A Generalised Methodology for the Investigation of Human Behavioural Responses to Hostile Attacks
File(s)
Author(s)
Shipman, Alastair
Type
Thesis
Abstract
Terrorist attacks have become more frequent and more deadly in recent decades. Understanding the movement responses of the individuals involved can save lives.
Pedestrian dynamics is an area of research and industry that investigates the movement of people. Using pedestrian dynamics to predict the responses to terrorist attacks could be hugely beneficial to the survival of those involved.
This PhD developed a generalised methodology for investigating movement responses to terrorist attacks, accommodating the wide range of attack vectors seen in modern terrorism and providing a detailed summary of the data that should be obtained. Initial work analysed the overlap between pedestrian dynamics and terrorism.
This concluded that current interdisciplinary research is not sufficiently advanced to allow for accurate predictions of pedestrian movement responses to terrorist attacks, due to a lack of usable data given the sensitivity of previous
incidents and limitations on experimental methodologies.
Two sets of experiments were then performed, examining responses to knife-based attacks by an unexpected and hostile individual. These experiments were split into physical reality (PR) and virtual reality (VR). The PR experiments
were performed in December 2018 with 80 participants, and the VR experiments in January-February 2020 with 55 participants.
The experiments obtained positional, physiological, psychological and memory-related
measurements. Analysis of these datasets investigated the participant responses to hostile aggressors, including the impact of demographic,
emotional state, and the similarity in responses between the VR and PR paradigms.
The analysis on the psychological and physiological datasets proved a stressed
response to the hostile individual, resulting in the first ever experimental dataset for these. The subsequent analysis on the positional and memory-related datasets provided key inputs for pedestrian dynamics models and concluded that VR is appropriate as a data-generating tool for human behaviour in emergencies.
Pedestrian dynamics is an area of research and industry that investigates the movement of people. Using pedestrian dynamics to predict the responses to terrorist attacks could be hugely beneficial to the survival of those involved.
This PhD developed a generalised methodology for investigating movement responses to terrorist attacks, accommodating the wide range of attack vectors seen in modern terrorism and providing a detailed summary of the data that should be obtained. Initial work analysed the overlap between pedestrian dynamics and terrorism.
This concluded that current interdisciplinary research is not sufficiently advanced to allow for accurate predictions of pedestrian movement responses to terrorist attacks, due to a lack of usable data given the sensitivity of previous
incidents and limitations on experimental methodologies.
Two sets of experiments were then performed, examining responses to knife-based attacks by an unexpected and hostile individual. These experiments were split into physical reality (PR) and virtual reality (VR). The PR experiments
were performed in December 2018 with 80 participants, and the VR experiments in January-February 2020 with 55 participants.
The experiments obtained positional, physiological, psychological and memory-related
measurements. Analysis of these datasets investigated the participant responses to hostile aggressors, including the impact of demographic,
emotional state, and the similarity in responses between the VR and PR paradigms.
The analysis on the psychological and physiological datasets proved a stressed
response to the hostile individual, resulting in the first ever experimental dataset for these. The subsequent analysis on the positional and memory-related datasets provided key inputs for pedestrian dynamics models and concluded that VR is appropriate as a data-generating tool for human behaviour in emergencies.
Version
Open Access
Date Issued
2021-02
Date Awarded
2021-08
Copyright Statement
Creative Commons Attribution NonCommercial Licence
License URL
Advisor
Majumdar, Arnab
Sponsor
Engineering and Physical Sciences Research Council
Grant Number
EP/L016826/1
Publisher Department
Civil and Environmental Engineering
Publisher Institution
Imperial College London
Qualification Level
Doctoral
Qualification Name
Doctor of Philosophy (PhD)