Intelligent quotient estimation of mental retarded people from different psychometric instruments using artificial neural networks

OBJECTIVE The estimation of a person's intelligence quotient (IQ) by means of psychometric tests is indispensable in the application of psychological assessment to several fields. When complex tests as the Wechsler scales, which are the most commonly used and universally recognized parameter for the diagnosis of degrees of retardation, are not applicable, it is necessary to use other psycho-diagnostic tools more suited for the subject's specific condition. But to ensure a homogeneous diagnosis it is necessary to reach a common metric, thus, the aim of our work is to build models able to estimate accurately and reliably the Wechsler IQ, starting from different psycho-diagnostic tools. METHODS AND MATERIALS Four different psychometric tests (Leiter international performance scale; coloured progressive matrices test; the mental development scale; psycho educational profile), along with the Wechsler scale, were administered to a group of 40 mentally retarded subjects, with various pathologies, and control persons. The obtained database is used to evaluate Wechsler IQ estimation models starting from the scores obtained in the other tests. Five modelling methods, two statistical and three from machine learning, that belong to the family of artificial neural networks (ANNs) are employed to build the estimator. RESULTS Several error metrics for estimated IQ and for retardation level classification are defined to compare the performance of the various models with univariate and multivariate analyses. Eight empirical studies show that, after ten-fold cross-validation, best average estimation error is of 3.37 IQ points and mental retardation level classification error of 7.5%. Furthermore our experiments prove the superior performance of ANN methods over statistical regression ones, because in all cases considered ANN models show the lowest estimation error (from 0.12 to 0.9 IQ points) and the lowest classification error (from 2.5% to 10%). CONCLUSIONS Since the estimation performance is better than the confidence interval of Wechsler scales (five IQ points), we consider models built very accurate and reliable and they can be used into help clinical diagnosis. Therefore a computer software based on the results of our work is currently used in a clinical center and empirical trails confirm its validity. Furthermore positive results in our multivariate studies suggest new approaches for clinicians.

[1]  R. Telford,et al.  A 14,000-Year Oxygen Isotope Record from Diatom Silica in Two Alpine Lakes on Mt. Kenya , 2001, Science.

[2]  Wm. R. Wright.  General Intelligence, Objectively Determined and Measured. , 1905 .

[3]  C. Drew.  Vineland Adaptive Behavior Scales , 2010 .

[4]  J. Raven.  Guide to using the Coloured Progressive Matrices. , 1958 .

[5]  L. Crocker,et al.  Introduction to Classical and Modern Test Theory , 1986 .

[6]  Bernard Widrow,et al.  Improving the learning speed of 2-layer neural networks by choosing initial values of the adaptive weights , 1990, 1990 IJCNN International Joint Conference on Neural Networks.

[7]  Donald F. Specht,et al.  A general regression neural network , 1991, IEEE Trans. Neural Networks.

[8]  Laurent Mottron,et al.  The Level and Nature of Autistic Intelligence , 2007, Psychological science.

[9]  Lianfen Qian,et al.  Regularized Radial Basis Function Networks: Theory and Applications , 2002, Technometrics.

[10]  S. Hyakin,et al.  Neural Networks: A Comprehensive Foundation , 1994 .

[11]  J. P. Guilford,et al.  Fundamental statistics in psychology and education, 3rd ed. , 1956 .

[12]  T. Kuhlmann,et al.  ICF (International Classification of Functioning, Disability and Health) , 2011 .

[13]  J. Freidman,et al.  Multivariate adaptive regression splines , 1991 .

[14]  D. Wechsler.  Wechsler Intelligence Scale for Children , 2020, Definitions.

[15]  Vincenzo Catania,et al.  Feedforward artificial neural network to estimate iq of mental retarded people from different psychometric instruments , 2009, 2009 International Joint Conference on Neural Networks.

[16]  John R. Wolberg,et al.  Data Analysis Using the Method of Least Squares: Extracting the Most Information from Experiments , 2005 .

[17]  Charles D. Spielberger,et al.  Encyclopedia of applied psychology , 2004 .

[18]  P ? ? ? ? ? ? ? % ? ? ? ? , 1991 .

[19]  J. D Gergonne,et al.  The application of the method of least squares to the interpolation of sequences , 1974 .

[20]  George McCloskey,et al.  Leiter International Performance Scale , 1987 .

[21]  Allan Pinkus,et al.  Approximation theory of the MLP model in neural networks , 1999, Acta Numerica.

[22]  Jyh-Shing Roger Jang,et al.  ANFIS: adaptive-network-based fuzzy inference system , 1993, IEEE Trans. Syst. Man Cybern..

[23]  Stephen L. Chiu,et al.  Fuzzy Model Identification Based on Cluster Estimation , 1994, J. Intell. Fuzzy Syst..

[24]  A. L. Edwards,et al.  An introduction to linear regression and correlation. , 1985 .

[25]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[26]  James C. Bezdek,et al.  Pattern Recognition with Fuzzy Objective Function Algorithms , 1981, Advanced Applications in Pattern Recognition.

[27]  J. Raven,et al.  Guide to using the coloured progressive matrices : sets A, Ab, B , 1963 .

[28]  Hiroaki Kazui,et al.  [Wechsler Adult Intelligence Scale-III (WAIS-III)]. , 2011, Nihon rinsho. Japanese journal of clinical medicine.

[29]  George Cybenko,et al.  Approximation by superpositions of a sigmoidal function , 1989, Math. Control. Signals Syst..

[30]  Shin Suzuki,et al.  Constructive function-approximation by three-layer artificial neural networks , 1998, Neural Networks.

[31]  Ke Zeng,et al.  A comparative study on sufficient conditions for Takagi-Sugeno fuzzy systems as universal approximators , 2000, IEEE Trans. Fuzzy Syst..

[32]  J. P. Guilford,et al.  Fundamental statistics in psychology and education , 1943 .

[33]  Philip D. Wasserman,et al.  Advanced methods in neural computing , 1993, VNR computer library.

[34]  S. Warschausky,et al.  WISC-III index growth curve characteristics following traumatic brain injury. Wechsler Intelligence Scale for Children--Third Edition. , 1999, Journal of clinical and experimental neuropsychology.

[35]  J. Sterne,et al.  Essential Medical Statistics , 2003 .

[36]  Serafino Buono,et al.  Diagnosis of Intellectual Disability: comparison between clinical criteria and automatized procedures , 2009 .

[37]  Nariman Sepehri,et al.  Modifying Weights Layer-By-Layer with Levenberg-Marquardt Backpropagation Algorithm , 2001, Intell. Autom. Soft Comput..

[38]  T.,et al.  Training Feedforward Networks with the Marquardt Algorithm , 2004 .

[39]  Geoffrey J. McLachlan,et al.  Analyzing Microarray Gene Expression Data , 2004 .

[40]  Simon Haykin,et al.  Regularized radial basis functional networks: theory and applications , 2001 .

[41]  J. Guilford.  Fundamental statistics in psychology and education , 1943 .

[42]  Michio Sugeno,et al.  Fuzzy identification of systems and its applications to modeling and control , 1985, IEEE Transactions on Systems, Man, and Cybernetics.

[43]  Alan C. Evans,et al.  Intellectual ability and cortical development in children and adolescents , 2006, Nature.

[44]  G. Chard.  International Classification of Functioning, Disability and Health , 2004 .

[45]  S. Stigler.  Gergonne's 1815 paper on the design and analysis of polynomial regression experiments , 1974 .