Initially published at the CIPD website here. 

Transforming Uncertainty from a Negative to a Positive: Assessing a Novel Intervention Designed to Increase Uncertainty Tolerance in Individuals and Teams

Dr Avri Bilovich1, Sam Conniff, Katherine Templar-Lewis, Stephen H. Dewitt2 

1Centre for the Study of Decision-Making Uncertainty, Division of Psychology and Language Sciences,University College London, UK

2Department of Experimental Psychology, University College London, UK

Introduction

The aim of this study was to evaluate the effectiveness of ‘Uncertainty Experts’ (UE), a novel three-part educational programme aimed at raising tolerance to uncertainty. The programme mixes science-based education, documentary-style testimonials from individuals who have been through periods of great uncertainty, and audience interaction. The goal is to empower the audience with effective coping mechanisms for dealing with the difficult emotions caused by uncertainty, and then move beyond these to gain the potential benefits that can be harnessed during uncertain times. UE was designed in response to the needs of UK employers seeking workshops around resilience, wellbeing, collaboration and psychological safety as a result of a perceived increase in uncertainty. This study sits within a body of burgeoning research exploring the benefits of uncertainty tolerance and related concepts for application across people management, employment, personal learning and development, and organisational development.

The study compares an audience going through the UE experience to a control group, comparing each before and after on several measures taken together to encapsulate ‘uncertainty tolerance’. These are the Positive and Negative Affect Schedule (PANAS), used to measure negative and positive responses to uncertainty (Watson, Clark & Tellegen, 1988); Need for Closure (NFC), used to measure a willingness to remain in a state of uncertainty / ambiguity (Webster & Kruglanski, 1994; Roets & Van Hiel, 2011) and the New General Self-Efficacy Scale (NGSE), to measure changing attitudes towards self around uncertainty (Chen, Gully & Eden, 2001). As well as these measures on changing emotional responses and attitudes to uncertainty, the study also tested changing behaviour using the Balloon Analogue Risk Task (BART) (Lejuez et al. 2002), which is a simple game designed to measure an individual’s willingness to take risks / tolerance to uncertainty.

The findings as discussed below, suggest that Uncertainty Experts was successful in improving tolerance to uncertainty. We also believe that this novel educational format combining documentary-style storytelling with interaction and introspection can be successfully adapted to other areas of workplace learning or behavioural change. 

Context

Uncertainty has always been a core feature of daily human existence, but the lives of individuals in the modern globally connected world are unprecedented in the complexity and uncertainty they have to deal with (Ahir, Bloom & Furceri, 2018). Uncertainty has often been primarily seen as a negative, impacting emotional states as well as cognitive and decision-making abilities, and so to be avoided (Ladouceur, Talbot & Dugas 1997; Anderson, Carleton, Diefenbach & Han, 2019; Carleton, 2016). However much psychological work shows that this orientation is not inevitable (Shamionov, 2017) and that times of uncertainty, if dealt with appropriately, can be used to break unhealthy habits, develop new creative ways of thinking, and undertake new ventures and opportunities (Clark. 2015, Garrison, Lee & Ali, 2017; Kim, Rhee, Ha, Yang & Lee, 2016).

This new conceptualisation of uncertainty is highly relevant across firms today. It is in the interest of firms that their employees at every level have high tolerance to uncertainty. Uncertainty is a core feature of daily decision making for every employee, and individuals with low tolerance for uncertainty are more ‘cognitively vulnerable’ (Dugas, Freeson & Ladoucer, 1997), more prone to anxiety (Grupe & Nitschke, 2013), incapable of taking calculated risks (Morriss, Christakou & Van Reekum, 2016) and indecisive (Rassin & Muris, 2005). Conversely, those high in tolerance to uncertainty are not only more resilient and less stressed (Blanuša, Barzut & Knežević, 2021), but crucially are also more creative (Zenasni, Besançon & Lubart, 2008), more innovative problem solvers (Jensen, Kind, Morrison & Heimberg, 2014) and more effective decision makers (Pavlova & Kornilova, 2013).

The Uncertainty Experts

        In the first episode of UE, the audience is taught to manage the often-negative emotions associated with uncertainty (Anderson, Carleton, Diefenbach & Han, 2019). They are taught to become more aware of rising fear responses, as well as maladaptive coping mechanisms for dealing with that fear (Morriss, Christakou & Van Reekum, 2016). They are also taught about the inherent malleability of the brain, and by extension that their current emotional and behavioural responses to uncertainty are not fixed but can be changed (Shamionov, 2017).

In the second episode, moving past merely dealing with the negative response, the focus is on becoming more open to uncertainty, laying the groundwork for greater possibility of positive responses (Roets & Van Hiel, 2011). The audience is taught to be more open-minded and to move towards a place of acceptance of uncertainty as an inevitable feature of daily life.

In the final episode, the audience is challenged to consider how uncertainty can in fact be a positive. They are taught to recognise it as a time of change and flux, which shakes us out of our normal limiting habits and routines (Lally & Gardner, 2013) into an opportunity to grow and learn as individuals with strategies that can be applied at personal and group level. The present paper seeks to determine, via a range of psychometric measures, if a pilot run of this programme achieved these aims.

Method

Three hundred and nineteen participants took part in the study and the demographic breakdown can be seen below in Table 1.

Table 1. Demographics for both the UE and control groups combined.

The study used a mixed between-within subjects design, taking pre/post responses on all measures for both the control and UE group. The control group (n=120) undertook a concurrent workshop on storytelling while the UE group (n=199) received three uncertainty-experts sessions, across three consecutive weeks. Participants in both groups completed the following range of measures before and after their sessions.

The Positive and Negative Affect Schedule (PANAS) was used to measure attitudes towards uncertainty. PANAS is a 20-item scale developed by Watson, Clark & Tellegen (1988). There are 10 ‘positive’ emotions and 10 ‘negative’, which are summed separately producing separate positive and negative scores. For this study participants were asked to fill out this scale particularly when thinking about their experience of uncertainty.

The New General Self-Efficacy (NGSE) scale, developed by Chen, Gully & Eden (2001), was also used. A final separate question was also developed based on the NGSE to ask about self-efficacy specifically relating to uncertainty. The Need for Closure (NFC) scale, developed by Roets and Van Hiel (2011), was used to rate participants’ ability to handle elements of uncertainty such as ambiguity.

The Balloon Analogue Risk Task (BART) is a simple game developed by Lejuez et al (2002) where each round participants press a button to pump up a balloon to earn rewards but which has a chance of exploding on each pump, and no reward. This provides a measure of willingness to take risks / engage with uncertainty. Coupled with this, participants were asked to complete the same PANAS 20-item scale again but specifically to reference their emotional experience of the BART.

Findings

A significant change in mean scores in the expected direction was found across all the measures for the UE group. Most measures also changed for the control group, with the exception of the BART and the PANAS related to the BART. Repeated measures ANOVA was used to compare the change between the two groups and for the majority of measures a significant or borderline interaction effect was found with the UE group outperforming the control group: the NGSE and the additional self efficacy question both increased more for the UE group, and both positive and negative PANAS scores related to the BART also increased significantly more. Both PANAS positive and the BART showed borderline effects in the direction of UE. The summary of results can be seen below in Table 2 and full details of all analyses can be found in Appendix 1.

Table 2. For each of the measures used in the study, in the first column the measure and the maximum possible score on that measure is given. In the next two columns the change, in bold, and pre->post scores in brackets are given for both conditions. In the final column the key results from the ANOVA of the interaction between pre->post and condition are given, providing the group which outperformed the other on that measure (and by what amount), as well as the p-value for the interaction. The full details can be found in Appendix 1.

Table 2. In all cases a * indicates a significant effect at p<.05.

Practical Implications

This study demonstrates the effectiveness of an innovative educational programme which can be delivered en masse to large numbers of individuals. The programme has been shown to improve not just attitudes and beliefs but also behaviours around uncertainty. These findings show important potential applications within the context of work as well as for individual wellbeing.  

Across both PANAS measures, UE appears to increase positive emotions towards uncertainty, and may also reduce negative emotions. It also appears to increase self-efficacy in its audience (NGSE), and particularly self-efficacy towards uncertainty. It may also increase the audience’s willingness to remain in a state of ambiguity (NFC), an important skill to explore within a work environment in the modern age.

Beyond this, it has also been shown to change behaviours and emotional responses to risk and uncertainty tolerance, as illustrated by the BART task and associated PANAS measures. In corporate roles where risk and uncertainty are often confused, attitudes towards uncertainty are vital to leadership skills (Pavlova & Kornilova, 2013). It is possible for individuals to engage positively with the principle of risk-taking but in practice find it uncomfortable. The results here indicate both an increase in attitude as well as the ability for taking risks.

UE has been shown to increase scores more than the control group on both the NGSE and a further question based upon the NGSE specifically focused on uncertainty. The NGSE measures the “belief in one’s capabilities to mobilize the motivation, cognitive resources, and courses of action needed to meet given situational demands” (Wood & Bandura, 1989, p.408). Higher scores on this general and fundamental attribute following the UE course are thought to lead to greater job attitudes (Saks, 1995), training proficiency (Martocchio & Judge, 1997), and job performance (Stajkovic & Luthans, 1998).

UE has also been shown to reduce scores on the ‘Need for Closure’ (NFC) scale. Kruglanski (1990) defined ‘need for closure’ as the desire for “an answer on a given topic, any answer … compared to confusion and ambiguity” (p.337). This represents an individual’s ability to be able to work under uncertain conditions without feeling the need to quickly come to a definitive answer – to tolerate ambiguity. This is related to intellectual humility, the ability to recognise one might be wrong (Leary et al, 2017) and it also favors creative thinking and behaviors as it enables individuals to resist the urge towards partial or non-optimal solutions to complex problems (Merrotsy, 2013). This is highly relevant for today’s world of work with critical thinking and problem solving consistently named the key skills for the ambiguity and upheaval of the 21st century, as summed up by The World Economic Forum's key skills for 2025.

Scores on the NFC scale decreased similarly in the control group. This may be because the control group was a storytelling workshop, delivered by a charismatic facilitator with more of a personal development angle than anticipated, offering participants ‘confidence in ‘telling your story’, and so this could have also had an effect on this measure. The use of a different control group in future may allow us to pick this apart. Alternatively, we may be seeing a demand characteristics effect in both groups where participants are responding to their beliefs about researcher expectations. However since we do not see this effect in the control group for a range of other measures, this seems unlikely to explain this result.

We have found an increase in positive emotions on the PANAS scale (general feelings towards uncertainty) for both groups and tentative evidence that the UE group increases this more than the control group. Conversely, we see the opposite pattern for the negative emotions with the control group tentatively leading to a greater decrease than the UE group. One possible explanation for this mixed finding is that participants may not have interpreted this question in the way we intended. Furthermore participants may struggle to think about how uncertainty as an abstract concept makes them feel. For this reason, given that the control group undertook a storytelling workshop, which is very likely to have an emotional effect upon the participants, it is possible that this question is not sufficiently tapping into participants’ actual emotional response to uncertainty per se, but instead may be overly influenced by their current emotional state (Askim & Knardahl, 2021). One way to improve this question in future may be to present participants with a particular uncertainty scenario and rate their emotions with respect to that. This should provide a more vivid focus for reflection rather than the abstract concept of uncertainty.

 Interestingly, participants in the EU group also showed a considerable increase in their willingness to take risks on the BART, on average being willing to pump around two more times after UE compared to the control group. Furthermore, we have found that while engaging in this they experienced fewer negative emotions and more positive emotions on the PANAS scale related specifically to that task compared to the control group. Overall therefore we seem to have found that UE makes participants both behaviourally and emotionally more tolerant of uncertainty. A relationship between uncertainty tolerance and a greater willingness to take risks has been demonstrated previously in the IOWA gambling task (Korniolva, Chumakova & Kornilov, 2018), so it is validating to see that UE also created this effect, as its primary aim is to increase tolerance to uncertainty.

Further Application

Reinforcing the results of this paper on the efficacy of Uncertainty Experts, the takeup of UE has been rapid, reflecting a felt need in the community for this content. As a result of the pilot, numerous organisations have come forward with approaches to secure places in future programs, or have ‘in-house’ adaptations of UE designed for their teams. These firms range from global brands including Google, Netflix and Apple to long established businesses going through industrial transformation including Mercedes, Twinings and Lego to public bodies including the civil service, home office, local authorities and the BBC.

Even though these organisations have booked small numbers of places, or commissioned exploratory work, the breadth of cross sector interest in the topic suggests potential for wide scale application, and the pervasiveness of the impact of uncertainty.

Conclusion

In summary, this pilot of the Uncertainty Experts programme has produced significant improvements in all measures taken related to uncertainty tolerance and self-efficacy when faced with uncertainty. The majority of these also improved more than the control group, which suggests that these effects are not due to demand characteristics and represent real effects upon the audience after taking part in an Uncertainty Experts series.

Heightened by the collision of Brexit, the climate crisis and the COVID-19 pandemic, uncertainty is at an all-time high (Ahir et al. 2018). Both in and out of the workplace, people are faced with high levels of uncertainty which impact them personally and professionally. Coping with such levels of uncertainty takes an emotional toll, and failing to manage these emotions effectively can lead to cognitive fatigue and lower levels of wellbeing, and ultimately to a less productive and resilient workforce. Further studies and research are now being carried out to better understand, test and explore the mechanism behind ‘The Uncertainty Experts’ intervention and how this can be scaled to provide both individuals and firms with an innovative solution to navigate the unprecedented complexity and uncertainty of the modern world.

References

Ahir, H., Bloom, N., & Furceri, D. (2022). The world uncertainty index (No. w29763). National Bureau of Economic Research.

Anderson, E. C. Carleton R. N. Diefenbach, M. Han, P. K. (2019) The Relationship Between Uncertainty and Affect. Frontiers in Psychology 10.

Askim, K., & Knardahl, S. (2021). The influence of affective state on subjective-report measurements: evidence from experimental manipulations of mood. Frontiers in psychology, 12, 159.

Blanuša, J., Barzut, V., & Knežević, J. (2021). Intolerance of uncertainty and fear of COVID-19 moderating role in relationship between job insecurity and work-related distress in the Republic of Serbia. Frontiers in Psychology, 12.

Chen, G., Gully, S. M., & Eden, D. (2001). Validation of a new general self-efficacy scale. Organizational research methods, 4(1), 62-83.

Clark, A. (2015). Surfing uncertainty: Prediction, action, and the embodied mind. New York, NY: Oxford University Press

Crawford, John R.; Henry, Julie D. (2004). "The Positive and Negative Affect Schedule (PANAS): Construct validity, measurement properties and normative data in a large non-clinical sample". British Journal of Clinical Psychology. 43 (3): 245–265.

Dugas M. J., Freeston M. H., Ladouceur R. (1997). Intolerance of uncertainty and problem orientation in worry. Cognit. Ther. Res. 21 593–606.

Garrison, Y. L., Lee, K. H., and Ali, S. R. (2017). Career identity and life satisfaction: the mediating role of tolerance for uncertainty and positive/negative affect. J. Career Dev. 44, 516–529.

Grupe, DW, Nitschke JB (2013) Uncertainty and anticipation in anxiety: an integrated neurobiological and psychological perspective. Nat Rev Neurosci. 14(7):488-501.

Carleton, R. N. (2016). Fear of the unknown: one fear to rule them all? J. Anxiety Disord. 41, 5–21.

Jensen, D., Kind, A. J., Morrison, A. S., & Heimberg, R. G. (2014). Intolerance of uncertainty and immediate decision-making in high-risk situations. Journal of Experimental Psychopathology, 5(2), 178-190.

Kim, B., Rhee, E., Ha, G., Yang, J., and Lee, S. M. (2016). Tolerance of uncertainty: links to happenstance, career decision self-efficacy, and career satisfaction. Career Dev. Q. 64, 140–152.

Kornilova, T. V., Chumakova, M. A., and Kornilov, S. A. (2018). Tolerance and intolerance for uncertainty as predictors of decision making and risk acceptance in gaming strategies of the Iowa gambling task. Psychol. Russ. State Art 11, 86–95.

Kruglanski, A. W. (1990). Motivations for judging and knowing: Implications for causal attribution. In E. T. Higgins & R. M. Sorrentino (Eds.), The handbook of motivation and cognition: Foundation of social behavior (Vol. 2, pp. 333–368). New York: Guilford Press.

Lally, P., & Gardner, B. (2013). Promoting habit formation. Health psychology review, 7(sup1), S137-S158.

Leary, M. R., Diebels, K. J., Davisson, E. K., Jongman-Sereno, K. P., Isherwood, J. C., Raimi, K. T., ... & Hoyle, R. H. (2017). Cognitive and interpersonal features of intellectual humility. Personality and Social Psychology Bulletin, 43(6), 793-813.

Ladouceur R, Talbot F, Dugas MJ. (1997) Behavioral expressions of intolerance of uncertainty in worry. Experimental findings. Behav Modif. (3):355-71

Lejuez, C. W., Read, J. P., Kahler, C. W., Richards, J. B., Ramsey, S. E., Stuart, G. L., ... & Brown, R. A. (2002). Evaluation of a behavioral measure of risk taking: the Balloon Analogue Risk Task (BART). Journal of Experimental Psychology: Applied, 8(2), 75.

Martocchio, J. J., & Judge, T. A. (1997). Relationships between conscientiousness and learning in employee training: Mediating influences of self-deception and self-efficacy. Journal of Applied Psychology, 82, 764-773.

Merrotsy, P. (2013). Tolerance of ambiguity: a trait of the creative personality?. Creativity Research Journal, 25(2), 232-237.

Morriss, J., Christakou, A., & Van Reekum, C. M. (2016). Nothing is safe: Intolerance of uncertainty is associated with compromised fear extinction learning. Biological psychology, 121, 187-193.

Rassin, E., & Muris, P. (2005). Indecisiveness and the interpretation of ambiguous situations. Personality and individual Differences, 39(7), 1285-1291.

Roets, A., & Van Hiel, A. (2007). Separating ability from need: Clarifying the dimensional structure of the need for closure scale. Personality and Social Psychology Bulletin, 33(2), 266-280.

Pavlova, E. M., & Kornilova, T. V. (2013). Creativity and tolerance for uncertainty predict the engagement of emotional intelligence in personal decision making. Psychology in Russia: State of the Art. Scientific Yearbook, 6(4), 34-46.

Saks, A. M. (1995). Longitudinal field investigation of the moderating and mediating effects of self-efficacy on the relationship between training and newcomer adjustment. Journal of Applied Psychology, 80, 211-225.

Shamionov, R. M. (2017). Attitude to change and tolerance to uncertainty as predictors of adaptability and adaptive readiness. Russ. Psychol. J. 14, 90–104

Stajkovic, A. D., & Luthans, F. (1998). Self-efficacy and work-related performance: A metaanalysis. Psychological Bulletin, 124, 240-261

Watson, D., Clark, L. A., & Tellegen, A. (1988b). Development and validation of brief measures of positive and negative affect: The PANAS Scales. Journal of Personality and Social Psychology, 47, 1063–1070.

Webster, D. M., & Kruglanski, A. W. (1994). Individual differences in need for cognitive closure. Journal of Personality and Social Psychology, 67, 1049-1062.

Wood, R., & Bandura, A. (1989). Impact of conceptions of ability on self-regulatory mechanisms and complex decision making. Journal of Personality and Social Psychology, 56, 407-415.

Zenasni, F., Besançon, M., & Lubart, T. (2008). Creativity and tolerance of ambiguity: An empirical study. The Journal of Creative Behavior, 42(1), 61-73.

Appendices

Appendix 1

Findings (expanded)

On the general measure of PANAS, where participants were asked to reflect on their experience of uncertainty. For the UE group, mean positive scores significantly increased by 3.2 which was confirmed by a paired-samples t-test (t[168]=7.4, p<.001). For the control group, mean positive scores significantly increased by 1.9 (t[112]=2.7, p=.007). A repeated measures ANOVA with pre-post positive PANAS scores as the within subjects IV and condition (0=control, 1=UE) as the between subjects IV found a borderline significant interaction with the UE group outperforming the control group (F[1,280]=61.3, p=.081).

For negative PANAS scores, within the UE group, mean negative scores significantly decreased by 2.2 (t[168]=-4.7, p<.001). For the control group, mean negative scores significantly decreased by 4.3 (t[118]=-5.9, p<.001). A repeated measures ANOVA found a significant interaction effect with the control group outperforming the UE group (F[1,286]=6.5, p=.012).

For the self-efficacy question about uncertainty, mean scores significantly increased in the UE group by 2.9 (t[198]=19.6, p<.001). Mean scores also significantly increased for the control group by 0.6 (t[112]=3.8,p<.001). A repeated measures ANOVA found a significant interaction effect with the UE group outperforming the control group (F[1,310]=96.2, p<.001).

For the NGSE, mean scores significantly increased in the UE group by 2.4 (t[137]=6.7, p<.001). Mean scores also significantly increased in the control group by 1.1 (t[110]=3.3, p=.001). A repeated measures ANOVA found a significant interaction effect with the UE group outperforming the control group (F[1,247]=6.5, p=.011).

For the NFC, mean scores significantly decreased in the UE group by -6.1 (t[198]=-10.2, p<.001). Mean scores also significantly decreased for the control group by -4.9 (t[113]=4.6,p<.001). A repeated measures ANOVA found no significant interaction effect between the two groups (F[1,311]=1.1, p=.297).

For the BART, mean scores significantly increased in the UE group by 2.4 (t[137]=3.6,p<.001). Mean scores did not significantly increase in the control group (t[110]=.960,p=.339). A repeated measures ANOVA found a borderline significant interaction effect with the UE group outperforming the control group (F[1,247]=3.5, p=.061).

For the PANAS positive scores focused on the BART, mean scores in the UE group significantly increased by 1.8 (t[137]=2.9,p=.005). Mean scores did not significantly increase in the control group however (t[110]=1.1,p=.317). A repeated measures ANOVA found a significant interaction effect with the UE group outperforming the control group (F[1,247]=7.3, p=.007).

For the PANAS Negative scores focused on the BART, mean scores in the UE group significantly decreased by 2.0 (t[137]=-5.4,p<.001). Mean scores in the control group also significantly decreased by 0.9 (t[110]=2.7,p=.007). A repeated measures ANOVA found a significant interaction effect with the UE group outperforming the control group (F[1,247]=4.2, p=.042).