Belief Busting: The value of rethinking how we form beliefs
How we need to consider different ways to think about ‘beliefs’ if we are to find ways to tackle them effectively
‘Vibecession’ was the term recently coined for when people are in a growing economy and yet they are anxious that everything is about to fall apart. This term was used to explain the 2023 apparent anomaly in the way that ordinary Americans were convinced that the economy was terrible whilst official numbers suggested the opposite. For example, in April, as unemployment hit a 60-year low, a record-high share of the public expressed negative views of the economy. At the same time, as Americans said it was the worst time to buy stocks in almost 20 years the S&P 500 had surged 14 percent in six months leading up to April.
So how can we have such different beliefs between economists and consumers? For something so central to our world, the study of beliefs themselves tend to be limited, focusing on the content of beliefs, (the specific ideas or propositions that people hold to be true) rather than on the form of beliefs, that is the structure, nature, or way in which these beliefs are held and integrated into our lives.
We typically assume that beliefs are the result of us acquiring and assessing knowledge, meaning we can arrive at ‘justified beliefs’. Deficits in our ability to do this are often cited by mis / disinformation researchers as the reason for holding seemingly outlandish beliefs. This has meant that a great deal of focus has been on improving knowledge acquisition strategies, as this is assumed to be the precursor of belief formation.
But is this really the full story? Can there be other ways in which beliefs are formed? Indeed, the notion that knowledge is the cornerstone of belief is itself a relatively modern phenomenon – there are other explanations we should consider.
Added to this, some commentators suggest we live at a time of history where there is a huge shift in beliefs - meaning we had better have a good understanding how people are navigating and forming these new beliefs.
Are we living in a period of radical belief change?
There are periods in history where there is an intense re-negotiation of beliefs: for example, the late 18th and early 19th centuries were characterized by a surge in religious and millenarian beliefs (anticipating an apocalyptic end, followed by a divinely created new order). Historian J.F.C. Harrison sets out how this period of ‘belief transition’ saw the birth of a range of movements such as Methodism and The Shakers. These groups shared a common belief in an imminent, transformative event—be it the Second Coming, a new age of enlightenment, or a societal reformation.
So why did this belief shift happen? Harrison cites a number of factors that came together at that time: industrialization disrupted lives, revolutions like those in America and France questioned established power, the Enlightenment challenged traditional beliefs, and an emerging capitalist society deepened social divides. Perhaps no wonder that millenarian movements arose, as they offered a set of beliefs to navigate changing societal norms and proposed more equitable, divinely inspired alternatives.
One question for our purposes then is whether we are in the midst of a similar landscape of belief revolution? Certainly some people such as political theorist Paolo Gerbaudo speculate that we are witnessing a moment of global transition of ideas, aligning with historical cycles in ideologies that take place every fifty years or so. And a cocktail of COVID, climate change, technology disruption is leading commentators to suggest we are at a point of ‘revolution’ in many of our beliefs.
If this is the case, then perhaps we can see the period we are now living in as a time of excitement, energy and opportunity to rethink established ways of operating that have been dominant for decades. Whilst there are opportunities for societal innovation there are also risks to navigate, such as uncertainty, polarization, sectarianism, increased inequality and psychological stress.
But overall, we would argue that reflecting on whether we live in an epoch of significant belief change means we do need to understand the building blocks of the time we live in – how beliefs are formed. And it is that that we now turn.
The belief journey is not simple
A good starting point to understand and explain belief formation is through a Bayesian model, such as Karl Friston's Free Energy Principle. This sets out how living systems, including the brain, maintain order and function by minimizing ‘free energy’ — a term from thermodynamics, here representing prediction error or the gap between expected and observed outcomes.
In this context, beliefs are internal models the brain uses to anticipate sensory inputs. Adjusting these models to reduce free energy, the brain strives for accurate perceptions. In this case, Friston's theory suggests our beliefs shape our perceptions and actions, with the brain also updating its beliefs to align predictions with reality, thereby learning from discrepancies to improve future predictions.
While this suggests a clear pathway of existing beliefs (priors) being updated by new information, then human deficits mean that the updating process is not always as effective as it could be. We have what Cass Sunstein calls a ‘crippled epistemology’ which means that this process can ‘go wrong’.
Indeed, this is arguably the dominant model of how beliefs work, underpinning most ‘media literacy’ approaches such as Sander Van De Linden’s inoculation model. In the case of the ‘vibecession’, economists considered that consumers were being irrational (in that the evidence does not stack up with what people believe) – and the human shortcoming in evaluating the evidence was attributed to people being disproportionately influenced by a small number of items going up in price (such as the cost of electronic goods) whilst failing to see the bigger picture of the wider economic improvement. But is it that simple?
The trade-off between knowledge and beliefs
This apparent trade-off between knowledge and beliefs – with human frailties intervening between them – certainly aligns with popular mental models of the way our minds works. This is what psychologist Teppo Felin calls the ‘Bucket theory’ of perception, that there is a world that exists independently to how we think about it, just waiting to be uncovered. Of course, we can only ever seek to represent the world in our heads and our limitations means we can get this wrong.
But Felin suggests an alternative approach to understanding beliefs – the so-called ‘Searchlight theory’. This suggests we mentally reflect the world in a similar way to a road map represents a landscape. Points on the map represent towns and cities, lines represent roads and motorways. We cannot externally verify and validate our beliefs: we cannot know the world independently from our own experience. Importantly this suggests the way we make sense of the world is active, using guesses, theories, questions, and hypotheses, meaning we comprehend things by directing perception and attention.
This arguably turns on its head more conventional explanations of belief – although perhaps in reality it simply changes the emphasis. We can see that we all operate with a combination of ‘buckets’ and ‘searchlights’ – in some areas we are open to evidence to help us form new beliefs, in others, the belief has greater primacy and may well shape the way we subsequently gather evidence.
In the case of ‘vibecession’, the priors held by economists are that economic indicators reflect what is happening ‘on the ground’. But official figures may not in fact reflect reality for the majority of people. Growing economic inequality means that economic improvements are often disproportionately accrued by a small segment of the population. This means that for most individuals, the struggle with living standards is not merely a perception issue but a tangible everyday reality.
In this instance perhaps the ‘vibecession’ is an example of belief derived via a ‘searchlight model’: we are invested in economic data offering us a coherent understanding of our world and then fail to take into account the evidence that is clearly available to many people who are suffering in the ongoing cost of living crisis.
Perhaps we need a greater emphasis on a searchlight model of belief
As practitioners, it is useful to consider that some beliefs are better understood as ‘’bucket’ in their formation while others are more ‘searchlight’. Just doing this can seem quite a radical step as all too often we assume beliefs are formed in a ‘bucket’ manner.
And perhaps there is a case to be made that in a time of change as we are arguably living in today, we can perhaps better understand belief formation via the ‘searchlight’ model even though it can at times be ill-fitting (as it seems the case for our economists). For example, in a changing environment there may simply be very scant, if any, evidence as Lorraine Daston suggested when COVID broke, we were living in a period of radical novelty and uncertainty, thrown into a state of ‘ground-zero empiricism’.
And are many examples throughout history that being willing to change our beliefs in the light of apparent evidence is not always very sensible. The notion that Saddam Hussein had weapons of mass destruction was vigorously contested by protestors and only subsequently found to be based on very shaky evidence. And before Edward Snowden made his disclosures, speculation about the existence of extensive, unauthorized data collection programs were often dismissed as conspiracy theories or paranoia about government overreach. So sometimes holding onto a belief which is counter to the evidence being presented is perfectly legitimate.
This suggests a departure from assuming beliefs are always ‘bucket’ and that some will be ‘searchlight’. The implications are significant for the way we think about misinformation and conspiracy theories: the reasons we hold ‘minority beliefs’ cannot always be assumed to be a function of poor quality reasoning on the part of the believer.
And in fact this is how science works, as Imre Lakatos suggests - a certain level of faith or provisional acceptance of a theory is typically necessary, even when not all evidence supports it initially. This approach allows for the development and refinement of knowledge over time, as theories are tested and improved.
Essentially, it's about being open to the potential of a belief to become substantiated through further inquiry and evidence, rather than requiring all supporting knowledge upfront. This is a philosophical version of the famous Christian quote “Blessed are they that have not yet seen and yet have believed”.
But can this tip over into something unhelpful?
Of course, just as an over-emphasis on the way beliefs are formed by our ability to properly process knowledge (bucket theory) can fail to properly describe the landscape of different beliefs, so an over-emphasis on our prior understanding of the world (searchlight) can lead us astray. The notion we may put aside knowledge that does not align with our preconceived beliefs is often referred to as motivated reasoning, when we interpret the evidence from the world in a way that protects the belief we hold (rather than adjusting our beliefs to take into account this new information.
For example Dan Kahan’s identity-protective cognition framework makes a case for the way in which the cultural values we hold define our social identities – which in turn then shape our beliefs about disputed matters of fact (e.g., whether humans are responsible for climate change; whether the death penalty prevents murder).
And Jonathan Haidt explores the concept of sacred beliefs in his work on moral psychology. Sacred beliefs, according to Haidt, are those that are considered inviolable and non-negotiable by an individual or within a culture. They are so central to a person's identity or a group's values that questioning or challenging them is often met with strong resistance or offense. On this basis, it means beliefs can be highly resistant to new information.
But holding on too tightly to beliefs in the face of overwhelming evidence to the contrary (as perhaps is illustrated in the notion of ‘vibecession’) is as equally problematic as being overly willing to change beliefs in the light of new evidence. Both ends of the spectrum are unhelpful.
A more sympathetic take on when we hold beliefs despite evidence to the contrary
But having said this, much of the problematising of belief tends to be focused on cases where they do not seem to bend to the available evidence. And whilst this can of course be a problem, it does seem that the time is right to have a more sympathetic narrative of the beliefs that we hold in the face of challenging evidence. In medieval times, beliefs were actually seen as distinct from knowledge, with figures like Ignatius of Loyola suggesting that beliefs are about trust over empirical evidence, emphasizing the fallibility of human experience and the reliability of church authority.
While this perspective on deference to institutional authority typically is out of step with contemporary views, much of our belief system implicitly trusts external sources of authority. For example, as cultural theorist Mark Fisher said, "It's easier to imagine the end of the world than the end of capitalism”.
Perhaps the wider point is that who determines what are legitimate beliefs and what is reasonable evidence will always be something that is the subject of debate. But for those of us involved in engaging with people to help effect change, it is surely important to understand these different mechanisms if we are able to support ways to negotiate positive outcomes.
But should we leave this to the experts?
The ‘vibecession’ discussion could be interpreted as reflecting the notion that experts know best, ignoring the way in which the majority of people are effectively navigating the evidence ‘on the ground’ and refuting the economists’ evidence as lacking nuance. This illustrates the importance of navigating beliefs with the wider population who will have legitimate perspectives about the evidence as well as their prior beliefs.
To illustrate, we can return to Harrison's observation about the attractiveness of millenarianism in the late 18th and early 19th centuries. He suggests people were seeking to be part of a conversation about their changing world, with all the difficulties of varying beliefs and contested evidence. Indeed, the millenarianism movements typically encouraged people to read, interpret, and find meaning in religious scriptures themselves, often in ways that diverged from traditional church teachings. Harrison suggested this democratized knowledge and empowered individuals to seek their own understanding of divine prophecy and in doing so the followers could see themselves as part of a divine narrative unfolding in real-time.
An interesting modern-day counterpart to this is arguably Taylor Swift with her millions of fans worldwide. Clearly Swift is very different from millenarianism movements but there are surely some parallels in the way people are seeking to engage and unpick her lyrics, which often conceal hidden meanings, personal anecdotes, and subtle references.
Through her storytelling, Swift engages listeners, connecting on emotional and intellectual levels. ‘Swifties’, as her fans are known, work together to decode her lyrics which are often on social issues such as feminism, mental health, and social justice, prompting discussions and creating new understanding and narratives about the world.
We can see the world we inhabit as being in a period of immense change meaning that many of us with no particular qualifications or evidence at hand, are understandably keen to engage in a conversation about our beliefs about its direction and our place within it.
In conclusion
All too often in applied behavioural science it is tempting to assume there is a single model for the way we make decisions – or if we do not align with that model then we side with the opposing one. Perhaps this is a reflection of academic work, in a somewhat combative way, picking apart the arguments of an alternative position and in doing so making progress.
But what this can at times fail to do is to see that both positions have merit and depending on the context, either are helpful. Practitioners need a multi-dimensionality to their work, to be able to see the way that quite different models can be true at the same time; our challenge is to know when one applies more and the other is less relevant.
The risk of failing to take this sort of approach that is that the solutions (interventions) we adopt are then implicitly assumed to be appropriate for all situations. So, for example, interventions such as the inoculation game may well be valuable for those instances where the beliefs are shaped around the ‘bucket’ style of belief (priors are not strongly held and open to new information) but may not be as effective for the ‘searchlight’ style of belief (strongly held priors and less open to new evidence). The challenge and opportunity for practitioners is therefore, first and foremost, to know what class of question is to be tackled.
There is a case to be made that we are living in an era of huge change, and with this we are necessarily less reliant on ‘evidence’ to form beliefs – or at the very least the evidence is much more contested than in the past. This places a greater need for people to be part of the decision making, finding ways to engage to reach meaningful conclusions.