Nothing Special   »   [go: up one dir, main page]
Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Article
  • Published:

Encoding and decoding in parietal cortex during sensorimotor decision-making

Nature Neuroscience volume 17, pages 1395–1403 (2014)Cite this article

Subjects

This article has been updated

Abstract

It has been suggested that the lateral intraparietal area (LIP) of macaques plays a fundamental role in sensorimotor decision-making. We examined the neural code in LIP at the level of individual spike trains using a statistical approach based on generalized linear models. We found that LIP responses reflected a combination of temporally overlapping task- and decision-related signals. Our model accounts for the detailed statistics of LIP spike trains and accurately predicts spike trains from task events on single trials. Moreover, we derived an optimal decoder for heterogeneous, multiplexed LIP responses that could be implemented in biologically plausible circuits. In contrast with interpretations of LIP as providing an instantaneous code for decision variables, we found that optimal decoding requires integrating LIP spikes over two distinct timescales. These analyses provide a detailed understanding of neural representations in LIP and a framework for studying the coding of multiplexed signals in higher brain areas.

Access through your institution
Buy or subscribe

This is a preview of subscription content, access via your institution

Access options

Access through your institution

Buy this article

  • Purchase on SpringerLink
  • Instant access to full article PDF
Buy now

Prices may be subject to local taxes which are calculated during checkout

Figure 1: Decision-making task and classical analysis of LIP responses.
Figure 2: Encoding model, fitted components, and rate predictions.
Figure 3: Encoding model captures responses across a heterogeneous population.
Figure 4: Model-based single trial predictions.
Figure 5: Spike history effects captured by the post-spike filter.
Figure 6: Single-trial prediction and model-based decoding of spike trains.
Figure 7: Decoding from low-dimensional decoding weight space.
Figure 8: Targets absent manipulation.

Similar content being viewed by others

Change history

  • 04 September 2014

    In the version of this supplementary file originally posted online, there were several errors because an obsolete version was posted. The supplementary figures were misnumbered (1 as 4, 2 as 5, 3 as 6, 4 as 7, 5 as 8, 6 as 1, 7 as 2, and 8 as 3). In Supplementary Figure 2b, the comma was missing after "consecutive intervals." Supplementary Figure 3 referred to "equal quantile buckets," "saccade-locked kernel" and "effect from the model" instead of "equal-quantile buckets," "saccade-locked kernels" and "effect predicted from the model." The Supplementary Figure 4 legend read "exponential model is better" where it should have read "exponential model better fits the data." Supplementary Figure 9 referred to a Table 1 instead of Supplementary Table 1. The title for Supplementary Table 1 was absent, and the legend referred to the "first 3 singular values" instead of "first 3 singular vectors." In addition, there were four instances in the PDF version where a figure or table citation was replaced or accompanied by "Error! Reference source not found." The errors have been corrected in this file as of 4 September 2014.

References

  1. Platt, M.L. & Glimcher, P.W. Neural correlates of decision variables in parietal cortex. Nature 400, 233–238 (1999).

    Article  CAS  Google Scholar 

  2. Shadlen, M.N. & Newsome, W.T. Neural basis of a perceptual decision in the parietal cortex (area LIP) of the rhesus monkey. J. Neurophysiol. 86, 1916–1936 (2001).

    Article  CAS  Google Scholar 

  3. Huk, A.C. & Shadlen, M.N. Neural activity in macaque parietal cortex reflects temporal integration of visual motion signals during perceptual decision making. J. Neurosci. 25, 10420–10436 (2005).

    Article  CAS  Google Scholar 

  4. Kable, J.W. & Glimcher, P.W. The neural correlates of subjective value during intertemporal choice. Nat. Neurosci. 10, 1625–1633 (2007).

    Article  CAS  Google Scholar 

  5. Yang, T. & Shadlen, M.N. Probabilistic reasoning by neurons. Nature 447, 1075–1080 (2007).

    Article  CAS  Google Scholar 

  6. Beck, J.M. et al. Probabilistic population codes for Bayesian decision making. Neuron 60, 1142–1152 (2008).

    Article  CAS  Google Scholar 

  7. Seo, H., Barraclough, D.J. & Lee, D. Lateral intraparietal cortex and reinforcement learning during a mixed-strategy game. J. Neurosci. 29, 7278 (2009).

    Article  CAS  Google Scholar 

  8. Cisek, P. & Kalaska, J.F. Neural mechanisms for interacting with a world full of action choices. Annu. Rev. Neurosci. 33, 269–298 (2010).

    Article  CAS  Google Scholar 

  9. Mazurek, M.E., Roitman, J.D., Ditterich, J. & Shadlen, M.N. A role for neural integrators in perceptual decision making. Cereb. Cortex 13, 1257 (2003).

    Article  Google Scholar 

  10. Lo, C.C. & Wang, X.J. Cortico-basal ganglia circuit mechanism for a decision threshold in reaction time tasks. Nat. Neurosci. 9, 956–963 (2006).

    Article  CAS  Google Scholar 

  11. Wong, K.F., Huk, A.C., Shadlen, M.N. & Wang, X.J. Neural circuit dynamics underlying accumulation of time-varying evidence during perceptual decision making. Front. Comput. Neurosci. 1, 6 (2007).

    Article  Google Scholar 

  12. Fusi, S., Asaad, W.F., Miller, E.K. & Wang, X.-J. A neural circuit model of flexible sensorimotor mapping: learning and forgetting on multiple timescales. Neuron 54, 319–333 (2007).

    Article  CAS  Google Scholar 

  13. Wang, X.-J. Decision making in recurrent neuronal circuits. Neuron 60, 215–234 (2008).

    Article  CAS  Google Scholar 

  14. Ganguli, S. et al. One-dimensional dynamics of attention and decision making in LIP. Neuron 58, 15–25 (2008).

    Article  CAS  Google Scholar 

  15. Soltani, A. & Wang, X.J. Synaptic computation underlying probabilistic inference. Nat. Neurosci. 13, 112–119 (2010).

    Article  CAS  Google Scholar 

  16. Pillow, J.W., Paninski, L., Uzzell, V.J., Simoncelli, E.P. & Chichilnisky, E.J. Prediction and decoding of retinal ganglion cell responses with a probabilistic spiking model. J. Neurosci. 25, 11003–11013 (2005).

    Article  CAS  Google Scholar 

  17. Pillow, J.W. et al. Spatio-temporal correlations and visual signaling in a complete neuronal population. Nature 454, 995–999 (2008).

    Article  CAS  Google Scholar 

  18. Jacobs, A.L. et al. Ruling out and ruling in neural codes. Proc. Natl. Acad. Sci. USA 106, 5936–5941 (2009).

    Article  CAS  Google Scholar 

  19. Fernandes, H.L., Stevenson, I.H., Phillips, A.N., Segraves, M.A. & Kording, K.P. Saliency and saccade encoding in the frontal eye field during natural scene search. Cereb. Cortex published online, 10.1093/cercor/bht179 (17 July 2013).

  20. Paninski, L., Fellows, M., Shoham, S., Hatsopoulos, N. & Donoghue, J. Superlinear population encoding of dynamic hand trajectory in primary motor cortex. J. Neurosci. 24, 8551–8561 (2004).

    Article  CAS  Google Scholar 

  21. Truccolo, W., Eden, U.T., Fellows, M.R., Donoghue, J.P. & Brown, E.N. A point process framework for relating neural spiking activity to spiking history, neural ensemble and extrinsic covariate effects. J. Neurophysiol. 93, 1074–1089 (2005).

    Article  Google Scholar 

  22. Yu, B.M. et al. Gaussian-process factor analysis for low-dimensional single-trial analysis of neural population activity. J. Neurophysiol. 102, 614 (2009).

    Article  Google Scholar 

  23. Brown, E.N., Frank, L., Tang, D., Quirk, M. & Wilson, M. A statistical paradigm for neural spike train decoding applied to position prediction from ensemble firing patterns of rat hippocampal place cells. J. Neurosci. 18, 7411–7425 (1998).

    Article  CAS  Google Scholar 

  24. Barbieri, R., Wilson, M.A., Frank, L.M. & Brown, E.N. An analysis of hippocampal spatio-temporal representations using a Bayesian algorithm for neural spike train decoding. IEEE Trans. Neural Syst. Rehabil. Eng. 13, 131–136 (2005).

    Article  Google Scholar 

  25. Rorie, A.E., Gao, J., McClelland, J.L. & Newsome, W.T. Integration of sensory and reward information during perceptual decision-making in lateral intraparietal cortex (LIP) of the macaque monkey. PLoS ONE 5, e9308 (2010).

    Article  Google Scholar 

  26. Park, J. & Zhang, J. Sensorimotor locus of the buildup activity in monkey lateral intraparietal area neurons. J. Neurophysiol. 103, 2664–2674 (2010).

    Article  Google Scholar 

  27. Jenison, R.L., Rangel, A., Oya, H., Kawasaki, H. & Howard, M.A. Value encoding in single neurons in the human amygdala during decision making. J. Neurosci. 31, 331–338 (2011).

    Article  CAS  Google Scholar 

  28. Rishel, C.A., Huang, G. & Freedman, D.J. Independent category and spatial encoding in parietal cortex. Neuron 77, 969–979 (2013).

    Article  CAS  Google Scholar 

  29. Huk, A.C. & Meister, M.L. Neural correlates and neural computations in posterior parietal cortex during perceptual decision-making. Front. Integr. Neurosci. 6, 86 (2012).

    Article  Google Scholar 

  30. Gottlieb, J. & Goldberg, M.E. Activity of neurons in the lateral intraparietal area of the monkey during an antisaccade task. Nat. Neurosci. 2, 906–912 (1999).

    Article  CAS  Google Scholar 

  31. Meister, M.L.R., Hennig, J.A. & Huk, A.C. Signal multiplexing and single-neuron computations in lateral intraparietal area during decision-making. J. Neurosci. 33, 2254–2267 (2013).

    Article  CAS  Google Scholar 

  32. Newsome, W.T. & Pare, E.B. A selective impairment of motion perception following lesions of the middle temporal visual area (MT). J. Neurosci. 8, 2201–2211 (1988).

    Article  CAS  Google Scholar 

  33. Roitman, J.D. & Shadlen, M.N. Response of neurons in the lateral intraparietal area during a combined visual discrimination reaction time task. J. Neurosci. 22, 9475 (2002).

    Article  CAS  Google Scholar 

  34. Sugrue, L.P., Corrado, G.S. & Newsome, W.T. Matching behavior and the representation of value in the parietal cortex. Science 304, 1782–1787 (2004).

    Article  CAS  Google Scholar 

  35. Foley, N.C., Jangraw, D.C., Peck, C. & Gottlieb, J. Novelty enhances visual salience independently of reward in the parietal lobe. J. Neurosci. 34, 7947–7957 (2014).

    Article  CAS  Google Scholar 

  36. Premereur, E., Vanduffel, W. & Janssen, P. Functional heterogeneity of macaque lateral intraparietal neurons. J. Neurosci. 31, 12307–12317 (2011).

    Article  CAS  Google Scholar 

  37. Ma, W.J., Beck, J.M., Latham, P.E. & Pouget, A. Bayesian inference with probabilistic population codes. Nat. Neurosci. 9, 1432–1438 (2006).

    Article  CAS  Google Scholar 

  38. Jazayeri, M. & Movshon, J.A. Optimal representation of sensory information by neural populations. Nat. Neurosci. 9, 690–696 (2006).

    Article  CAS  Google Scholar 

  39. Britten, K.H., Newsome, W.T., Shadlen, M.N., Celebrini, S. & Movshon, J.A. A relationship between behavioral choice and the visual responses of neurons in macaque MT. Vis. Neurosci. 13, 87–100 (1996).

    Article  CAS  Google Scholar 

  40. Steenrod, S.C., Phillips, M.H. & Goldberg, M.E. The lateral intraparietal area codes the location of saccade targets and not the dimension of the saccades that will be made to acquire them. J. Neurophysiol. 109, 2596–2605 (2013).

    Article  Google Scholar 

  41. Brunton, B.W., Botvinick, M.M. & Brody, C.D. Rats and humans can optimally accumulate evidence for decision-making. Science 340, 95–98 (2013).

    Article  CAS  Google Scholar 

  42. Gottlieb, J. & Balan, P. Attention as a decision in information space. Trends Cogn. Sci. 14, 240–248 (2010).

    Article  Google Scholar 

  43. Bollimunta, A., Totten, D. & Ditterich, J. Neural dynamics of choice: single-trial analysis of decision-related activity in parietal cortex. J. Neurosci. 32, 12684–12701 (2012).

    Article  CAS  Google Scholar 

  44. Schall, J.D., Purcell, B.A., Heitz, R.P., Logan, G.D. & Palmeri, T.J. Neural mechanisms of saccade target selection: gated accumulator model of the visual-motor cascade. Eur. J. Neurosci. 33, 1991–2002 (2011).

    Article  Google Scholar 

  45. Purcell, B.A., Schall, J.D., Logan, G.D. & Palmeri, T.J. From salience to saccades: multiple-alternative gated stochastic accumulator model of visual search. J. Neurosci. 32, 3433–3446 (2012).

    Article  CAS  Google Scholar 

  46. Gold, J.I. & Shadlen, M.N. The neural basis of decision making. Annu. Rev. Neurosci. 30, 535–574 (2007).

    Article  CAS  Google Scholar 

  47. Churchland, A.K. et al. Variance as a signature of neural computations during decision making. Neuron 69, 818–831 (2011).

    Article  CAS  Google Scholar 

  48. Leathers, M.L. & Olson, C.R. In monkeys making value-based decisions, LIP neurons encode cue salience and not action value. Science 338, 132–135 (2012).

    Article  CAS  Google Scholar 

  49. Newsome, W.T., Glimcher, P.W., Gottlieb, J., Lee, D. & Platt, M.L. Comment on “In monkeys making value-based decisions, LIP neurons encode cue salience and not action value”. Science 340, 430 (2013).

    Article  CAS  Google Scholar 

  50. Leathers, M.L. & Olson, C.R. Response to comment on “In monkeys making value-based decisions, LIP neurons encode cue salience and not action value”. Science 340, 430 (2013).

    Article  CAS  Google Scholar 

  51. Cunningham, J.P., Yu, B.M., Shenoy, K.V. & Sahani, M. Inferring neural firing rates from spike trains using Gaussian processes. Adv. Neural Inf. Process. Syst. 20, 329–336 (2007).

    Google Scholar 

Download references

Acknowledgements

We thank J. Yates, K. Latimer and L. Cormack for discussions, K. Eastman for assistance with data collection, and C. Rorex and S. Winston for technical support. This project was supported by grants from the National Eye Institute (EY017366 to A.C.H.) and the National Institute of Mental Health (MH099611 to J.W.P. and A.C.H.), by the Sloan Foundation (J.W.P.), McKnight Foundation (J.W.P.), and a National Science Foundation CAREER award (IIS-1150186 to J.W.P.).

Author information

Authors and Affiliations

  1. Center for Perceptual Systems, The University of Texas at Austin, Austin, Texas, USA

    Il Memming Park, Alexander C Huk & Jonathan W Pillow

  2. Department of Psychology, The University of Texas at Austin, Austin, Texas, USA

    Il Memming Park, Alexander C Huk & Jonathan W Pillow

  3. Institute for Neuroscience, The University of Texas at Austin, Austin, Texas, USA

    Il Memming Park, Miriam L R Meister, Alexander C Huk & Jonathan W Pillow

  4. Department of Physiology and Biophysics, University of Washington, Seattle, Washington, USA

    Miriam L R Meister

  5. Department of Neuroscience, The University of Texas at Austin, Austin, Texas, USA

    Alexander C Huk

  6. Department of Statistics and Data Science, The University of Texas at Austin, Austin, Texas, USA

    Jonathan W Pillow

Authors
  1. Il Memming Park
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Miriam L R Meister
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Alexander C Huk
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Jonathan W Pillow
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

J.W.P. and A.C.H. conceived the study. M.L.R.M. and A.C.H. designed the experiments. M.L.R.M. performed the experiments. I.M.P., J.W.P. and A.C.H. designed the analysis. I.M.P. performed the analysis. A.C.H., I.M.P., M.L.R.M. and J.W.P. wrote the manuscript.

Corresponding author

Correspondence to Jonathan W Pillow.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Integrated supplementary information

Supplementary Figure 1 Rate model captures variance in spike counts.

(a) Peri-stimulus time variance (PSTV) histogram (thin line), and PSTH (think line) plotted simultaneously. These are targets-ON trials conditioned on coherence and decision. Both traces are computed with 50 ms boxcar moving average and smoothed with a Gaussian of 50 ms standard deviation. Note that the estimated mean and variance are close to each other, consistent with Poisson variability. (b) Population summary of PSTV power explained analogous to Fig. 3b. On average, 84.1% of the power is captured by the rate model.

Supplementary Figure 2 Post-spike filter corrects for non-Poisson interval statistics.

We use the time-rescaling theorem to transform the observed spike trains and check if the transformed interval statistics resulting from post-spike filter are consistent with the Poisson assumption. (a) Quantile-quantile plot of time-rescaled interval distribution. Diagonal represents Poisson process model assumptions. (b) Subsamples of adjacent time-rescaled inter-spike intervals. Note the uniformity of the intervals resulting from the model with the post-spike filter. The post-spike filter decorrelates the consecutive intervals, which is consistent with the Poisson assumption, indicated by the reduction of correlation coefficient (R) values.

Supplementary Figure 3 Empirical quantification of multiplicative effects.

We compare an alternative nonlinearity with (soft rectification/threshold linear function). Each column represents a neuron corresponding to Fig. 5. (a) Exponential function (black) and empirical predicted nonlinearity on test-data (red). Empirical nonlinearities are estimated by first partitioning the net linear output for each time bin into eight equal-quantile buckets, then estimating the expected spike count conditioned on each bucket. The expected means are plotted at the mean net linear drive within each bucket. (b) Threshold linear function (black) and empirical predicted nonlinearity on test-data (red). Note the systematic deviation towards an exponential nonlinearity. (c) To check for multiplicative interactions, we divided the net linear input into two components, one from saccade-locked kernels and the rest. Then, we compared model prediction to that expected from multiplicative interaction between those two components. Each colored trace represents conditioning on different amounts of contribution from the saccade-locked kernels. Solid line is the true multiplicative effect predicted from the model (with exponential nonlinearity), while dotted lines are empirical rate prediction from the data.

Supplementary Figure 4 Nonlinearity comparison with cross-validated log-likelihoods for exponential and linear rectification.

Positive cross-validated log-likelihood difference implies exponential model better fits the data. We used Fano factor as a measure of neural variability, to test for a relation between variability and the quality of the fit with an exponential nonlinearity. There is a noticeable trend (correlation coefficient -0.44) indicating that underdispersed neurons (Fano factor less than 1) are better modeled with an exponential nonlinearity, while overdispersed neurons are often better modeled with linear rectification.

Supplementary Figure 5 Comparing temporal extent of post-spike filters.

Majority of neurons showed a long self-exciting tail in the post-spike filter (Supplementary Fig. 7). We quantify how much information the tail contributes to, and test if there is a trial length scale modulation of gain. The post-spike filter used throughout the paper consists of ten 1 ms bins to model fast response components, in addition to ten raised cosine bases stretched logarithmically over 250 ms; here we also consider a model with shorter post-spike filter which has 7 temporal bases over 100 ms, and a model with longer post-spike filter which has 25 temporal bases over 5 s, which is long enough to cover the entire trial. (a) Resulting fits for 3 example cells, comparing short/medium/long spans for post-spike filter basis functions. Inset shows 5 seconds scale. (b) Spike prediction accuracy comparison across 3 models averaged over 80 neurons. Errorbar indicates standard error. Our "original" model used throughout the paper explained 7.9% more information per spike compared to short filter model (p < 10−10), and the long filter model explained 0.7% more information (p < 0.001). Thus, longer post-spike effect is negligible, and there is no systematic trial time scale gain modulation.

Supplementary Figure 6 Visualization of all kernels.

(Left) Kernels corresponding to each event. Thinner line shows the corresponding weight for the model with post-spike filter. Note that Fig. 2 is fit with a rank-2 constraint on the motion coherence kernels. Here we show the unconstrained kernel in the middle panel. (Right) Post-spike filter (blue), which is represented as a sum of weighted temporal basis functions (gray).

Supplementary Figure 7 Population diversity in all kernels.

Each column represents a different kernel. The motion coherence kernels are convolved with a 500 ms boxcar for easier interpretation. (Top 9 rows) Each row shows one of the 9 example cells. They are sorted by decoding performance: The top cell is the best performing, followed by example neurons performing at 11%, 22%, …, 89% percentile. (Bottom row) Population average for all 80 neurons. Average kernel was transformed with exponential nonlinearity to represent gain. These fits are from models with the post-spike filter.

Supplementary Figure 8 Single trial prediction of the model with and without post-spike filter.

In Fig. 4, we only showed best and median predicted trials. Here are randomly selected trials per coherence strength for which the monkey was correct. Each column represents a cell corresponding to Fig. 5 and each row corresponds to a coherence level. Light blue lines are predictions from the model with the post-spike filter. Although the fine time scale predictions are significantly better with the post-spike filter, we avoided showing the post-spike filter prediction in Fig 4, because it is difficult to visually assess the quality of fit. This is because a typical sharp self-excitatory gain appears after each spike, which also makes it difficult to smooth without inducing deceiving results.

Supplementary Figure 9 Dimensionality of saccade kernels compared to the decoding kernel.

(Top row) Individual traces of the kernel shapes of the population. Difference corresponds to Fig. 7a. (Bottom row) Dimensionality analysis corresponding to Supplementary Table 1. The first few dimensions explain most of the variance. Note that the difference of IN and OUT kernel has the tightest subspace.

Supplementary Figure 10 Visualization of temporal basis functions for the kernels.

(Top) Raised cosine basis functions spaced in 50 ms steps. (Bottom) 10 single bin boxcars followed by log-time scaled raised cosine basis functions. This is only used for the post-spike filter.

Supplementary information

Supplementary Text and Figures

Supplementary Figures 1–10 and Supplementary Table 1 (PDF 3156 kb)

Supplementary Methods Checklist (PDF 328 kb)

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Park, I., Meister, M., Huk, A. et al. Encoding and decoding in parietal cortex during sensorimotor decision-making. Nat Neurosci 17, 1395–1403 (2014). https://doi.org/10.1038/nn.3800

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/nn.3800

This article is cited by

Search

Advanced search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing