Risk taking? Effect of cannabis on traffic safety Effect THC ○ Positive effect: altered perception, relaxation, increased awareness of sensations ○ Possible effects: anxiety, panic, dissocation ○ We see a increase in SDLP Prevalence around 1%, quite common to use → accident risk increased in 2 fold Drugs are often combined → combination with alcohol large effect on SDLP, so on vehicle control. People react to drugs in different way, effects are smaller on heavy users, that’s why a THC limit is difficult to assess. Effects on Simulator driving → SDLP increase with higher dosages, RT increases in car following, other driving tasks not affected Stimulant drugs likes (cocaine / XTC) Improved neuropsycholoigcal tasks → better reaction time, impulse control and tracking performance Sometimes even improved performance. Discrepancy between epidemiological research and lab research. Amphetamines Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 increase risk taking. Vehicle control is better but risk people take are larger Reason to take drugs differ (can be used for medication as well) Medical categorisation based on impairing properties: 1 presumed safe, 2 moderate adverse effects, 3 potentially dangerous In a aging society people take more drugs and keep participating in traffic ○ 2.7 million mensen aan de antidepressiva (benzodiazepines) Reduce alertness, impairments depend on dose, time after intake, half life of the drugs (hoe snel wordt medicatie afgebouwd in je bloed) Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 Multidrug: Gedownload door Linde Brunink ([email protected]) lOMoARcPSD|7065412 Minor Tranquilizers, Benzodiazepines, and Other Medications: These substances have legitimate therapeutic use but can impair mental and motor functions, leading to drowsiness and light-headedness. However, the impact varies, and some studies suggest that some drivers can perform better when treated with certain medications. Lecture 7: Human factors and VRU Gedownload door Linde Brunink ([email protected])

Traffic Psychology Notes PDF

Document Details

Tags

Related

Summary

Full Transcript

Upgrade to continue