Article | Published:

Anxious individuals have difficulty learning the causal statistics of aversive environments

Nature Neuroscience volume 18, pages 590596 (2015) | Download Citation



Statistical regularities in the causal structure of the environment enable us to predict the probable outcomes of our actions. Environments differ in the extent to which action-outcome contingencies are stable or volatile. Difficulty in being able to use this information to optimally update outcome predictions might contribute to the decision-making difficulties seen in anxiety. We tested this using an aversive learning task manipulating environmental volatility. Human participants low in trait anxiety matched updating of their outcome predictions to the volatility of the current environment, as predicted by a Bayesian model. Individuals with high trait anxiety showed less ability to adjust updating of outcome expectancies between stable and volatile environments. This was linked to reduced sensitivity of the pupil dilatory response to volatility, potentially indicative of altered norepinephrinergic responsivity to changes in this aspect of environmental information.

Access optionsAccess options

Rent or Buy article

Get time limited or full article access on ReadCube.


All prices are NET prices.


  1. 1.

    & Cognitive processes in anxiety. Adv. Behav. Res. Ther. 5, 51–62 (1983).

  2. 2.

    & Anxiety and decision-making. Biol. Psychiatry 72, 113–118 (2012).

  3. 3.

    & Emotion and decision-making: affect-driven belief systems in anxiety and depression. Trends Cogn. Sci. 16, 476–483 (2012).

  4. 4.

    , , & Toward a definition of intolerance of uncertainty: a review of factor analytical studies of the Intolerance of Uncertainty Scale. Clin. Psychol. Rev. 31, 1198–1208 (2011).

  5. 5.

    , & Fearing the unknown: a short version of the Intolerance of Uncertainty Scale. J. Anxiety Disord. 21, 105–117 (2007).

  6. 6.

    & Uncertainty, neuromodulation, and attention. Neuron 46, 681–692 (2005).

  7. 7.

    , , & The neural representation of unexpected uncertainty during value-based decision making. Neuron 79, 191–201 (2013).

  8. 8.

    , , & Learning the value of information in an uncertain world. Nat. Neurosci. 10, 1214–1221 (2007).

  9. 9.

    & Reinforcement Learning (MIT Press, 1998).

  10. 10.

    & A theory of Pavlovian conditioning: variations in the effectiveness of reinforcement and nonreinforcement. in Classical Conditioning II: Current Research and Theory (eds. A.H. Black & W.F. Prokasy) 64–99 (Appleton-Centuary-Crofts, 1972).

  11. 11.

    , , & Associative learning of social value. Nature 456, 245–249 (2008).

  12. 12.

    , , , & Fear-conditioning mechanisms associated with trait vulnerability to anxiety in humans. Neuron 69, 563–571 (2011).

  13. 13.

    , & Pupil dilation signals surprise: evidence for noradrenaline's role in decision making. Front. Neurosci. 5, 115 (2011).

  14. 14.

    et al. Rational regulation of learning dynamics by pupil-linked arousal systems. Nat. Neurosci. 15, 1040–1046 (2012).

  15. 15.

    , & The effects of neural gain on attention and learning. Nat. Neurosci. 16, 1146–1153 (2013).

  16. 16.

    , , & Pupil diameter tracks changes in control state predicted by the adaptive gain theory of locus coeruleus function. Cogn. Affect. Behav. Neurosci. 10, 252–269 (2010).

  17. 17.

    & An integrative theory of locus coeruleus–norepinephrine function: adaptive gain and optimal performance. Annu. Rev. Neurosci. 28, 403–450 (2005).

  18. 18.

    , & New developments in the neurobiological basis of anxiety disorders. Psychopharmacol. Bull. 36 (suppl. 2), 49–67 (2002).

  19. 19.

    & Neurobiological basis of anxiety disorders. in Neuropsychopharmacology: the Fifth Generation of Progress (eds. Davis, K.L., Charney, D.S., Coyle, J.T. & Nemeroff, C.) 901–930 (American College of Neuropsychopharmacology, 2002).

  20. 20.

    , & Manual for the State-Trait Anxiety Inventory (STAI) (Consulting Psychologists Press, 1983).

  21. 21.

    et al. Influence of uncertainty and surprise on human corticospinal excitability during preparation for action. Curr. Biol. 18, 775–780 (2008).

  22. 22.

    Information Theory, Inference and Learning Algorithms (Cambridge University Press, 2003).

  23. 23.

    et al. Dissociable effects of surprise and model update in parietal and anterior cingulate cortex. Proc. Natl. Acad. Sci. USA 110, 3660–3669 (2013).

  24. 24.

    et al. DSM-5 and RDoC: progress in psychiatry research? Nat. Rev. Neurosci. 14, 810–814 (2013).

Download references


The authors would like to thank S. Hicks and C. Kennard for use of the eye-tracking system. This research was supported by European Research Council grant GA 260932 and US National Institutes of Health grant R01MH091848.

Author information


  1. Functional MRI of the Brain Centre, Nuffield Department of Clinical Neurosciences, John Radcliffe Hospital, Oxford, UK.

    • Michael Browning
    • , Timothy E Behrens
    • , Gerhard Jocham
    • , Jill X O'Reilly
    •  & Sonia J Bishop
  2. Center for Behavioral Brain Sciences, University of Magdeburg, Magdeburg, Germany.

    • Gerhard Jocham
  3. Department of Psychology and Helen Wills Neuroscience Institute, University of California, Berkeley, California, USA.

    • Sonia J Bishop


  1. Search for Michael Browning in:

  2. Search for Timothy E Behrens in:

  3. Search for Gerhard Jocham in:

  4. Search for Jill X O'Reilly in:

  5. Search for Sonia J Bishop in:


M.B. and S.J.B. wrote the manuscript. M.B., S.J.B. and T.E.B. designed the task. M.B. collected the data. All of the authors contributed to data analysis. T.E.B. developed the Bayesian model. All of the authors commented on the manuscript.

Competing interests

The authors declare no competing financial interests.

Corresponding author

Correspondence to Sonia J Bishop.

Integrated supplementary information

Supplementary figures

  1. 1.

    Calibration of Electrical Shocks.

  2. 2.

    Effects of volatility upon participants’ learning rates in a structurally equivalent reward task.

  3. 3.

    Influence of potential shock magnitude and prior outcome history on actual and simulated behavior choice as a function of block volatility (a,b) and participant anxiety (c-f).

  4. 4.

    Relationship between trait anxiety and Bayesian volatility when multiple parameters, including a decay function, are allowed to compete for influence over a dynamic learning rate.

  5. 5.

    Shock receipt on difficult trials (where the two options were close in expected value) was positively correlated with trait anxiety (a), and negatively correlated with both change in learning rate between stable and volatile blocks (b) and modulation of post-outcome pupil dilation by trial volatility (c).

  6. 6.

    No significant association was observed between Trait Anxiety and Risk Preference or Inverse Decision Temperature.

  7. 7.

    Analyses of reaction time and pupil dilation data using non-Bayesian estimates of trial volatility and outcome surprise

  8. 8.

    Baseline corrected, z-transformed eyetracker traces from the aversive learning task.

  9. 9.

    The relationship between trait anxiety and post outcome pupil dilation as a function of trial-wise estimates of volatility, controlling for trial number within each block.

  10. 10.

    Results from additional pupil analyses investigating the interaction between outcome (shock received versus no shock received), and volatility (a) and surprise (b).

  11. 11.

    Interaction of estimated trialwise volatility and surprise on pupil dilation post outcome.

Supplementary information

PDF files

  1. 1.

    Supplementary Text and Figures

    Supplementary Figures 1–11 and Supplementary Modeling Note

About this article

Publication history





Further reading