Key concepts
Horizon scanning involves crowd-sourcing information and drawing on communities of practice to sort, verify and analyse that information to look for early indications of poorly recognised threats, opportunities, and research questions.
‘Exploratory’ horizon scanning identifies novel issues by searching for the first signals of change.
‘Issue-centred’ scanning monitors issues that have already been identified by searching for additional signals to confirm, or provide further details about, their emergence.
Horizon scanning is a variety of foresight. Foresight is not about predicting the future but exploring the range of possible futures that may emerge. This makes it different to forecasting, which is the process of making predictions about how the future might turn out, based on past and present data and trends.
Purpose of the tool
The purpose of horizon scanning tools is to make use of distributed information by providing a structured process that both helps improve people’s judgements and aggregates those judgements into a single result.
Judgements may be qualitative, in terms of describing emerging trends, issues, or threats, and quantitative, in terms of ascribing probabilities or other numerical characteristics to these. However, CSER tends to restrict quanitative judgements to estimations of the timeframe in which something might be expected to emerge and the relative priority of conducting further research into it.
Horizon scanning is thus a form of structured expert elicitation, that aims to improve upon individual expert opinion. It has long been used for thinking about future possibilities or for responding to scientific problems that involve a very high degree of uncertainty and/or demand a high degree of interdisciplinary cooperation to address. It can be used to study emerging trends, issues, and technologies and to identify opportunities, threats, and research questions.
History and background
Luke Kemp talks about different approaches to foresight and forecasting for Existential Risk, including Horizon Scanning and the Delphi technique
The most famous approach to horizon scanning is the ‘Delphi technique’, which was developed by RAND in the 1950s. This technique is often known as Estimate-Discuss-Estimate (EDE) and uses a panel of experts who are asked to respond to a series of questions across two or more rounds. After each round, a facilitator provides an anonymised summary of the results, along with the reasons each expert provided for their answers. Extreme outliers must substantiate their position and experts can then revise their judgements given the broader knowledge-based gained through considering the responses of others; leading, hopefully, to experts converging towards more accurate and informed judgement.
Another approach that has recently been developed, in part by Existential Risk researchers, is the IDEA (Investigate, Discuss, Estimate and Aggregate) protocol. This drops the focus on seeking consensus and allows participants to discuss differences of opinion and defend probability estimates directly in participatory workshops, rather than responding to anonymised statements of reasons. The final independent estimates are then given as anonymous submissions to the facilitator who aggregates them, either on their own or with further feed-in from the group.
CSER’s involvement
Luke Kemp talks about how CSER has made use of Horizon Scanning in our work and some of the things we have learned
CSER researchers have been involved in several horizon scanning techniques using variations on the IDEA protocol and related techniques, both as facilitators and participants. These have included:
- Emerging trends and technologies: a horizon scan for global public health
- WHO Horizon Scan on Dual-Use Research of Concern
- 80 Questions for UK Biological Security
- Bioengineering horizon scan 2020
- A transatlantic perspective on 20 emerging issues in biological engineering
- A 2017 Horizon Scan of Emerging Issues for Global Conservation and Biological Diversity
- A Horizon Scan of Global Conservation Issues for 2016
In addition, horizon scanning techniques have influenced CSER’s wider work on using participatory methods to study future threats, including our reports on Epistemic Security and the Malicious Use of Artificial Intelligence.
Where to get started
CSER researchers have produced a number of guides to assist people wishing to make use of horizon scanning to support their work
For an overview of different approaches to horizon scanning see this chapter on scanning horizons in research, policy and practice
For an introductory guide to the IDEA protocol see this paper on a practical guide to structured expert elicitation using the IDEA protocol
For suggestions on how horizon scanning can support better policy-making in relation to risk assessment and risk planning see this evidence submission on foresight for unknown, long-term and emerging risks, Approaches and Recommendations.
Finally, see the methods sections for each of our horizon scans for more detail on the specific techniques used in each of them and how these have been applied, revised, and updated over time.
CSER's 3 top tips for using Horizon Scanning
- The success of a horizon scanning exercise depends upon what is put into it. This includes the questions that are asked, the group that is recruited to participate, the degree of participation and engagement of that group, and the quality of analysis conducted on the results. If the wrong question is asked to the wrong people who engage in the wrong ways and the results get uncritically written up as truths, then the horizon scanning process will likely have done more harm than good.
- Having a genuine diversity of perspectives is essential for the success of horizon scanning. This should include diversity in discipline, profession, and practice, geography, culture, and lived experience, relationship to the policy process, prior beliefs, knowledge base, and expertise. The more diversity is present among the participants, the more information and understanding will be possessed by the group as a whole, the more this will be subjected to critical analysis and engagement from individual group members, and the more valuable any resulting convergence and agreement will be.
- Even a well-run exercise involving diverse participants can be subject to groupthink. Employing ‘devil's advocates’, who are specifically (secretly) requested to be critical of any emerging consensus or to make proposals that run counter to expected trends can help to tackle this and improve the novelty and robustness of results.