, Volume 59, Issue 1, pp 21–47 | Cite as

The statistical analysis of general processing tree models with the EM algorithm

  • Xiangen Hu
  • William H. Batchelder


Multinomial processing tree models assume that an observed behavior category can arise from one or more processing sequences represented as branches in a tree. These models form a subclass of parametric, multinomial models, and they provide a substantively motivated alternative to loglinear models. We consider the usual case where branch probabilities are products of nonnegative integer powers in the parameters, 0≤θs≤1, and their complements, 1 - θs. A version of the EM algorithm is constructed that has very strong properties. First, the E-step and the M-step are both analytic and computationally easy; therefore, a fast PC program can be constructed for obtaining MLEs for large numbers of parameters. Second, a closed form expression for the observed Fisher information matrix is obtained for the entire class. Third, it is proved that the algorithm necessarily converges to a local maximum, and this is a stronger result than for the exponential family as a whole. Fourth, we show how the algorithm can handle quite general hypothesis tests concerning restrictions on the model parameters. Fifth, we extend the algorithm to handle the Read and Cressie power divergence family of goodness-of-fit statistics. The paper includes an example to illustrate some of these results.

Key words

EM algorithm multinomial models processing trees power divergence family 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Batchelder, W. H. (1991). Getting wise about minimum distance measures [Review ofGoodness-of-fit statistics for discrete multivariate data by T. R. C. Read & N. A. C. Cressie].Journal of Mathematical Psychology, 35, 267–273.Google Scholar
  2. Batchelder, W. H., Hu, X., & Riefer, D. M. (in press) Analysis of a model for source monitoring. In G. H. Fischer & D. Laming (Eds.),Mathematical psychology: New developments. Berlin: Springer-Verlag. (Available as Technical Report No. 92-07, Institute for Mathematical Behavior Sciences, School of Social Sciences, UC, Irvine)Google Scholar
  3. Batchelder, W. H., & Riefer, D. M. (1986). Statistical analysis of a model for storage and retrieval processes in human memory.British Journal of Mathematical and Statistical Psychology, 39, 129–149.Google Scholar
  4. Batchelder, W. H., & Riefer, D. M. (1990). Multinomial processing models of source monitoring.Psychological Review, 97, 548–564.Google Scholar
  5. Bäuml, K.-H. (1991). Experimental analysis of storage and retrieval processes involved in retroactive inhibition: The effect of presentation mode.Acta Psychologica, 77(2), 103–119.Google Scholar
  6. Bernstein, F. (1925). Zusammenfassende Betrachtungen über die erblichen Blutenstructuren des Menschen [Summarizing considerations on the inheritable blood structures of mankind].Z. Abstamm. Vererbgsl., 37, 237–270.Google Scholar
  7. Boyles, R. A. (1983). On the convergence of the EM algorithm.Journal of Royal Statistical Society, Series B, 45, 47–50.Google Scholar
  8. Ceppellini, R., Siniscalco, M., & Smith, C. A. B. (1955). The estimation of gene frequencies in random mating populations.Annals of Human Genetics, 20, 97–115.Google Scholar
  9. Chechile, R., & Meyer, D. L. (1976). A Bayesian procedure for separately estimating storage and retrieval components of forgetting.Journal of Mathematical Psychology, 13, 269–295.Google Scholar
  10. Dempster, A. P., Laird, N. M., & Rubin, D. B. (1977). Maximum likelihood from incomplete data via the EM algorithm.Journal of Royal Statistical Society, Series B, 39, 1–38.Google Scholar
  11. Efron, B., & Hinkley, D. V. (1978). The observed versus expected information.Biometrika, 65, 457–487.Google Scholar
  12. Elandt-Johnson, R. C. (1971).Probability models and statistical methods in genetics. New York, Wiley & Sons.Google Scholar
  13. Erdfelder, E., & Bayen, U. J. (1991). Episodisches Gedächtnis im Alter: Methodologische und empirische Arguments für einen Zugang über mathematische Modelle [Episodic memory in old age: Methodological and empirical arguments for an access through mathematical models]. In D. Frey (Eds.),Bericht über den 37. Kongreß der Deutschen Gesellschaft für Psychologie in Kiel 1990, Band 2 (pp. 172–180). Göttingen: Hogrefe.Google Scholar
  14. Hartley, H. O. (1958). Maximum likelihood estimation from incomplete data.Biometrics, 14, 174–194.Google Scholar
  15. Harvey, P. D. (1985). Reality monitoring in mania and schizophrenia.The Journal of Nervous and Mental Disease, 173, 67–72.Google Scholar
  16. Hu, X. (1990).Source monitoring program (Version 1.0). Irvine: University of California. (available upon request)Google Scholar
  17. Hu, X. (1991).General program for processing tree models (Version 1.0). Irvine: University of California. (available upon request)Google Scholar
  18. Humphreys, M. S., & Bain, J. D. (1983). Recognition memory: A cue and information analysis.Memory and Cognition, 11, 583–600.Google Scholar
  19. Johnson, M. K., & Raye, C. L. (1980). Reality monitoring.Psychological Review.88, 67–85.Google Scholar
  20. Landsteiner, K. (1901). Über Agglutinationserscheinungen normalen menschlichen Blutes. [On agglutination appearances of normal human blood]Wien. Klin. Wschr., 14, 1132–1134.Google Scholar
  21. Lazarsfeld, P. F., & Henry, N. W. (1968).Latent structure analysis. New York: Houghton Mifflin.Google Scholar
  22. Little, R. J. A., & Rubin, D. B. (1987).Statistical analysis with missing data. New York: Wiley.Google Scholar
  23. Louis, T. A. (1982). Finding the observed information matrix when using the EM algorithm.Journal of Royal Statistical Society, Series B, 44, 226–233.Google Scholar
  24. Meng, X. L., & Rubin, D. B. (1991). Using EM to obtain asymptotic variance-covariance matrices: The SEM algorithm.Journal of the American Statistical Association, 86, 899–909.Google Scholar
  25. Read, T. R. C., & Cressie, N. A. C. (1988).Goodness-of-fit statistics for discrete multivariate data. New York: Springer-Verlag.Google Scholar
  26. Riefer, D. M., & Batchelder, W. H. (1988). Multinomial modeling and the measurement of cognitive processes.Psychological Review, 95, 318–339.Google Scholar
  27. Riefer, D. M., & Batchelder, W. H. (1991). Statistical inference for multinomial processing tree models. In J.-P. Doignon & J.-C. Falmagne (Eds.),Mathematical psychology: Current developments (pp. 313–335). New York: Springer-Verlag.Google Scholar
  28. Riefer, D. M., & Rouder, J. M. (1992). A multinomial modeling analysis of the mnemonic benefits of bizarre imagery.Memory and Cognition, 20, 601–611.Google Scholar
  29. Rosenbloom, P. S., Laird, J. E., Newell, A., & McCarl, R. (1991). A preliminary analysis of the SOAR architecture as a basis for general intelligence.Artificial Intelligence, 47, 289–325.Google Scholar
  30. Ross, B. H., & Bower, G. H. (1981). Comparisons of models of associative recall.Memory & Cognition, 9, 1–16.Google Scholar
  31. Rubin, D. B. (1991). EM and beyond.Psychometrika, 56, 241–254.Google Scholar
  32. Rumelhart, D. E., & McClelland, J. L. (1986).Parallel distributed processing (Vol. 1). Cambridge: MIT Press.Google Scholar
  33. Ruud, P. A. (1991). Extensions of estimation methods using the EM algorithm.Journal of Econometrics, 49, 305–341.Google Scholar
  34. Smith, C. A. B. (1957). Counting methods in genetical statistics.Annals of Human Genetics, 21, 97–115.Google Scholar
  35. Wickens, T. D. (1982).Models for behavior: Stochastic processes in psychology. San Francisco: Freeman.Google Scholar
  36. Weir, B. S. (1990).Genetic data analysis. Sunderland, MA: Sinaver Associates.Google Scholar
  37. Wu, J. (1983). On the convergence properties of the EM algorithm.The Annals of Statistics, 11, 95–103.Google Scholar

Copyright information

© The Psychometric Society 1994

Authors and Affiliations

  • Xiangen Hu
    • 1
  • William H. Batchelder
    • 1
  1. 1.School of Social SciencesUniversity of CaliforniaIrvine

Personalised recommendations