Traffic Psychology and Sustained Mobility Lecture notes
LECTURE 1 16/09/2022
A good psychological theory/model:
- It should be descriptive
- It should predict
o Which should make it falsifiable (i.e. testable)!
Theories and models in Traffic Psychology are generally about driving.
- Driving is generally self-paced
- It requires sustained attention, hazard identification, and motor-coordination
- In general, things go well → often due to the environment being forgiving (e.g. giving you space to
react, others being very responsive, etc.)
There are many theories, of which none are really widely accepted.
These are theories that will be covered in this course:
- Skill Models
- Attitude theories
- Utility theories
- Risk theories
- Safety margins
- Performance models
For now, it’s important that you remember the central elements of these theories; you don’t need to know all
of them by heart.
Skills Models: low level models concerned with reaction time, vision, and the level of driver training.
- Crashes occur when Task Demands exceed Driver Skills.
- Skills Models are too simplistic.
- There was no relationship found (or a very weak one at most) between visual attributes and reaction
time, and accidents.
o So, the things seen as important in these models are not as important as they were thought
to be.
o Biographical and exposure factors are more important.
o Marital status, mileage, traffic conviction record, and socio-economic status are also
important.
Attitude theories: for example, the Theory of Planned Behaviour (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
behaviour
- Intention: what I say I will do
- Behaviour: what I actually do
See the image below for the TPB visualized and applied to the concept of speeding:
,Traffic Psychology and Sustained Mobility Lecture notes
Safe or enjoyable
All my friends do it We intend to and we speed
Empty road or no cameras
TPB applied to drink driving:
- Intention: e.g. I often intend to drive my car after drinking alcohol at a party. Agree/disagree.
- Attitude: e.g. Driving my car after drinking alcohol at the party is: pleasant/unpleasant.
- Subjective norms: e.g. Most people who are important to me would be disappointed in me if I drove my
car after drinking alcohol at the party. Likely/unlikely.
- Perceived behavioural control: e.g. If I decide to, I can easily drive my car after drinking alcohol at the
party. Likely/unlikely.
But: there’s a gap between intention/attitude and behaviour! TBP explains intentions quite well, but it only
explains 10-20% of the variance in actual observed behaviour.
Behaviour may change (due to an intervention for example), but attitudes not necessarily so. → the link
between attitude/intention and behaviour is very weak.
Attitude theories are popular, though.
- They’re often about self-reported behaviour;
- It seems like “common sense”, which actually is the attribution bias at play.
Behaviour in traffic is to a large extent automated: habits play a large role.
Intentions were only significantly related to behaviour when habit was weak.
,Traffic Psychology and Sustained Mobility Lecture notes
Utility theories: humans want to maximise gain and minimise loss.
- Or at least we want to trade these variables off.
- These models are rational decision models.
o But are we really that rational?
▪ Perhaps in some situations;
▪ But not often in actual traffic, since being rational is a slow process.
- Some utility theories:
o Utility Maximisation
o Subjective Expected Utility
o Peltzman’s Driving Intensity Theory
- They’re popular in modelling, in incentives research, and in economics.
Risk/Motivational Theories: arose out of research into Galvanic Skin Response (GSR; the principle on which lie
detectors are based).
- These theories take motives into account, or so they say :)
- Some examples:
o Risk Homeostasis Theory (Wilde)
o Risk Allostasis Theory (Fuller)
o Zero-Risk Theory (Näätänen & Summala)
o Risk/Threat Avoidance Theory (Fuller)
o The Safety Margin Model (Summala)
Risk Homeostasis Theory (RHT):
o Humans want to feel a certain level of risk all the time; target risk.
o We take action if we don’t experience this level of risk (either too much or too little risk).
▪ E.g. by going towards the limits.
▪ So sometimes, introducing safety measures makes it less safe!
• Safety measures cause our level of experienced risk to decrease, so we to
riskier things to get back to our target level.
o This model does not make full testable (and falsifiable) predictions…
o It relies on a constant (accurate) perception of risk.
o It predicts full compensation of safety measures (such as seatbelts, road improvements etc.)
if the underlying target risk of the population is not changed.
▪ But there has been a decline in motor vehicle deaths actually! So these safety
measures did in fact have an effect.
o RHT implies that the population learns from crashes, while only a small proportion of the
population actually experience a crash.
o It is not specific as to through what pathways this occurs, and as to over what timeframe.
o But: we do adapt to changes in our environment!
Risk Allostasis Theory (RAT):
o Risk is not the main thing that is monitored in this model.
o Stimuli are “marked” with emotions.
o Emotions are body states.
▪ Cognition arises from the body.
o But that would mean that people who lack emotion should be irrational, which is not the
case.
, Traffic Psychology and Sustained Mobility Lecture notes
o Important concepts in this theory: task demands and (perceived) task difficulty.
Comparing RHT with RAT:
- RAT is more dynamic, and is a better representation of changing motivational influences on the driver.
- RHT posits a fixed target level of task difficulty or risk.
Zero-Risk Theory:
o This theory argues that people do not like risk, and that risk is hardly ever experienced.
o When we do experience it, it acts as a warning.
o We only experience risk if it surpasses a risk threshold.
▪ But if risk isn’t constantly monitored, how does a risk threshold operate?
o This theory is not easy to test or to form predictions from.
Risk/Threat Avoidance Theories:
o Risk is aversive and avoided.
o Risk means something different to everyone: what is risky depends on experience.
o And risk is not constant!
o Context matters a lot!!
o There are many definitions of “risk”, which is problematic:
▪ Risk as a hazard;
▪ Risk as probability;
▪ Risk as consequence;
▪ Risk as potential adversity or threat;
o Calculation for the level of risk: the probability of an outcome multiplied by the consequence
of that outcome. (It is effortful and biased)
o “People run risks, but they do not take them.” → we’re not always consciously taking risks.
Safety Margin Model:
o This is also a threshold model.
o People are motivated to be comfortable (it’s an unconscious motivation).
o Multiple safety margins/zones are monitored.
▪ Like personal space;
▪ Time-based safety margins: TTC (time to collision) and TLC (time to line crossing).
All these theories fall into one of two categories: theories with a constant perception, or theories with a
threshold perception.
Constant perception models:
- Risk Homeostasis Theory
- Task Difficulty Homeostasis
- Risk Allostasis Theory
Threshold perception models:
- Zero-Risk Theory
- Threat Avoidance Theory
- The Multiple Comfort Zone model (Summala)
