Smart Cities, Smarter Citizens

The rise of digital intelligence and how the internet of things is altering spatial decision-making: mapping the journey from carrying a mobile phone to embracing cognitive prosthetics…

---

I was invited to give the following presentation at the 7th Annual Internet of Things European Summit, held in Brussels on 18th – 19th May 2016. It was the first opportunity to present some of the research I am conducting in to digitally sensing and evoking behavioural change – how the internet of things and real-time data can be used to enhance situation awareness and spatial cognition.

The presentation was titled ‘Using the Internet of Things (IoT) to Enhance Situation Awareness and Spatial Cognition’

The slides have been edited for online viewing, and as always the online version is never as good as being there in person 🙂

I also wrote a follow-up paper for the City & Cognition research workshop held at EPFL in Lausanne, Switzerland during June 2016, titled ‘Smart Cities, Smarter Citizens: The Rise of Digitally Augmented Intelligence, available for download using the link below.

 Smart Cities, Smarter Citizens (PDF, 1.6MB)

The following is a shortened version of the paper.

Introduction

Many city administrations and national governments have published plans and funded projects to modernise urban infrastructure and services, and even to build entirely new cities with digital technology embedded at the core. To create ‘smart’ cities. What the plans rarely discuss is the impact of digitisation on citizens, either in terms of altering how citizens interact with city services or how we are adapting to a world of near pervasive digital connectivity. As with many visions of the future, we imagine great technological feats whilst assuming the social fabric of our world will remain unchanged.

Spatial Decisions

Behaviour results from a person-situation interaction, the outcome from a cycle of ‘intelligence – decision – action’. We react to a situation by processing available intelligence, make a decision based on that intelligence and act on the decision. Actions create new intelligence and so the cycle continues.

Psychological studies identify two different mental systems that control decision-making: one that is fast and one that is slow (Kahneman, 2012). Whilst both systems follow the same pattern of intelligence – decision – action when responding to situations, they process information differently. The fast system functions continuously and responds automatically, generating instinctive ‘gut’ reactions. The slow system is deliberately activated to direct attention to reasoning and problem-solving. This includes checking reactions for errors and requires an effort that is limited by mental capacity.

Spatial interactions exhibit both types of decision-making. When facing an unexpected situation, fast decision-making drives the initial ‘fight or flight’ response. However, when tackling uncertainty, even when there is a sense of urgency, a conscious effort is made to evaluate options. In both cases, decisions can be influenced by available information and how it is presented.

Mobile Knowledge

In a spatial context, before the arrival of mobile phones, available intelligence was limited to previous experience and locally accessible information about current conditions. Remote knowledge could usually only be accessed if there was a fixed telephone line within reach. Radios provided an early form of mobile communication but their use remained limited to specific uses in both professional and amateur capacities. As ownership of mobile phones began to accelerate, it brought about a change in how people access information to inform localised decision-making.

One of the earliest published reports demonstrating the use of mobile phones to alter spatial judgement and decision making was from an analysis of the emergency response efforts at the World Trade Centre towers on 11th September, 2001 (Averill, 2004). People inside the towers were given the standard advice issued by emergency callers: stay put and await rescue. No further information was given about the situation. However, many people were carrying Blackberry devices and were able to contact families who informed them that this was not an isolated incident. This information changed their perspective. Nobody knew whether or not further airplanes would crash into the towers. People decided to disobey authority and leave, including using the elevators as far as possible. They paused along the way to encourage others to leave and assisted the incapacitated.  There was no evidence of panic – the reason given by officials to justify withholding information from citizens in a crisis. The report concluded that approximately 2,500 lives were saved.

Over the following decade, there have been numerous examples of mobile devices and social media being used to provide information to, and coordinate actions by, individuals within urban spaces under extreme conditions. From responding to unexpected emergencies, including natural disasters and terrorist acts, to organised disruption such as flash mobs and protestors occupying public spaces. In the same time period, the use of mobile devices to assist decision-making has become normalised in everyday interactions. In 2013, it was reported that during the previous Thanksgiving holiday period nearly half of US adults used a mobile phone to contact a family member or friend to discuss a purchasing decision whilst in the physical store (Pew Research Center, 2013). Over a quarter reported looking up product prices and checking online reviews whilst in-store.

In modern everyday interactions, there is a growing expectation and assumption that global online information will be accessible from any location to inform and support local decision-making.

Sensory Interactions

Whilst early uses of mobile devices focused on accessing and communicating information verbally and via text messaging, a more recent trend is the use of built-in sensors within mobile devices to track local behaviours and interact directly with urban services. This includes the use of low-cost plastic cards and wristbands for digital ‘contactless’ payments when using public transport and making small-value purchases. Such payments are limited to £30 per transaction in the UK.

The number of contactless payment transactions in the UK has increased by 228% in the past 12 months (UK Card Association, 2016). The behavioural benefits include ease of movement, particularly at entrances to transport hubs and shops which are known to be crime hotspots for pickpocketing (Braga and Weisburd, 2010). One potential risky behaviour is in an unintentional increase in expenditure. A recent news report suggested that the contactless nature of the payment makes the money feel less real (Hooker, 2016). This could be an example of fast decision-making that would benefit from being interrupted by a slow, more reasoned, choice.

It’s so accessible, I spend more on the little things…”

– Will paying with contactless cards make us less healthy? 

A second form of sensory interaction is the use of mobile and wearable devices to track personal behaviours, dominated by a growing trend to wear fitness devices that monitor movement and basic health readings such as heart rate and skin temperature. The devices include a small screen providing immediate feedback, such as number of steps walked and calories burned in the last 24 hours. The devices also store the data online to provide longer-term trends.

Whilst the devices are predominantly used by individuals to track their personal health and well-being, including publishing results online to compare with others, there has been at least one instance where the data has been used by medical professionals in an emergency department to decide on the best course of treatment (Rudner et al, 2016). A patient was admitted having suffered a seizure and the treatment options depended on knowing if the seizure was an isolated incident or indicative of an ongoing condition, with potential fatal consequences if treated incorrectly. The emergency team were given permission to access the patient’s Fitbit™ data to examine his heart rate trend and were able to determine that it was indeed an isolated seizure. The patient made a full recovery.

The increasing reliance on digital devices to monitor and interact with our surroundings is altering judgement and decision-making, creating both positive and negative effects. When transactions become effortless, such as with contactless payment systems, they appear to encourage effortless decision-making when deliberation may be more appropriate given the financial consequences. The granular level of data capture from everyday interactions is enabling new forms of monitoring and analysis that can lead to more informed decisions in situations containing risk and uncertainty.

Cognitive Enhancements

The third behavioural development linked to the mainstream adoption of ‘smart’ devices is in the field of artificial intelligence, and the potential to delegate human decision-making to technology. Mobile and wearable devices are now incorporating forms of virtual personal assistance – software connecting multiple different applications and sensors that alter uses of such devices from being a reactive tool assisting decisions to becoming proactive in response to changing conditions.

Early versions of virtual personal assistance via mobile devices included the ability to be automatically notified about disruptions to transportation networks that may affect travel plans, linking live online travel information with location data in a diary app on the device to provide situation awareness. The next iteration included predictions such as recommending a revised departure time to accommodate travel disruptions. The logical future development is the delegation of decision-making to the device, such as granting permission for travel plans to be automatically rearranged if there is a disruption that can be overcome via an alternative route. Early examples are already in existence, utilising an online framework ‘If this… then that…’ (IFTTT). Devices, applications and and real-time data sources can be programmatically linked to automatically create an action if specific criteria are met.

Mobile and wearable devices are increasingly acting as an extension to both cognitive and physical abilities in spatial situations. Examples have included using bio-sensing methods to help blind and partially-sighted people navigate cities (Cities Unlocked, 2014). The provision of a new form of taxi service booked entirely via mobile device is disrupting traditional taxi companies within many countries. Car-sharing firms such as Uber have been a benefit to deaf drivers since the passenger interaction can happen entirely without verbal communication. A recent news article (BBC News, 2016) detailed how a mother used Siri – a voice-activated virtual personal assistant included with the Apple iPhone – to call an ambulance when her one-year-old daughter stopped breathing. The woman, who had dropped her phone whilst running to attend to her daughter, was able to shout at her handset and communicate with emergency services whilst attempting resuscitation. The child survived.

Whilst we consider it normal to overcome physical limitations using technology, whether it is to improve eyesight, artificially sustain damaged organs or replace missing limbs, cognitive enhancements are still considered a novelty and abnormal. Their benefits are embraced in certain scenarios, such as those outlined above. However in many situations the use of technology to enhance our intelligence meets disapproval. Mobile phones are banned from many educational settings for fear of students ‘cheating’. Whilst we now have evidence of doctors using our personal data to make better health decisions, patients themselves are not expected to challenge expert opinion.

As mobile and wearable technology continues to decrease in size and cost, embedded enhancements and increased delegation of decision-making are likely. At some point, it may not be possible to differentiate between organic intelligence and digitally-augmented decision-making.

Conclusion

Smart City plans and the embracing of digital technology in modernising urban environments are failing to consider the altered behaviour of citizens resulting from access to the same technology and using it to adapt their own individual decisions and actions within urban settings.

The rapid rise in using mobile devices and online networks to support physical shopping decisions demonstrates how quickly new behaviours can become normalised. But such technology is still considered and treated as a separate tool rather than as an extension to our cognitive abilities. We don’t judge people differently because they wear glasses to improve eyesight yet we consider using technology to improve intelligence as abnormal, acceptable in some scenarios but unacceptable in others. As the devices continue to decrease in size and visibility, it may become impossible to differentiate between natural, artificial and hybrid forms of intelligence.

This is a challenge spanning government, healthcare, education and industry. The rising adoption of mobile, wearable and embedded technology is creating digitally-augmented intelligence accessible for all. Informed citizens bring different expectations when engaging with urban services.

References

Averill Jason D et al. (2005) Federal Building and Fire Safety Investigation of the World Trade Center Disaster: Occupant Behaviour, Egress, and Emergency Communications. National Institute of Standards and Technology NCSTAR 1-7.

Braga Anthony A and Weisburd David L. (2010) Policing Problem Places: Crime Hot Spots and Effective Prevention. Oxford University Press.

Cities Unlocked. (2014) Realising the potential of people & places. Online report, as of 5th June 2016. http://www.citiesunlocked.org.uk/wp-content/uploads/2014/11/CUReport_WEB.pdf

Hooker Lucy. (2016) Will paying with contactless cards make us less healthy? BBC News Online, as of 8th June 2016. http://www.bbc.co.uk/news/business-36358030

IFTTT web site https://ifttt.com

Kahneman Daniel. (2012) Thinking, Fast and Slow. Penguin Books edition.

Kleinman Zoe. (2016) Apple’s Siri calls ambulance for baby. BBC News online, as of 8th June 2016. http://www.bbc.co.uk/news/technology-36471180

Smith Aaron. (2013) In-store Mobile Commerce During the 2012 Holiday Shopping Season. Pew Research Center online report, as of 5th June 2016. http://www.pewinternet.org/2013/01/31/in-store-mobile-commerce-during-the-2012-holiday-shopping-season/

Rudner Joshua et al. (2016) Interrogation of Patient Smartphone Activity Tracker to Assist Arrhythmia Management. Annals of Emergency Medicine. DOI: http://dx.doi.org/10.1016/j.annemergmed.2016.02.039

UK Card Association. (2016) Contactless Statistics. Web page as of 5th June 2016. http://www.theukcardsassociation.org.uk/contactless_contactless_statistics/

Images

iStockPhoto images were licensed for use in this article and presentation and not for reuse. The following images were kindly shared under Creative Commons license, sourced via Photopin:

• Contactless payment by Vodafone Median
• Alcatel watch with fitness reading by Maurizio Pesce
• Reality check by GollyGForce

… and finally

If you made it this far – thanks for reading!!! 🙂 A couple of the slides from the presentation were not detailed in the paper. They included a quick outline of some research projects at CASA (The Bartlett Centre for Advanced Spatial Analysis, University College London) where I am currently undertaking doctoral research in spatial and behavioural data science.

Slide 7 is a summary of a data analysis I completed using social media spatially located in the Queen Elizabeth Olympic Park, London. The dataset covered the 4th to 31st March 2016. During that period there were two large events in the park. The first was the World Track Cycling Championships held in the Velodrome on the 4th – 6th March, with the major finals being held on the 5th. The second was a national charity event, Sports Relief. Just using a very small dataset – geotagged Tweets and Foursquare venue check-ins – and simple statistics, it was possible to see different temporal and spatial patterns in the data for event and non-event days. For example, Foursquare appeared to mimic local behaviours whilst Twitter appears to be more sensitive to special events likely to attract visitors to the area…

Slide 11 is a collection of images involving the ‘internet of things’, using low-cost sensors in urban settings. My own projects including building a nano-satellite and creating an interactive virtual fountain display that responded to facial expressions via a webcam. CASA was also involved in a digital empathy project ‘4Candles‘ (fans of an old UK TV show The Two Ronnies will appreciate the title) where an urban and rural church have digitally-connected fonts and are able to share thoughts and prayers. The common lesson learned across all the different projects was that people interacted for far longer than expected with the technology. Possibly because the experience was direct between the individual and the device with no intervening authority or oversight. It was a positive indication for the acceptance of such technology as it becomes increasingly embedded in our urban lives, when deployed for fun or to assist.

If you are interested and want to find out more, please get in touch!

Related links

• Slideshare of presentation
• PDF of paper
• Building a nano-satellite

Featured image: Smart London exhibit held at the NLA in London, December 2014 (author’s own photo). A model of London highlighting the new transport links under construction