AI and the future of behavioural science | LSE Event

We recently held an event at LSE on AI and the Future of Behavioural SCience. IT was a very stimulating session. The video and description are available below.

 

Artificial Intelligence (AI) is transforming various aspects of behavioural science. For example, AI-driven models are being used to predict human behaviour and decision-making, and to design personalized behavioural interventions. AI can also be used to generate artificial research participants on whom behavioural interventions can be tested instead of on humans. AI is creating many new opportunities and challenges in behavioural science, disrupting the discipline to the degree that researchers, practitioners, and any behavioural science enthusiasts are trying to keep up with the new developments and understand how to best navigate the rapidly changing landscape. In this public event, speakers who are associated with pioneering work on AI in relation to behavioural science, as part of their own research or organisational initiatives, will discuss their views on how AI will change and is already changing behavioural science. This will involve touching upon topics such as the implications of AI for behavioural scientists in academia, public, and private sectors, new skills that will be required by behavioural scientists of the future, and impact on behavioural science education. 

Speakers: Alexandra Chesterfield Elisabeth Costa Professor Oliver Hauser Dr Dario Krpan Professor Susan Michie Professor Robert West Chair:

Professor Liam Delaney

Biden-Harris Sludge Initiative

A fascinating new memo from the White House.  Various literature and practice areas have evolved examining the role of administrative design in decision making and in the potential for manipulation of consumers. The language of the memo is combative "Americans are tired of being taken for suckers". But it is clearly a very interesting area in terms of policy, regulation and consumer outcomes. The combination of ideas on reducing misuse of administrative frictions, promoting a good environment for user reviews, and ethical deployment of chatbots is quite an exciting agenda from a research and policy perspective. 

New actions will take on corporate tricks and scams like excessive paperwork, long wait times, and more that pad the profits of big business at the expense of everyday Americans’ time and money.

Today, President Biden and Vice President Harris are launching “Time Is Money,” a new governmentwide effort to crack down on all the ways that corporations—through excessive paperwork, hold times, and general aggravation—add unnecessary headaches and hassles to people’s days and degrade their quality of life.

Americans are tired of being played for suckers, and President Biden and Vice President Harris are committed to addressing the pain points they face in their everyday lives. The Administration is already cracking down on junk fees—those hidden costs and surcharges in everything from travel to banking services—that hit people in their pocketbooks. Now the Biden-Harris Administration is taking on the corporate practice of giving people the run around, wasting their precious time and money.

Americans know these practices well: it’s being forced to wait on hold just to get the refund we’re owed; the hoops and hurdles to cancel a gym membership or subscription; the unnecessary complications of dealing with health insurance companies; the requirements to do in-person or by mail what could easily be done with a couple of clicks online; and confusing, lengthy, or manipulative forms that take unnecessary time and effort.

These hassles don’t just happen by accident. Companies often deliberately design their business processes to be time-consuming or otherwise burdensome for consumers, in order to deter them from getting a rebate or refund they are due or canceling a subscription or membership they no longer want—all with the goal of maximizing profits.

In addition to robbing hardworking families of their valuable time and adding frustration to our daily lives, these hassles cost us money. When, after endless hours on hold or piles of incomprehensible paperwork, we give up pursuing a service, rebate or refund we’re due, we take a hit to our pocketbooks, and companies profit

Today and in the coming months, the Biden-Harris Administration will take wide-ranging action to crack down on these unfair practices and save Americans time and money. Key actions include:

Making it easier to cancel subscriptions and memberships. Businesses often trick consumers into paying for subscriptions—on everything from gym memberships to newspapers to cosmetics—that they no longer want or didn’t sign up for in the first place. Consumers shouldn’t have to navigate a maze just to cancel unwanted subscriptions and recurring payments. The Federal Trade Commission (FTC) has proposed a rule that, if finalized as proposed, would require companies to make it as easy to cancel a subscription or service as it was to sign up for one. The agency is currently reviewing public comments about its proposal. And today, the Federal Communications Commission (FCC) is initiating an inquiry into whether to extend similar requirements to companies in the communications industry.

Ending airline runarounds by requiring automatic cash refunds. The Department of Transportation’s (DOT) new automatic refunds rule requires airlines to pay you back the airfare when your flight is canceled or significantly changed for any reason, and you are not offered, or choose not to accept, alternatives such as rebooking. This rule prevents airlines from switching up their policies to make it hard to get your money back when they don’t deliver and requires them to tell you when you’re owed a refund. DOT’s rule also puts an end to airline runarounds by requiring refunds to be automatic, prompt, in the original form of payment, and for the full amount paid. No more jumping through hoops or getting stuck with expiring flight credits.

Allowing you to submit health claims online. Health coverage can be full of headaches and hassles, as many plans and insurance companies make it unnecessarily difficult to access information or send in claims. For example, many of the largest plans still require some customers to print out and either scan or physically mail health claims forms, and people seeking help can encounter inaccurate or confusing websites, extended wait times, or narrow call center hours that force them to step away from work to talk to an agent. Today, Department of Health and Human Services (HHS) Secretary Becerra and Department of Labor (DOL) Acting Secretary Su are calling on health insurance companies and group health plans to take concrete actions to save people time and money when interacting with their health coverage, and in the coming months will identify additional opportunities to improve consumers’ interactions with the health care system. In addition, the Office of Personnel Management plans to require Federal Employees Health Benefits and Postal Service Health Benefits plans, covering eight million Americans, to make it easier to submit out of network claims online, provide clear information about what health plan providers are in-network, and make it easier to find information on how to appeal claim denials.

Cracking down on customer service “doom loops.” Too often customers seeking assistance from a real person are instead sent through a maze of menu options and automated recordings, wasting their time and failing to get the support they need. In a recent survey, respondents said that being forced to listen to long messages before being permitted to speak to a live representative was their top customer service complaint. To tackle these “doom loops,” the Consumer Financial Protection Bureau (CFPB) will initiate a rulemaking process that would require companies under its jurisdiction to let customers talk to a human by pressing a single button. The FCC will launch an inquiry into considering similar requirements for phone, broadband, and cable companies.  HHS and DOL will similarly call on health plan providers to make it easier to talk to a customer service agent.

Ensuring accountability for companies that provide bad service. People shopping for products or services should be able to rely on customer reviews to assess which companies will provide streamlined service and not waste their time. The FTC has proposed a rule that, if finalized as proposed, would stop marketers from using illicit review and endorsement practices such as using fake reviews, suppressing honest negative reviews, and paying for positive reviews, which deceive consumers looking for real feedback on a product or service and undercut honest businesses.

Taking on the limitations and shortcomings of customer service chatbots. While chatbots can be useful for answering basic questions, they often have limited ability to solve more complex problems and disputes. Instead, chatbots frequently provide inaccurate information and give the run-around to customers seeking a real person. The CFPB is planning to issue rules or guidance to crack down on ineffective and time-wasting chatbots used by banks and other financial institutions in lieu of customer service. The CFPB will identify when the use of automated chatbots or automated artificial intelligence voice recordings is unlawful, including in situations in which customers believe they are speaking with a human being.

Helping streamline parent communication with schools.  Between communicating with teachers, viewing school policies, completing forms and permission slips, and more, school processes, platforms, and paperwork can sometimes be a hassle for families that already have a lot on their plates. The Department of Education will issue new guidance to schools on how they can help make these processes less time-consuming for parents to handle, and to build effective family engagement through two-way communications. This will include new resources for schools to address time-wasting technology and offer more streamlined processes for engaging and communicating with parents.

What else should we take on? The White House is calling for Americans to share their ideas for how federal action can give them their time back. Interested parties can submit their ideas and comments at this portal, and may consider the following principles:

Companies should make it as easy to do things that you want to do as it is to do things they want to do.

It should be as easy cancel a subscription or membership as it is to enroll.

It should be as easy to obtain rebates and refunds as it was to purchase, with no needlessly cumbersome paperwork.

Refunds and rebates should be paid as quickly as companies take funds from your credit card or bank account.

Americans should be able receive customer service on their terms and their own time without significant hassle or hardship.

If you want to talk to a human, you should be able to talk to a human at convenient times and without interminable waits.

If you prefer to interact electronically – such as by text, email, or online portal – there should be simple and easily identified ways to do so securely.

Technology – such as chatbots – should be used to enhance customer service with speedy response times, not used to shirk on basic responsibilities, such as receiving a refund.

Americans should not be subject to confusing, manipulative, or deceptive practices online.

If you want to understand what you must do to obtain a good or service, the requirements should be clear and transparent.

You should not be subject to hidden fees or to requirements that are obscured through confusing language and small print.

Time Is Money builds on landmark efforts by the Biden-Harris Administration to improve customer service for people accessing government programs and services. In December 2021 the President signed an Executive Order, Transforming Federal Customer Experience and Service Delivery to Rebuild Trust in Government, directing federal agencies to streamline services and simplify customer experiences.

Already, agencies are making progress: the State Department launched a public beta to renew  your passport online; all 50 states have been invited to offer the Internal Revenue Service’s Direct File tool, an easy, secure, and—most importantly—free way for Americans to file their federal taxes; HHS has taken steps to allow more than 5 million Americans to automatically renew their health coverage without filling out paperwork, saving over 2 million hours in estimated processing time; and the Department of Homeland Security (DHS) announced that it has reduced the amount of time the public spends accessing DHS services per year by 21 million hours in fiscal year 2023, and is targeting reduction of 10 million more hours per year in fiscal year 2024. For more examples of progress and to learn more information about how agencies across the federal government are improving customer experience and reducing burden, visit performance.gov/cx and the Burden Reduction Initiative website.

Irish Department of Health: 10 practical tips for using behavioural science in health policy

The Irish Department of Health have recently released this document on applications of behavioural science in healthcare. I have worked with them on a number of projects. Behavioural science is a capacity embedded into healthcare policy and was a strong component of covid response in Ireland. There is a very strong link between academia, policy and a number of NGOs in this area in Ireland and it is evolving into a mature and interesting capacity for healthcare in Ireland. 

The 10 practical areas outlined in the paper are:

1. Increase hospital productivity by changing default options, such as changing the default to day case rather than inpatient for certain hospital procedures.

2. Increase hospital productivity by reducing readmissions and average length of stay by using improved discharge templates and additional supports.

3. Increase productivity by reducing did not attends for hospital appointments by using behaviorally informed content (shown to increase patient engagement) in correspondence such as appointment offer letters and text reminders.

4. Increase productivity by growing uptake of online services and improving administrative processes by reducing administrative burden through use of sludge audits.

5. Increase productivity by using behavioural insights to support use of key good practices in elective care such as using low-complexity pathways for low-risk patients, using enhanced recovery practices, and increasing throughput in theatres by measuring, communicating and managing the number of cases per theatre session.

6. Reduce influenza’s influence on demand and supply by increasing flu vaccination uptake with behaviourally informed letters / correspondence for the community, and by using multifaceted approaches for health workers.

7. Reduce unnecessary demand for Emergency Departments and outpatient consultations by using audit and feedback with checklists for GPs and by providing written guidance for patients and parents.

8. Reduce unnecessary antibiotics prescribing in general practice through greater use of behavioural interventions shown to reduce unnecessary prescribing, such as education, communication training, point of care testing, other decision support tools, and delayed prescribing. This will result in both financial and health gains, as reducing unnecessary antibiotic prescribing is central to reducing antimicrobial resistance.

9. Reduce future demands and costs by growing cancer screening through better correspondence by incorporating insights from behavioural science into screening offer letters and processes.

10. Reduce future healthcare demands and costs through better food choice architecture by using calorie posting on online menus for fast-food chains, front of pack nutritional labelling for food products in supermarkets, and recommended portion sizes supported by using reduced size tableware in canteens in public buildings.

Advances in Well-Being and Policy June 17th Geary Institute

I am organising a roundtable workshop on June 17th. It will take place between 11am and 4pm in the Geary Institute boardroom with kind venue support from the UCD Geary Institute. The workshop will hear a set of short talks on ongoing developments in the study of subjective well-being and public policy, including the development of national well-being frameworks, advances in subjective well-being measurement, current state of knowledge of connections between well-being, climate, health and policy, and other topics.  Presentation of policy papers, work in progress and/or potential project ideas are encouraged alongside regular research paper presentations. Bloomsday the day to celebrate James Joyce's famous book Ulysses takes place the day before so there will be at least some space devoted to day reconstruction methods! I will also organise the session to allow for a focused discussion on publication, research funding, and policy application. 

Provisional Programme (Suggestions by email welcome) 

1030 - 11am: Coffee and Welcome

11.00am to 11-30am: Leonhard Lades (UCD and Stirling), "Pro-Environmental Behaviour in Everyday Life".

11.30am to 12pm: Sharon Raj (LSE), "The Interrelationship Between Employment Status, Time Structure, Time Perceptions & Wellbeing". 

12pm to 1230pm: David Staunton (EY) "Workplace wellbeing - a study of professional services in a hybrid working environment".

1230pm to 1pm: Michael Daly (NUIM), "Meaning in Everyday Life: A Population-Level Analysis of Activities, Temporal Patterns, and Demographic Trends". 

115pm to 2pm: Working Lunch on Issues and Advances in Subjective Well-being and Policy 

2pm to 230pm: Cheng Ma (UCD) "Studying the experience of travel on the Dingle Peninsula using Day Reconstruction Methods". 

230pm to 3pm: Fiachra Kennedy (PER)  “Unravelling the Puzzle of Time: Well-being and Public Policy”.

3pm to 330pm Caoimhe Dempsey and Selina McCoy (ESRI). "The Department of Education Wellbeing Policy Statement: Objectives and Experiences".

330pm to 4pm: Chris Krekel (LSE), "Back to Edgeworth? Estimating the Value of Time Using Hedonic Experiences". 

Some resources and links: 

Arulsamy and Delaney "The impact of automatic enrolment on the mental health gap in pension participation: Evidence from the UK". Journal of Health Economics.

CSO Well-being Data Hub for Ireland.  

Daly and Macchia "Global Trends in Emotional Distress". PNAS. 

EPA Environment and Well-Being Conference 2024 

Farrell and Mahon "Well-being in the Irish secondary school: Reflections on a curricular approach".

LSE Department of Psychological and Behavioural Science Well-Being Research Page 

Martin, Delaney, and Doyle "Everyday Administrative Burdens and Inequality". Public Administration Review.

Irish Government Publication on a Well-Being Framework for Ireland.

Frijters and Krekel: A Handbook for Well-Being Policy Making. 

Krekel and McKerron: "Back to Edgeworth? Estimating the Value of Time Using Hedonic Experiences". IZA Working Paper. 

OECD Well-Being page 

Leo Lades presentation on pro-environmental behaviour and mood in everyday life 

Kyle Moore and David Maher "Adopting a Well-being Approach to Assessing Climate Action in The Transport Sector". Strategic Research and Analysis Division Department of Transport 

Irish Government Economic and Evaluation Service Well-Being publications 

The implications of behavioural science for effective climate policy (CAST)

The UK Climate Change Council council recently released this report that they commissioned from the the Centre for Climate Change and Social Transformations (CAST) at Bath. Details from their summary below. 

The eight key areas reviewed are:

• Diet change – reducing high-carbon foods in people’s diets (e.g. meat and dairy).
• Reducing end-user consumption – encouraging people to reuse and repair goods, recycle, and reduce food waste.
• Aviation demand – reducing how much people fly.
• Adaptation – adapting to a changing climate (e.g. extreme heat or flooding).
• Net Zero skills and careers – helping people develop Net Zero skills and enter Net Zero careers.
• Business leaders and the transition to sustainability – encouraging businesses to be more sustainable.
• Land use and farming – increasing tree-planting on farmland.
• Policy acceptability – increasing the public acceptability of climate policies.

Several areas were omitted where evidence is being collated as part of separate projects or has been collated elsewhere, such as surface transport, energy, and buildings.

In total, 336 academic papers and 62 non-academic reports (e.g. from government or charities) were reviewed. The review prioritised UK-based evidence but drew on international literature where UK evidence was lacking.

2. Key messages

In the eight areas reviewed, low-carbon and climate-resilient behaviours were primarily driven by:

• An individual’s knowledge about a behaviour, their values, and emotions.
• Social factors (e.g. what individuals perceive others do or expect of them, and which social groups individuals most identify with).
• Practical factors (e.g. how difficult or expensive it is to carry out certain behaviours).

Most research explored how effective it is to provide people with information (e.g. about the impacts of certain diets). These information-based interventions tend to be less effective at changing behaviour compared to measures that make green choices and behaviours the default or shift social expectations (e.g. by the government or businesses demonstrating leadership or introducing new policies).

The focus in the evidence base on informational approaches has limited our understanding of whether some interventions are more effective than others, how varying social groups might respond differently to interventions, and how large-scale behaviour change can be achieved.

Overall, the report notes a need for clearer communication of climate policies and more public involvement in policy design. This may be achieved with the introduction of a dedicated public engagement strategy. There is also a need for policies that make low-carbon behaviours easier, cheaper, and more attractive, alongside regulations and incentives to shift behaviour. Generally, interventions are more likely to be successful when introduced in ‘moments of change’, when habits are disrupted and behaviour is more malleable.

The key findings and implications for each of the eight key areas are summarised below.

• Reducing high-carbon foods in people’s diets (e.g. meat and dairy). Providing information about a food’s impact on one’s health, the environment, or animal welfare is not an effective way to change diets in isolation. Information-based interventions work best in combination with other approaches, such as making plant-based foods more available, convenient, attractive, and affordable. Policymakers may achieve this by making plant-based options more visible and the default in supermarkets and restaurants, alongside introducing financial incentives (e.g. reducing the price of plant-based foods).

• Encouraging people to reuse and repair goods, recycle, and reduce food waste. People are more likely to reuse and repair goods (rather than buying new ones) if they have strong environmental values or perceive others to reuse and repair. However, a lack of relevant skills and low availability of repair services are key barriers. There is some evidence that financial incentives might be effective at encouraging recycling (e.g. getting money back for recycling), but more studies are needed to understand if incentives might be effective at encouraging people to reuse and repair (e.g. policies to reduce VAT on repair services). There are minimal real-world trials studying how to encourage people to buy fewer material goods. Policymakers may opt for a combined approach – providing useful information (e.g. skills training for repairs) and providing financial incentives (e.g. making repairs cheaper).

• Reducing how much people fly. People are generally reluctant to reduce how much they fly, as flying is often linked to ideas about success, freedom, and identity. People’s concern about climate change does not necessarily mean they reduce how much they fly, but giving information about a flight’s emissions could lead consumers to choose more efficient flights. The public broadly supports taxes that make flying more expensive for frequent fliers (Frequent Flyer Levies), but such measures need to be accompanied by changes to legislation, and government leadership. It is important that policies aiming to reduce how much people fly are perceived as fair for all, and that the need to fly less is communicated consistently. Generally, research about the effectiveness of aviation policies is very limited.

• Adapting to a changing climate. Most people know very little about how they can prepare for changes such as extreme heat, droughts, or flooding, making it important for policymakers to clearly communicate what types of behaviours are most effective in preparing for a changing climate. More research is needed to investigate if regulations can be effective in creating more climate-resilient buildings (e.g. regulations for planners), and helping people make changes to their homes (e.g. interest-free loans for homeowners). There is also a need for more research comparing the effectiveness of interventions for different types of risks (e.g. flooding compared to heatwaves), and further evidence about how to encourage people to change their behaviour in situations like heatwaves or droughts. Policymakers should therefore focus on delivering clear advice tailored to the type of risk, alongside other measures like financial incentives for homeowners. UK-based evidence in this area was limited, meaning more global evidence was used.

• Helping people develop Net Zero skills and enter Net Zero careers. People know little about the type of skills that are needed to achieve Net Zero in the UK, making providing young people with information about Net Zero skills and careers an important first step in encouraging them to choose a green career. To achieve this, policymakers would need to give a clearer definition of what green skills and careers are alongside opportunities to develop these skills. Schools and businesses could play an important role in educating about Net Zero careers, but more evidence must test this.

• Encouraging businesses to be more sustainable. Businesses are more likely to adopt sustainable business practices when managers are aware of climate change and its impacts. Policymakers may therefore seek to educate managers about sustainable business models. Alongside this, businesses often shift to more sustainable practices when consumers demand more eco-friendly products, and their competitors adopt sustainable practices. Market-based policies could be as successful as regulations, but more UK-based evidence in larger companies is needed to determine this.

• Increasing tree-planting on farmland. Providing farmers with financial aid might encourage increased tree-planting. However, grants are often overly complex to apply for, limiting how effective they can be. Simplifying the grant application process could be a low-cost way of increasing grant uptake. The loss of control over land that comes with using grant money is also a concern for farmers. Policies that focus on providing examples of woodland creation by other farmers, and information campaigns and training about woodland creation are likely to be effective, especially when focusing on the profitability of creating new woodland, and engaging the wider farming community.

• Increasing the public acceptability of climate policies. The public generally favour policies that provide grants or subsidies (known as pull measures) over bans and regulations (push measures). However, there are factors that tend to increase support across a variety of policies. People are more supportive of policies that they perceive as effective and fair for all. It is also important that the policy decision-making process is viewed as transparent, and that policies are implemented by trusted leaders. Policies should be designed in line with these principles. Providing information about other benefits (e.g. for people’s health) and involving the public in designing policies can also increase support. Grouping policies in ‘bundles’ may also bolster support, but further research is needed to test this.

Behaviourally Informed Health Policy in Ireland

Below from Robert Murphy in the Irish Department of Health details the latest in a series of reports and publications on behaviourally-informed health policy in Ireland. Further details including links to papers are available on the following website. 

This paper reports the impact of a collaborative quality improvement project between the Department of Health, the National Treatment Purchase Fund (NTPF), and the Health Service Executive (HSE). It shows that using amended text (SMS) reminder content, informed by behavioural science, reduced DNA rates for outpatient hospital appointments by a substantial amount.

A “did not attend” (DNA) occurs when a patient unexpectedly does not attend an outpatient hospital appointment. This means that hospital staff expected the patient to attend but the patient did not attend and did not signal in advance that they would not attend. International research shows that DNAs for hospital outpatient appointments can lead to the inefficient use of staff time, worse care for patients, and increase waiting times for patients.

Sending a text (SMS) reminder to patients a few days in advance of appoints has been found to reduce DNA rates. International evidence suggests the effectiveness of these reminders at reducing DNA rates can be further enhanced by applying findings from behavioural science to enhance the content of SMS reminder messages. The quality improvement project sought to identify the best performing SMS reminder for Ireland.

We used a randomised control trial (RCT) to test the impact of four re-designed SMS appointment reminders (interventions) against an existing SMS reminder (control) on patient DNA rates in Naas General Hospital.

Previous international research shows that the most common reason reported by patients for not attending is that they forgot. All interventions aimed to help the patient remember better or to improve their recall.

 Intervention 1 (Recall I) included two design elements reported to increase engagement, namely personalisation and reciprocity. It also included the day of the appointment in words to help with recall.

All other interventions included these three design elements of Intervention 1 along with additional design features. Intervention 2 (Recall II) additionally included the name of the consultant and clinic name on the basis that this might help patients to remember their appointment.

Another commonly reported reason for not attending is that a patient felt the appointment appeared to have no benefit. Therefore, Intervention 3 additionally stated the importance of attendance for health (Recall II + Importance for health). Some previous international trials show an effect on DNAs of including the cost to hospitals of non-attendance, so Intervention 4 included text on the avoided loss if patients who cannot attend signal this in advance (Recall II + Avoided loss to patients and staff).

We found that the redesigned SMS of Intervention 2 (Recall II) is the best performing reminder. It reduced DNA rates by 12.7% compared to the control reminder. This small change to the content of the reminder resulted in one in eight non-attendees changing their behaviour. The intervention is highly cost effective. Applying the findings to 2022 national data suggests DNAs could be reduced by about 61,000 by using a better worded text reminder. 

Commenting on the publication, Minister for Health, Stephen Donnelly said:

“I welcome the findings of this impact evaluation. The paper identifies the best performing SMS reminder, which based on findings from behavioural science makes it easier to remember the appointment. Widespread adoption of the recommended SMS templates as part of national practice will substantially reduce non-attendance and improve hospital productivity.”

A special thank you to my co-authors Liam Delaney Deirdre A. Robertson Helen Ryan Alex Wood Alison Green Glenn Murphy and to my colleagues in the NTPF (especially Alison Green, Glenn Murphy, and Elva Powell) and the HSE (in particular Jenny O’Rourke) who made this project possible.