Abstract
In many laboratory visuo-motor decision tasks, subjects compensate for their own visuo-motor error, earning close to the maximum reward possible. To do so, they must combine information about the distribution of possible error with values associated with different movement outcomes. The optimal solution is a potentially difficult computation that presupposes knowledge of the probability density function (pdf) of visuo-motor error associated with each possible planned movement. It is unclear how the brain represents such pdfs or computes with them. In three experiments, we used a forced-choice method to reveal subjects' internal representations of their spatial visuo-motor error in a speeded reaching movement. Although subjects' objective distributions were unimodal, close to Gaussian, their estimated internal pdfs were typically multimodal and were better described as mixtures of a small number of distributions differing only in location and scale. Mixtures of a small number of uniform distributions outperformed other mixture distributions, including mixtures of Gaussians.
This is a preview of subscription content, access via your institution
Access options
Subscribe to this journal
Receive 12 print issues and online access
$209.00 per year
only $17.42 per issue
Buy this article
- Purchase on Springer Link
- Instant access to full article PDF
Prices may be subject to local taxes which are calculated during checkout
Similar content being viewed by others
References
Bach, D.R. & Dolan, R.J. Knowing how much you don't know: a neural organization of uncertainty estimates. Nat. Rev. Neurosci. 13, 572–586 (2012).
Maloney, L.T. & Zhang, H. Decision-theoretic models of visual perception and action. Vision Res. 50, 2362–2374 (2010).
Trommershäuser, J., Maloney, L.T. & Landy, M.S. Decision making, movement planning and statistical decision theory. Trends Cogn. Sci. 12, 291–297 (2008).
Battaglia, P.W. & Schrater, P.R. Humans trade off viewing time and movement duration to improve visuomotor accuracy in a fast reaching task. J. Neurosci. 27, 6984–6994 (2007).
Faisal, A.A. & Wolpert, D.M. Near optimal combination of sensory and motor uncertainty in time during a naturalistic perception-action task. J. Neurophysiol. 101, 1901–1912 (2009).
Hudson, T.E., Maloney, L.T. & Landy, M.S. Optimal compensation for temporal uncertainty in movement planning. PLOS Comput. Biol. 4, e1000130 (2008).
Körding, K.P. & Wolpert, D.M. Bayesian integration in sensorimotor learning. Nature 427, 244–247 (2004).
Trommershäuser, J., Maloney, L.T. & Landy, M.S. Statistical decision theory and trade-offs in the control of motor response. Spat. Vis. 16, 255–275 (2003).
Jazayeri, M. & Shadlen, M.N. Temporal context calibrates interval timing. Nat. Neurosci. 13, 1020–1026 (2010).
Wei, K. & Körding, K. Uncertainty of feedback and state estimation determines the speed of motor adaptation. Front. Comput. Neurosci. 4, 11 (2010).
Trommershäuser, J., Landy, M.S. & Maloney, L.T. Humans rapidly estimate expected gain in movement planning. Psychol. Sci. 17, 981–988 (2006).
Ma, W.J., Beck, J.M., Latham, P.E. & Pouget, A. Bayesian inference with probabilistic population codes. Nat. Neurosci. 9, 1432–1438 (2006).
Huys, Q.J., Zemel, R.S., Natarajan, R. & Dayan, P. Fast population coding. Neural Comput. 19, 404–441 (2007).
Pouget, A., Beck, J.M., Ma, W.J. & Latham, P.E. Probabilistic brains: knowns and unknowns. Nat. Neurosci. 16, 1170–1178 (2013).
Maloney, L.T. Statistical decision theory and biological vision. in Perception and the Physical World: Psychological and Philosophical Issues in Perception (eds. D. Heyer & R. Mausfeld) 145–189 (Wiley, New York, 2002).
Haruno, M., Wolpert, D.M. & Kawato, M. Hierarchical MOSAIC for movement generation. Int. Congr. Ser. 1250, 575–590 (2003).
Maloney, L.T. & Mamassian, P. Bayesian decision theory as a model of human visual perception: testing Bayesian transfer. Vis. Neurosci. 26, 147–155 (2009).
Akaike, H. A new look at the statistical model identification. IEEE Trans. Automat. Contr. 19, 716–723 (1974).
Hurvich, C.M. & Tsai, C.-L. Regression and time series model selection in small samples. Biometrika 76, 297–307 (1989).
Stephan, K.E., Penny, W.D., Daunizeau, J., Moran, R.J. & Friston, K.J. Bayesian model selection for group studies. Neuroimage 46, 1004–1017 (2009).
Zhang, H., Daw, N.D. & Maloney, L.T. Testing whether humans have an accurate model of their own motor uncertainty in a speeded reaching task. PLOS Comput. Biol. 9, e1003080 (2013).
Oruç, I., Maloney, L.T. & Landy, M.S. Weighted linear cue combination with possibly correlated error. Vision Res. 43, 2451–2468 (2003).
Acerbi, L., Wolpert, D.M. & Vijayakumar, S. Internal representations of temporal statistics and feedback calibrate motor-sensory interval timing. PLOS Comput. Biol. 8, e1002771 (2012).
Daw, N.D., Courville, A.C. & Dayan, P. Semi-rational models of conditioning: the case of trial order. in The Probabilistic Mind: Prospects for Bayesian Cognitive Science (eds. N. Chater & M. Oaksford) 431–452 (Oxford University Press, Oxford, 2008).
Gershman, S. & Wilson, R. The neural costs of optimal control. Adv. Neural Inf. Process. Syst. 23, 712–720 (2010).
Vul, E., Goodman, N.D., Griffiths, T.L. & Tenenbaum, J.B. One and done? Optimal decisions from very few samples. Cogn. Sci. 38, 599–637 (2014).
Sanborn, A.N., Griffiths, T.L. & Navarro, D.J. Rational approximations to rational models: alternative algorithms for category learning. Psychol. Rev. 117, 1144 (2010).
Vul, E., Hanus, D. & Kanwisher, N. Attention as inference: selection is probabilistic; responses are all-or-none samples. J. Exp. Psychol. Gen. 138, 546–560 (2009).
Daw, N.D. & Courville, A. The pigeon as particle filter. in Advances in Neural Information Processing Systems (ed. J.C. Platt, D. Koller, Y. Singer & S. Roweis) 369–376 (MIT Press, 2007).
Maloney, L.T. Evaluation of linear models of surface spectral reflectance with small numbers of parameters. J. Opt. Soc. Am. A 3, 1673–1683 (1986).
Körding, K.P. & Wolpert, D.M. The loss function of sensorimotor learning. Proc. Natl. Acad. Sci. USA 101, 9839–9842 (2004).
Todorov, E. & Jordan, M.I. Optimal feedback control as a theory of motor coordination. Nat. Neurosci. 5, 1226–1235 (2002).
Harris, C.M. & Wolpert, D.M. Signal-dependent noise determines motor planning. Nature 394, 780–784 (1998).
Wolpert, D.M., Ghahramani, Z. & Jordan, M.I. An internal model for sensorimotor integration. Science 269, 1880–1882 (1995).
Hamilton, B.H. Does entrepreneurship pay? An empirical analysis of the returns to self-employment. J. Polit. Econ. 108, 604–631 (2000).
Harvey, C.R. & Siddique, A. Conditional skewness in asset pricing tests. J. Finance 55, 1263–1295 (2000).
Kraus, A. & Litzenberger, R.H. Skewness preference and the valuation of risk assets. J. Finance 31, 1085–1100 (1976).
Moskowitz, T.J. & Vissing-Jørgensen, A. The returns to entrepreneurial investment: a private equity premium puzzle? Am. Econ. Rev. 92, 745–778 (2002).
Pelli, D.G. The VideoToolbox software for visual psychophysics: transforming numbers into movies. Spat. Vis. 10, 437–442 (1997).
Brainard, D.H. The psychophysics toolbox. Spat. Vis. 10, 433–436 (1997).
Erev, I. et al. A choice prediction competition: choices from experience and from description. J. Behav. Decis. Mak. 23, 15–47 (2010).
Rasmussen, C.E. & Williams, C.K.I. Gaussian Processes for Machine Learning (MIT Press, 2006).
Acknowledgements
The authors would like to thank J. Tee for inspiring discussions. H.Z. and L.T.M. were supported by grant EY019889 from the US National Institutes of Health and L.T.M. by an award from the Alexander v. Humboldt Foundation. N.D.D. was supported by a Scholar Award from the McKnight Foundation and a James S. McDonnell Foundation Award in Understanding Human Cognition.
Author information
Authors and Affiliations
Contributions
H.Z. designed and performed the experiments, analyzed the data and wrote the manuscript. N.D.D. and L.T.M. supervised the project and improved the manuscript.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing financial interests.
Integrated supplementary information
Supplementary Figure 1 Visualization of subjects’ internal pdfs in Experiment 1.
Each panel is for one subject. Shaded regions denote ±SEM. is in the unit of the subject’s horizontal standard deviation estimated from the reaching task.
Supplementary Figure 2 Equivalent ratio as a function of Triple Number in Experiment 1.
Each panel is for one subject. In the choice task, there were 12 different Triples, whose paired Singles were adjusted in width by adaptive procedures (Fig. 2e). For each subject and each Triple, we defined the equivalent width of a Triple as the width of the Single that the subject judged to be as “hittable” as the Triple. We defined the equivalent ratio as the ratio of the equivalent width of the Triple to the total width of its three rectangles. The equivalent ratio provides a measure for the probability density function in subjects’ internal pdf, roughly indicating the mean probability density over the side rectangles of the Triple relative to the probability density around the center. The 12 Triples in the plot are sorted by the horizontal position of the right rectangle in the Triple. Note the measured equivalent ratios (black dots and line) had abrupt changes at adjacent Triples. The U-mix prediction (green line) captured the abrupt changes better than the predictions of the Gaussian prediction (blue line). The predictions of other mixture models (especially mG-mix) were close to that of U-mix and were omitted in the plot for simplicity.
Supplementary Figure 3 Visualization of subjects’ internal pdfs in Experiment 2.
Each panel is for one subject. Shaded regions denote ±SEM. is in the unit of the subject’s vertical standard deviation estimated from the reaching task.
Supplementary Figure 4 Model fits of subjects’ internal pdfs in Experiment 2.
(a) mG-mix model. (b) U-mix model. Each panel is for the probability density function of one subject. is in the unit of the subject’s vertical standard deviation estimated from the reaching task. Subjects are in the same order as in Supplementary Figure 3.
Supplementary Figure 5 Model fits of subjects’ internal pdfs in Experiment 3 for the mG-mix model.
Each panel is for one subject. The pair of numbers inside each panel denotes the numbers of components for the horizontal and vertical directions. Green and gray curves respectively denote probability density functions in the horizontal and vertical directions. is in the unit of the subject’s standard deviation (averaged across the horizontal and vertical directions) estimated from the reaching task.
Supplementary Figure 6 Model fits of subjects’ internal pdfs in Experiment 3 for the U-mix model.
Each panel is for one subject. The pair of numbers inside each panel denotes the numbers of components for the horizontal and vertical directions. Green and gray curves respectively denote probability density functions in the horizontal and vertical directions. is in the unit of the subject’s standard deviation (averaged across the horizontal and vertical directions) estimated from the reaching task. Subjects are in the same order as in Supplementary Figure 5.
Supplementary Figure 7 mG-mix simulation for Körding and Wolpert31.
Similar to Figure 7, except that the U-mix representation was replaced by an mG-mix representation (Methods online). (a) Illustration of the mG-mix representation. (b) Simulated aim points vs. data.
Supplementary information
Supplementary Text and Figures
Supplementary Figures 1–7 (PDF 981 kb)
Rights and permissions
About this article
Cite this article
Zhang, H., Daw, N. & Maloney, L. Human representation of visuo-motor uncertainty as mixtures of orthogonal basis distributions. Nat Neurosci 18, 1152–1158 (2015). https://doi.org/10.1038/nn.4055
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1038/nn.4055
This article is cited by
-
Discrete confidence levels revealed by sequential decisions
Nature Human Behaviour (2020)
-
Sub-optimality in motor planning is not improved by explicit observation of motor uncertainty
Scientific Reports (2019)
-
Towards a neuroscience of active sampling and curiosity
Nature Reviews Neuroscience (2018)
-
Functional brain networks involved in decision-making under certain and uncertain conditions
Neuroradiology (2018)
-
Sub-optimality in motor planning is retained throughout 9 days practice of 2250 trials
Scientific Reports (2016)