Article | Published:

Elucidating the underlying components of food valuation in the human orbitofrontal cortex

Nature Neurosciencevolume 20pages17801786 (2017) | Download Citation

Subjects

Abstract

The valuation of food is a fundamental component of our decision-making. Yet little is known about how value signals for food and other rewards are constructed by the brain. Using a food-based decision task in human participants, we found that subjective values can be predicted from beliefs about constituent nutritive attributes of food: protein, fat, carbohydrates and vitamin content. Multivariate analyses of functional MRI data demonstrated that, while food value is represented in patterns of neural activity in both medial and lateral parts of the orbitofrontal cortex (OFC), only the lateral OFC represents the elemental nutritive attributes. Effective connectivity analyses further indicate that information about the nutritive attributes represented in the lateral OFC is integrated within the medial OFC to compute an overall value. These findings provide a mechanistic account for the construction of food value from its constituent nutrients.

Access optionsAccess options

Rent or Buy article

Get time limited or full article access on ReadCube.

from $8.99

All prices are NET prices.

Additional information

Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

  1. 1.

    Clithero, J. A. & Rangel, A. Informatic parcellation of the network involved in the computation of subjective value. Soc. Cogn. Affect. Neurosci. 9, 1289–1302 (2014).

  2. 2.

    Padoa-Schioppa, C. & Assad, J. A. Neurons in the orbitofrontal cortex encode economic value. Nature 441, 223–226 (2006).

  3. 3.

    Rich, E. L. & Wallis, J. D. Decoding subjective decisions from orbitofrontal cortex. Nat. Neurosci. 19, 973–980 (2016).

  4. 4.

    Rudebeck, P. H. & Murray, E. A. The orbitofrontal oracle: cortical mechanisms for the prediction and evaluation of specific behavioral outcomes. Neuron 84, 1143–1156 (2014).

  5. 5.

    Grabenhorst, F. & Rolls, E. T. Value, pleasure and choice in the ventral prefrontal cortex. Trends Cogn. Sci. 15, 56–67 (2011).

  6. 6.

    McNamee, D., Rangel, A. & O’Doherty, J. P. Category-dependent and category-independent goal-value codes in human ventromedial prefrontal cortex. Nat. Neurosci. 16, 479–485 (2013).

  7. 7.

    Chikazoe, J., Lee, D. H., Kriegeskorte, N. & Anderson, A. K. Population coding of affect across stimuli, modalities and individuals. Nat. Neurosci. 17, 1114–1122 (2014).

  8. 8.

    Howard, J. D., Gottfried, J. A., Tobler, P. N. & Kahnt, T. Identity-specific coding of future rewards in the human orbitofrontal cortex. Proc. Natl. Acad. Sci. USA 112, 5195–5200 (2015).

  9. 9.

    Lebreton, M., Jorge, S., Michel, V., Thirion, B. & Pessiglione, M. An automatic valuation system in the human brain: evidence from functional neuroimaging. Neuron 64, 431–439 (2009).

  10. 10.

    Small, D. M. et al. Dissociation of neural representation of intensity and affective valuation in human gustation. Neuron 39, 701–711 (2003).

  11. 11.

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

  12. 12.

    Stalnaker, T. A. et al. Orbitofrontal neurons infer the value and identity of predicted outcomes. Nat. Commun 5, 3926 (2014).

  13. 13.

    Gross, J. et al. Value signals in the prefrontal cortex predict individual preferences across reward categories. J. Neurosci. 34, 7580–7586 (2014).

  14. 14.

    Chib, V. S., Rangel, A., Shimojo, S. & O’Doherty, J. P. Evidence for a common representation of decision values for dissimilar goods in human ventromedial prefrontal cortex. J. Neurosci. 29, 12315–12320 (2009).

  15. 15.

    Levy, D. J. & Glimcher, P. W. Comparing apples and oranges: using reward-specific and reward-general subjective value representation in the brain. J. Neurosci. 31, 14693–14707 (2011).

  16. 16.

    Suzuki, S. et al. Learning to simulate others’ decisions. Neuron 74, 1125–1137 (2012).

  17. 17.

    Suzuki, S., Adachi, R., Dunne, S., Bossaerts, P. & O’Doherty, J. P. Neural mechanisms underlying human consensus decision-making. Neuron 86, 591–602 (2015).

  18. 18.

    Foerde, K., Steinglass, J. E., Shohamy, D. & Walsh, B. T. Neural mechanisms supporting maladaptive food choices in anorexia nervosa. Nat. Neurosci. 18, 1571–1573 (2015).

  19. 19.

    Carnell, S., Gibson, C., Benson, L., Ochner, C. N. & Geliebter, A. Neuroimaging and obesity: current knowledge and future directions. Obes. Rev. 13, 43–56 (2012).

  20. 20.

    Barron, H. C., Dolan, R. J. & Behrens, T. E. J. Online evaluation of novel choices by simultaneous representation of multiple memories. Nat. Neurosci. 16, 1492–1498 (2013).

  21. 21.

    Klein-Flügge, M. C., Barron, H. C., Brodersen, K. H., Dolan, R. J. & Behrens, T. E. J. Segregated encoding of reward-identity and stimulus-reward associations in human orbitofrontal cortex. J. Neurosci. 33, 3202–3211 (2013).

  22. 22.

    Ongür, D. & Price, J. L. The organization of networks within the orbital and medial prefrontal cortex of rats, monkeys and humans. Cereb. Cortex 10, 206–219 (2000).

  23. 23.

    Tang, D. W., Fellows, L. K. & Dagher, A. Behavioral and neural valuation of foods is driven by implicit knowledge of caloric content. Psychol. Sci. 25, 2168–2176 (2014).

  24. 24.

    Zuker, C. S. Food for the brain. Cell 161, 9–11 (2015).

  25. 25.

    de Araujo, I. E. et al. Food reward in the absence of taste receptor signaling. Neuron 57, 930–941 (2008).

  26. 26.

    Tellez, L. A. et al. Separate circuitries encode the hedonic and nutritional values of sugar. Nat. Neurosci. 19, 465–470 (2016).

  27. 27.

    Haynes, J.-D. A primer on pattern-based approaches to fMRI: principles, pitfalls, and perspectives. Neuron 87, 257–270 (2015).

  28. 28.

    Nichols, T., Brett, M., Andersson, J., Wager, T. & Poline, J.-B. Valid conjunction inference with the minimum statistic. Neuroimage 25, 653–660 (2005).

  29. 29.

    Kriegeskorte, N. & Kievit, R. A. Representational geometry: integrating cognition, computation, and the brain. Trends Cogn. Sci. 17, 401–412 (2013).

  30. 30.

    Efron, B. & Tibshirani, R. J. An Introduction to the Bootstrap (CRC Press, 1993).

  31. 31.

    Vickery, T. J., Chun, M. M. & Lee, D. Ubiquity and specificity of reinforcement signals throughout the human brain. Neuron 72, 166–177 (2011).

  32. 32.

    Kahnt, T., Park, S. Q., Haynes, J.-D. & Tobler, P. N. Disentangling neural representations of value and salience in the human brain. Proc. Natl. Acad. Sci. USA 111, 5000–5005 (2014).

  33. 33.

    Gottfried, J. A., O’Doherty, J. & Dolan, R. J. Encoding predictive reward value in human amygdala and orbitofrontal cortex. Science 301, 1104–1107 (2003).

  34. 34.

    Mishkin, M., Ungerleider, L. G. & Macko, K. A. Object vision and spatial vision: two cortical pathways. Trends Neurosci. 6, 414–417 (1983).

  35. 35.

    Howard, J. D. & Kahnt, T. Identity-specific reward representations in orbitofrontal cortex are modulated by selective devaluation. J. Neurosci. 37, 2627–2638 (2017).

  36. 36.

    Noonan, M. P. et al. Separate value comparison and learning mechanisms in macaque medial and lateral orbitofrontal cortex. Proc. Natl. Acad. Sci. USA 107, 20547–20552 (2010).

  37. 37.

    Rozin, P. & Vollmecke, T. A. Food likes and dislikes. Annu. Rev. Nutr. 6, 433–456 (1986).

  38. 38.

    Hare, T. A., Malmaud, J. & Rangel, A. Focusing attention on the health aspects of foods changes value signals in vmPFC and improves dietary choice. J. Neurosci. 31, 11077–11087 (2011).

  39. 39.

    Becker, G. M., DeGroot, M. H. & Marschak, J. Measuring utility by a single-response sequential method. Behav. Sci. 9, 226–232 (1964).

  40. 40.

    Deichmann, R., Gottfried, J. A., Hutton, C. & Turner, R. Optimized EPI for fMRI studies of the orbitofrontal cortex. Neuroimage 19, 430–441 (2003).

  41. 41.

    Hebart, M. N., Görgen, K. & Haynes, J.-D. The Decoding Toolbox (TDT): a versatile software package for multivariate analyses of functional imaging data. Front. Neuroinform. 8, 88 (2015).

  42. 42.

    Tzourio-Mazoyer, N. et al. Automated anatomical labeling of activations in SPM using a macroscopic anatomical parcellation of the MNI MRI single-subject brain. Neuroimage 15, 273–289 (2002).

  43. 43.

    Allefeld, C., Görgen, K. & Haynes, J.-D. Valid population inference for information-based imaging: From the second-level t-test to prevalence inference. Neuroimage 141, 378–392 (2016).

  44. 44.

    Kriegeskorte, N., Goebel, R. & Bandettini, P. Information-based functional brain mapping. Proc. Natl. Acad. Sci. USA 103, 3863–3868 (2006).

  45. 45.

    McNamee, D., Liljeholm, M., Zika, O. & O’Doherty, J. P. Characterizing the associative content of brain structures involved in habitual and goal-directed actions in humans: a multivariate fMRI study. J. Neurosci. 35, 3764–3771 (2015).

  46. 46.

    Kriegeskorte, N., Mur, M. & Bandettini, P. Representational similarity analysis - connecting the branches of systems neuroscience. Front. Syst. Neurosci. 2, 4 (2008).

  47. 47.

    Friston, K.J., Ashburner, J.T., Kiebel, S.J., Nichols, T.E. & Penny, W.D. Statistical Parametric Mapping: the Analysis of Functional Brain Images (Academic Press, 2006).

Download references

Acknowledgements

This work was supported by the JSPS Postdoctoral Fellowship for Research Abroad (S.S.), JSPS KAKENHI Grants JP17H05933 and JP17H06022 (S.S.) and the NIMH Caltech Conte Center for the Neurobiology of Social Decision Making (J.P.O.).

Author information

Affiliations

  1. Division of the Humanities and Social Sciences, California Institute of Technology, Pasadena, CA, USA

    • Shinsuke Suzuki
    •  & John P. O’Doherty
  2. Frontier Research Institute for Interdisciplinary Sciences, Tohoku University, Sendai, Japan

    • Shinsuke Suzuki
  3. Institute of Development, Aging and Cancer, Tohoku University, Sendai, Japan

    • Shinsuke Suzuki
  4. Computation and Neural Systems, California Institute of Technology, Pasadena, CA, USA

    • Logan Cross
    •  & John P. O’Doherty

Authors

  1. Search for Shinsuke Suzuki in:

  2. Search for Logan Cross in:

  3. Search for John P. O’Doherty in:

Contributions

S.S., L.C. and J.P.O. designed the research; S.S. and L.C. carried out the experiment; S.S. and L.C. analyzed the data; and S.S., L.C. and J.P.O. wrote the paper.

Competing interests

The authors declare no competing financial interest.

Corresponding author

Correspondence to Shinsuke Suzuki.

Integrated supplementary information

  1. Supplementary Figure 1 Ratings of nutrient factor.

    (a) Subjective ratings about the 56 food items. For each item, we plot the participants’ ratings about the six nutrient factors (cyan: fat; magenta: sodium; black: carbohydrate; red: sugar; green: protein; and blue: vitamin). See Table S2 for the item list. The rating data were z-normalized across the food items, within each participant and each nutrient factor. (b) Pair-wise correlations among subjective ratings of the nutrient factors (MEAN across participants; n = 23).

  2. Supplementary Figure 2 Supplementary results of the neural representation of subjective value.

    (a) Anatomical OFC ROIs used in this study. The ROIs are defined based on the AAL database42 as follows: lOFC, bilateral MNI_Frontal_Mid_Orb + MNI_Frontal_Inf_Orb + MNI_Frontal_Sup_Orb; and mOFC, bilateral MNI_Frontal_Med_Orb. lOFC, lateral orbitofrontal cortex; and mOFC, medial orbitofrontal cortex. (b) Evidence for significant decoding of subjective value at the Bid phase (left) and at the Feedback phase (right) (n = 23 participants). The format is the same as in Fig. 2a (left). **P < 0.01 and *P < 0.05, t-test against 50% (Bid phase, lOFC: t 22 = 2.78, P = 0.005; mOFC: t 22 = 1.96, P = 0.031; and Feedback phase, lOFC: t 22 = 2.23, P = 0.018; mOFC: t 22 = 3.40, P = 0.001). (c) Weights of voxels in the value classifiers obtained from the ROI analyses (see Fig. 2a). We plot the weights of the voxels for each participant within the lOFC (left) and the mOFC (right) ROIs separately. Format of the box and whisker plots is the same as in Fig. 1c. (d) Weights of voxels in the value classifiers obtained from the searchlight analyses (see Fig. 2b). We plot the weights of the voxels within a radius of 3 voxels (i.e., 9 mm) around the peak voxels in lOFC (left) and mOFC (right). See Fig. 2b for information about the peak voxels. Format of the box and whisker plots is the same as in Fig. 1c.

  3. Supplementary Figure 3 Classification scores in the decoding analysis for subjective nutrient factors.

    We plot the classification scores in the lOFC ROI obtained by the classifier trained on fat, carb., protein and vitamin respectively as functions of the subjective nutrititive ratings (MEAN ± SEM across participants; n = 23; see Fig. 3a). Note that ratings for each nutrient factors were binned based on the rank order; that each classifier is trained to discriminate high vs. low ratings (i.e., 1 & 2 vs. 3 & 4); and that the classification weights of each voxel were estimated on a subset of the data and the classification scores were computed on the other subset of the data (i.e., leave-one-run-out cross-validation; see Methods for details). lOFC, lateral orbitofrontal cortex; and Carb., carbohydrade.

  4. Supplementary Figure 4 Supplementary results of the neural representation of subjective nutrient factors.

    (a) Decoding accuracies of subjective nutrient factors at the time of bidding revealing a lack of significant decoding at this time-point (n = 23 participants). The format is the same as in Fig. 3a (left). Left. t-test against 50% (fat: t 22 = 1.38, P = 0.091; carb.: t 22 = -2.47, P = 0.989; protein: t 22 = 1.11, P = 0.139; and vitamin: t 22 = 0.48, P = 0.320). Right. t-test (fat: t 22 = -0.42, P = 0.660; carb.: t 22 = 0.14, P = 0.444; protein: t 22 = 1.26, P = 0.110; and vitamin: t 22 = 0.42, P = 0.339). lOFC, lateral orbitofrontal cortex; mOFC, medial orbitofrontal cortex; and Carb., carbohydrate. (b) Decoding accuracies of subjective nutrient factors at the time of feedback revealing little evidence for significant decoding at this time-point (n = 23 participants). The format is the same as in Fig. 3a (left). Left. t-test against 50% (fat: t 22 = 0.72, P = 0.239; carb.: t 22 = -0.43, P = 0.664; protein: t 22 = 1.38, P = 0.090; and vitamin: t 22 = 0.18, P = 0.431). Right. t-test (fat: t 22 = 0.90, P = 0.190; carb.: t 22 = -0.80, P = 0.783; protein: t 22 = 1.07, P = 0.149; and vitamin: t 22 = 1.01, P = 0.162). (c) Weights of voxels in the classifiers for each of the subjective nutrient factors obtained from the ROI analyses (see Fig. 3a). We plot the weights of the voxels for each participant within the lOFC ROI. Format of the box and whisker plots is the same as in Fig. 1c. (d) Weights of voxels in the classifiers for each of the subjective nutrient factors obtained from the search analyses (see Fig. 3c). We plot the weights of the voxels within a radius of 3 voxels (i.e., 9 mm) around the peak voxels in lOFC. See Fig. 3c for information about the peak voxels. Format of the box and whisker plots is the same as in Fig. 1c. (e) Decoding accuracies of the subjective nutrient factors for novel food items (n = 23 participants). The format is the same as in Fig. 3a (left). + P < 0.10, *P < 0.05 and **P < 0.01 for each factor, t-test against 50% (fat: t 22 = 2.42, P = 0.012; carb.: t 22 = 1.90, P = 0.035; protein: t 22 = 1.41, P = 0.087; and vitamin: t 22 = 1.84, P = 0.039). (f) Decoding accuracies of the subjective nutrient factors in the reduced lOFC ROIs (n = 23 participants). In this analysis, (i) we randomly re-sampled adjacent 533 voxels from the lOFC ROI (i.e., forming a continuous cluster consisting of the 553 voxels); then (ii) we tested if information about the subjective nutrient factors could be decoded from the re-sampled voxels; and (iii) the above procedure was repeated 100 times (the decoding accuracies were averaged). The format is the same as in Fig. 3a (left). *P < 0.05 and **P < 0.01 for each factor, t-test against 50% (fat: t 22 = 2.45, P = 0.011; carb.: t 22 = 1.81, P = 0.042; protein: t 22 = 2.59, P = 0.008; and vitamin: t 22 = 2.46, P = 0.011). (g) Pair-wise correlations among the classifiers’ weights for the four nutrient factors (MEAN across participants; n = 23). For each pair of the nutrient factors, we obtained the correlation coefficient in the classification weights of the voxels within the lOFC ROI. (h) Decoding accuracies in the cross-decoding analyses (n = 23 participants). Format of the box and whisker plots is the same as in Fig. 1c. Two nutrient factors in each parenthesis denote the pair used for the cross-decoding. **P < 0.01, two-tailed t-test against 50% ([fat, carb.]: t 22 = 1.59, P = 0.127; [fat, protein]: t 22 = -0.35, P = 0.729; [fat, vitamin]: t 22 = -3.18, P = 0.004; [carb., protein]: t 22 = -1.96, P = 0.062; [carb., vitamin]: t 22 = -1.06, P = 0.299; and [protein, vitamin]: t 22 = 0.84, P = 0.408). (i) Decoding accuracies on the re-sampled food items (see the main text; n = 23 participants). Format of the box and whisker plots is the same as in Fig. 1c. Left, accuracy of fat and vitamin (one classifier was trained and tested on fat; the other classifier was on vitamin; and the accuracy scores were averaged). Right, accuracy in the cross-decoding analysis between fat and vitamin. Two nutrient factors in the parenthesis denote the pair used for the cross-decoding. That is, we trained a classifier on one factor and tested it on the other factor (and the reverse; and the decoding accuracy was assessed by the average across both directions). *P < 0.05, two-tailed t-test against 50% ([fat, fat] & [vitamin, vitamin]: t 22 = 2.10, P = 0.048; and [fat, vitamin]: t 22 = -1.10, P = 0.282). (j) Significant decoding of sugar but not sodium content in lOFC (n = 23 participants). The accuracies are plotted for the lOFC ROI. Format of the box and whisker plots is the same as in Fig. 1c. **P < 0.01, t-test (Sodium: t 22 = 0.06, P = 0.474; and Sugar: t 22 = 2.67, P = 0.007). (k) Neither sodium nor sugar content was significantly decodable in mOFC (n = 23 participants). t-test against 50% (Sodium: t 22 = -0.15, P = 0.557; and Sugar: t 22 = 1.34, P = 0.100). mOFC, medial orbitofrontal cortex. Format of the box and whisker plots is the same as in Fig. 1c. (l) Decoding accuracies of objective nutrient factors at the time of valuation, demonstrating relatively weak effects of objective nutrient factors (n = 23 participants). The format is the same as in Fig. 3a (left). Left. *P < 0.05, t-test against 50% (fat: t 22 = -0.33, P = 0.626; carb.: t 22 = 1.58, P = 0.064; protein: t 22 = 2.05, P = 0.026; and vitamin: t 22 = 2.26, P = 0.017). Right. t-test (fat: t 22 = -3.10, P = 0.997; carb.: t 22 = 0.29, P = 0.387; protein: t 22 = 0.78, P = 0.222; and vitamin: t 22 = 0.08, P = 0.469).

  5. Supplementary Figure 5 Procedure and results of the representational similarity analysis (RSA).

    (a) Procedure for construction of the voxel-wise Representational Dissimilarity Matrix (RDM). The voxel-wise RDM is created based on the correlation across the voxels’ activities for each pair of the items. See Methods for details. Corr., Pearson’s correlation coefficient. (b) Procedure for construction of the behavioral RDM. The behavioral RDM is created based on the correlation in bundles of the four subjective nutrient factors for each item pair. See Methods for details. (c) Results of the ROI analyses. Spearman’s rank correlation (z-transformed) between the voxel-wise neural and the behavioral RDMs is plotted for the lOFC and the mOFC ROIs (n = 23). Format of the box and whisker plots is the same as in Fig. 1c. **P < 0.01, t-test (lOFC: t 22 = 2.85, P = 0.005; and mOFC: t 22 = 1.13, P = 0.135). lOFC, lateral orbitofrontal cortex; and mOFC, medial orbitofrontal cortex. (d) Results of the searchlight analysis. The RSA correlation map is thresholded at P < 0.005 (uncorrected) for display purposes, generated by performing a t-test (n = 23 participants). Peak voxels, [MNI: x, y, z = 12, 23, -23] and [-21 38 -23] (P < 0.05 small-volume corrected) for right and left OFC respectively. OFC, orbitofrontal cortex. (e) Pattern of fMRI response to each of the 56 food items in a space of the pair-wise correlation across voxels’ activities in the lOFC. We plot the voxel-wise neural RDM averaged over the participants (top left, n = 23). To visualize the approximate geometric structure, we also show the same data as a two-dimensional MDS plot (top center) and a dendrogram plot obtained by an agglomerative hierarchical clustering (bottom right). In the MDS plot, the digits depict the food items’ ID. See Table S2 for detailed information about the food items. MDS, multi dimensional scaling.

  6. Supplementary Figure 6 Effective connectivity between OFC subregions at the time of bidding and the time of feedback.

    (a) Results of an effective connectivity analysis at the time of bidding. Effect sizes of the PPI regressors are plotted (n = 23 participants). The format is the same as in Fig. 5ab. **P < 0.01 for each factor. Left. t-test (fat: t 22 = 1.37, P = 0.092; carb.: t 22 = 1.28, P = 0.108; protein: t 22 = 3.20, P = 0.002; and vitamin: t 22 = 1.40, P = 0.088). Right. t-test (fat: t 22 = 1.59, P = 0.063; carb.: t 22 = 1.61, P = 0.060; protein: t 22 = 1.54, P = 0.068; and vitamin: t 22 = 1.33, P = 0.099). lOFC, lateral orbitofrontal cortex; mOFC, medial orbitofrontal cortex; Carb., carbohydrate; and PPI, psychophysiological interaction. (b) Results of an effective connectivity analysis at the time of feedback. Effect sizes of the PPI regressors are plotted (n = 23 participants). The format is the same as in Fig. 5ab. **P < 0.01 for each factor. Left. t-test (fat: t 22 = 1.30, P = 0.104; carb.: t 22 = 1.67, P = 0.055; protein: t 22 = 1.22, P = 0.118; and vitamin: t 22 = 1.02, P = 0.159). Right. t-test (fat: t 22 = 1.64, P = 0.058; carb.: t 22 = 1.32, P = 0.100; protein: t 22 = 2.80, P = 0.005; and vitamin: t 22 = 1.01, P = 0.161).

  7. Supplementary Figure 7 Decoding of subjective value and nutrient factors in other brain regions.

    (a) Anatomical ROIs used in the additional post hoc analyses. The ROIs are defined based on the AAL database42 as follows: dmPFC, bilateral MNI_Frontal_Sup_Medial + MNI_Cingulum_Ant; dlPFC, bilateral MNI_Frontal_Mid + MNI_Frontal_Sup; vlPFC, bilateral MNI_Frontal_Inf_Oper + MNI_Frontal_Inf_Tri; PPC, bilateral MNI_Parietal_Inf + MNI_Parietal_Sup; Insula, bilateral MNI_Insula; and Amygdala, bilateral MNI_Amygdala. dmPFC, dorsomedial prefrontal cortex; dlPFC, dorsolateral prefrontal cortex; vlPFC, ventrolateral prefrontal cortex; and PPC, posterior parietal cortex. (b) Decoding accuracies of subjective value across the ROIs (n = 23 participants). The format is the same as in Fig. 2a (left). **P < 0.01 for each region, t-test against 50% (dmPFC: t 22 = 4.55, P < 0.001; dlPFC: t 22 = 7.30, P < 0.001; vlPFC: t 22 = 4.39, P < 0.001; PPC: t 22 = 6.52, P < 0.001; Insula: t 22 = 3.48, P = 0.001; and Amygdala: t 22 = 2.92, P = 0.004). (c) Decoding accuracies of subjective nutrient factors (n = 23 participants). The format is the same as in Fig. 3a (left). *P < 0.05 and **P < 0.01 for each factor. Top left. t-test against 50% (fat: t 22 = 1.57, P = 0.066; carb.: t 22 = 1.12, P = 0.137; protein: t 22 = 0.98, P = 0.168; and vitamin: t 22 = 1.71, P = 0.050). Top middle. t-test (fat: t 22 = 2.25, P = 0.018; carb.: t 22 = 1.26, P = 0.111; protein: t 22 = 1.52, P = 0.071; and vitamin: t 22 = 2.57, P = 0.009). Top right. t-test (fat: t 22 = 0.51, P = 0.307; carb.: t 22 = 0.52, P = 0.305; protein: t 22 = 2.44, P = 0.012; and vitamin: t 22 = 0.90, P = 0.189). Bottom left. t-test (fat: t 22 = 4.11, P < 0.001; carb.: t 22 = 2.37, P = 0.014; protein: t 22 = 4.46, P < 0.001; and vitamin: t 22 = 4.50, P < 0.001). Bottom middle. t-test (fat: t 22 = 0.65, P = 0.261; carb.: t 22 = 0.83, P = 0.209; protein: t 22 = 0.83, P = 0.208; and vitamin: t 22 = 0.05, P = 0.481). Bottom right. t-test (fat: t 22 = -0.79, P = 0.780; carb.: t 22 = 0.73, P = 0.236; protein: t 22 = 0.50, P = 0.312; and vitamin: t 22 = -0.23, P = 0.592). (d) Decoding accuracies of low-level visual features and comparison with subjective nutrient factors in lOFC, PPC and V1. The decoding accuracies of the low-level visual features (averaged over the eight features) and the subjective nutrient factors (averaged over the four factors identified as value predictors) are plotted for the lOFC, the PPC and the V1 (BA17) anatomical ROIs (n = 23 participants). Format of the box and whisker plots is the same as in Fig. 1c. * and ** on each plot respectively denote P < 0.05 and P < 0.01 for each factor, t-test against 50%. * and ** on the horizontal lines denote significant differences between the indicated pairs of data at P < 0.05 and P < 0.01 respectively, two-tailed paired t-test. Left. t-test (subjective nutrient factors: t 22 = 4.72, P < 0.001; low-level visual features: t 22 = 0.70, P = 0.247; and subjective nutrient factors vs. low-level visual features: t 22 = 3.18, P = 0.004). Middle. t-test (subjective nutrient factors: t 22 = 7.04, P < 0.001; low-level visual features: t 22 = 4.01, P < 0.001; and subjective nutrient factors vs. low-level visual features: t 22 = 2.62, P = 0.0157). Right. t-test (subjective nutrient factors: t 22 = 5.85, P < 0.001; low-level visual features: t 22 = 8.34, P < 0.001; and subjective nutrient factors vs. low-level visual features: t 22 = -3.15, P = 0.005). lOFC, lateral orbitofrontal cortex; PPC, posterior parietal cortex; V1, primary visual cortex; and BA17, Brodmann area 17.

  8. Supplementary Figure 8 A region of V1 in which all of the four subjective nutrient factors can be decoded.

    The decoding accuracy map obtained from the whole-brain searchlight analysis is thresholded at P < 0.05 (cluster-level FWE correction with the cluster-forming threshold P = 0.001; n = 23 participants), conjunction-test. Peak voxel, [MNI: x, y, z = -9, -94, 7]. V1, primary visual cortex.

Supplementary information

  1. Supplementary Text and Figures

    Supplementary Figures 1–8 and Supplementary Tables 1 and 2

  2. Life Sciences Reporting Summary

About this article

Publication history

Received

Accepted

Published

DOI

https://doi.org/10.1038/s41593-017-0008-x