Personalizing the theme park:Psychometric profiling and physiological monitoring
Stefan Rennick EgglestoneUniversity of Nottingham
1. The theme park as a target for UMAP
2. Proof-of-concept profiling study in a theme park
3. Questions for future research
Overview
• More than 100 million visits per year
• Little published research
• Few operators (but all large scale)
• Substantial investment in innovation
Why the theme park?
Challenges
Challenge one: The theme park recommender system
Challenge two: Personalised rides
Challenges
A common approach involves building a user profile!
• Information collected before the visit– Psychometric personality profiling
• Information collected during the visit– Physiological monitoring
Research into profile design
Personality profiling overview
Sensation Seeking Scale40 questions
Thrill seeking: 8/10Experience seeking: 7/10Disinhibition: 8/10Boredom susceptibility: 7/10
Big 538 questions
Openness to experience: 9/10Conscientiousness: 6/10
Extraversion: 9/10Agreeableness: 5/10Neuroticism: 5/10
Questionnaire on entry or during on-line ticket purchase?
• Heart-rate
• Skin conductance
• Breathing rate
Physiological monitoring
Affective computing: Analysis of physiological data reveals emotional response
What does physiological monitoring reveal about ride experience?
Research questions
Can personality profiles predict experiences on rides?
1. Negotiate access to local theme park
2. Choose single ride
3. Recruit cohort of participants
4. Profile: Personality tests on entry to theme park
5. Profile: Heart-rate response recorded on single ride
6. Profile: Participant quantifies their experience on the ride
7. Analysis: Relationships between profile and experience
Approach
Oblivion @ Alton Towers
Arousal: How much do you feel alert, with your body pumped up and buzzing, ready for action? (1,9)
Valence:How positive or negative do you feel? (-4,+4)
The circumplex model
Self-report data
Arousal Valence
During drop
What does heart-rate reveal?
0 50 100 150 200
050
100
150
200
Median heart-rate during the before phase
Med
ian
hear
t rat
e du
ring
the
drop
and
cur
ves
phas
e
• Correlations between personality dimensions and self-reports at various places on the ride (r~0.3, all at p=0.001)
– Thrill seeking– Extraversion– Openness to Experience
• Two different ways of using these dimensions to cluster participants into groups who report similar experience
– Thrill seeking– Extraversion and Openness to Experience
Results – personality profiling
• Potential use of heart-rate as a measure of whether visitors are excited or relaxed
• Evidence exists for the efficacy of personality profiling in predicting experience
Conclusions profile design
• Extend analysis to multiple rides
• Consider other attractions in theme park
• How to make recommendations for groups?
• Different physiological measures
• Different psychometric measures
• Considering patterns of queuing?
Further work – theme park
• When is personality modelling applicable in profiling?
• Trade-off between accuracy of model and time taken to fill out questionnaire
• High-value applications?– E.g internet dating
• Issues of multiple identity / personality
• Cross-cultural issues
Personality profiling
