Given that technology is moving at such an extraordinary pace, for this holiday weekend we thought we’d publish a triptych of blog posts exploring the psychology of how we imagine and predict the future.
The first, published yesterday, sheds light on the psychological and social processes behind making predictions.
Today we’re going to zoom in on the mistakes that creep into this process, and illustrate how these fallacies distort our vision of the future.
Tomorrow, we’ll put these insights to good use and have a closer look at a vision of 2001, as it was imagined in 1900… We’ll explore what the author got right, why most of his recommendations missed the mark, and how we can use his analysis to assess current imaginations of the future in a hundred years’ time.
If you want to read up on the basics of mind-travel to the future, you can check out yesterday’s blogpost here.
Ready? Then let’s begin…
Imagining the future is tricky
Generally, humans are pretty good at imagining what might happen in the near future and how we might feel about it – but it is with hindsight that the inevitable mistakes in our predictions become evident.
Both our predictions of future emotions and the scenarios on which they are based are prone to errors, mostly because any simulation we might make (and our emotional response to it) is influenced by our internal model of the future, and contextual factors in the present.
Only if the future actually turns out the way you imagined it, and your context (e.g. life circumstances, personality, etc.) haven’t changed much, will your predictions and imagined emotional response be accurate, as outlined in the table below.
Unfortunately, this kind of accuracy is pretty rare (Wilson & Gilbert, 2005).
|Simulation (X)||You react emotionally to a simulation of the future (1)||You simulate the future in context A, which impacts your reaction.|
|Experience (Y)||You react emotionally (and maybe differently) to an experience in the future. (2)||You experience the future in context B, which also impacts your reaction (maybe differently).|
|X = Y if||1 = 2 and||A = B|
We feel differently about simulations vs experiences
Simulations of the future, insofar as they are assembled from our memory of our past experiences (see yesterday’s blog post), tend to be unrepresentative of the events that actually take place.
Rather than an accurate, objective recording of reality, when we remember an experience our memory tends to be strongly influenced by the peak highs, lows, and the end of what we went through. Known in behavioural economics as the ‘peak end rule’, this can explain why our memories may vary so wildly to those around us, and why we only recall the most emotionally salient moments (Kahneman et al. 1993).
As if that weren’t enough, we are also more likely to memorise recent and unusual events, further skewing the memories on which we then build our simulations and predictions (Gilbert & Wilson, 2007).
Our simulations of the future also tend to be abbreviated, as we don’t take into account each moment of the experience. Instead, we focus more on short-term emotional reactions (rather than long-term), which means that we can’t account for the process of ‘hedonic adaptation’ (the tendency to make sense of and adapt to our circumstances, thus causing our emotional reactions to dissipate over time) (Gilbert & Wilson, 2007).
All of these elements can contribute towards inaccurate predictions and ‘impact bias’, our propensity towards ‘overestimat[ing] the intensity and duration of [our] emotional reactions to future events’ (Wilson & Gilbert, 2005, p.131).
We ignore the context in which we’ll experience an event
Predictions of how we will feel tend to focus on essential aspects of the experience while leaving out auxiliary details, however it is often these non-essential details that can strongly affect our perception of an experience.
We might imagine, for instance, that moving to a hip neighbourhood would make us happy.
Once we have moved, we may well come to realise that we miss our old neighbours, can’t stand the street noise and find the lack of parking space a real hassle. Simulations lack the ‘richness and reality of genuine perceptions’, which can lead to big differences between our emotional reactions to simulations versus reality (p. 1354, Gilbert & Wilson, 2007).
Still with me? Good, because we’re not yet done… There are yet more things that cloud our vision.
The problem with heuristics
When faced with complex mental operations such as assessing the probability of a future event, we often rely on heuristics (mental shortcuts or rules of thumb) to replace these complicated processes with a simpler one. While these heuristics tend to work decently most of the time, they are nonetheless prone to predictable systematic errors (Tversky & Kahneman, 1974).
Given that our imagination of the future is based on our memory of the past, our ability to imagine disruptive change, non-continuous innovation and unlikely events can can be limited at best.
Our imaginations are restricted by that which we have experienced or imagined before (as explored in yesterday’s post) and transcending these boundaries of imagination is both rare and difficult (Makridakis & Taleb, 2009, Zittoun & Gillespie, 2016).
The certainty with which we hold our beliefs about the future matters as well. If we are very confident of our own predictions and emotional reactions, we are less likely to consider and plan for alternative futures, which (if our confidence happens to be unfounded) can obviously have deleterious effects (Makridakis & Taleb, 2009).
Lastly (phew!), our individual judgments about the likelihood of future events tend to be unduly optimistic, especially when we’re thinking about our own futures. When predicting the probability of, say, successfully attending the gym, we tend to compare ourselves unrealistically to inactive others, inflating our confidence that we’ll be able to make decisions that will positively impact our future (Weinstein, 1980).
How can you tell good predictions from bad ones?
All this is may be well and good, but it’s only useful if we can apply it.
If you want to accurately evaluate the quality of a prediction, you first have to identify the goal you wish to achieve by making that prediction. Because imaginations of the future can have a wide range of benefits, various criteria might be relevant.
For instance, imagining the future might be a way for us to escape from reality and seek entertainment, or it might enable us to gain a new perspective on the present and oneself. On a larger scale, it could serve to help us assess our current situation, imagine alternatives to the status quo, and potentially motivate us towards collective action to change the future (Zittoun & Gillespie, 2016).
Each of these goals lead us to different criteria with which to evaluate the future, and below are just some that you may wish to consider:
- Are you using this simulation to entertain yourself?
- Does it help you see yourself in a new light?
- Is the prediction of your future emotional reactions prone to typical prediction errors?
- Does your prediction of a future event focus on both accuracy and certainty?
- Does it allow you to consider the full range of impact it will have, such as the social and psychological changes it will lead to or the political reactions that it might entice?
- Does it adequately take into account varying positions within an affected group that might lead to a variety of perspectives and emotional reactions?
- Does it go beyond what has been imagined before?
- Does it connect to the present in a way that bridges what is and what could be?
- Does it take into account potentially unintended consequences?
- Does it motivate you towards action?
- Does it give you a feeling of agency in shaping the future?
Dive deeper and come back tomorrow for more
In tomorrow’s blog post, we’ll use these criteria to evaluate predictions of the year 2001 made one hundred years earlier, and attempt to evaluate predictions people made in 2015 about what is to come in 2115…
If you want to dive deeper into the problems of forecasting the future, you can check out these resources:
A Ted talk by Dan Gilbert (whom we have cited above) on what makes us happy and what we falsely believe to make us happy.
Daniel Kahneman’s Ted talk on the difference between experience and memory (and therefore also predictions).
A video interview with Philip Tetlock on ‘Why an open mind is key to making better predictions’, a very short interview on his work and an article citing Tetlock on forecasting the future by Lisa Landskron.
Gilbert, D. T., & Wilson, T. D. (2007). Prospection: Experiencing the future. Science, 317(5843), 1351-1354.
Kahneman, D., Fredrickson, B. L., Schreiber, C. A., & Redelmeier, D. A. (1993). When more pain is preferred to less: Adding a better end. Psychological science, 4(6), 401-405.
Makridakis, S., & Taleb, N. (2009). Living in a world of low levels of predictability. International Journal of Forecasting, 25(4), 840-844.
Tversky, A., & Kahneman, D. (1974). Judgment under uncertainty: Heuristics and biases. Science, 185(4157), 1124-1131.
Weinstein, N. D. (1980). Unrealistic optimism about future life events. Journal of personality and social psychology, 39(5), 806-820
Wilson, T. D., & Gilbert, D. T. (2005). Affective forecasting knowing what to want. Current Directions in Psychological Science, 14(3), 131-134.
Zittoun, T. & Gillespie, A. (2016). Imagination in human and cultural development. London: Rout- ledge. (Chapter 7).