Abstract—Human emotions are deeply intertwined with cognition. Emotions direct cognitive processes and processing strategies of humans. The goal of this work is to design a model with the capability of classifying the uncertainty, contradiction and the cognitive nature of the emotions. For achieving this, 3D cognitive model is designed. This model enhances our vision of classification of emotions produced by reinforcing stimuli. In this model the dimensions represent the positive reinforcers, the negative reinforcers and the emotion content present. The positive reinforcer increases the probability of emission of a response on which it is contingent, whereas the negative reinforcer increases the probability of emission of a response that causes the reinforcer to be omitted. This model increases the number of emotions, that can be classified. Presently this model can classify 22 emotions subject to the presence of a facial expression database. It has the flexibility to increase upon the number of emotions. For emotion (pattern) identification, the pose and illumination factor are removed using Gabor wavelet transforms and the size is reduced by finding its principle components (PCA). This component vector is used for training the neural network. The test result shows the recognition accuracy of 85.7% on The Cohn-Kanade Action Unit Coded Facial Expression Database. The real time processing for identification, aids in applying emotions to real time audio player. An environment, that is all pervasive or ubiquitous, that would sense one’s mental state and play the appropriate musical track to maintain the positive emotional state or ease from a negative emotional state. Index Terms— Cognitive model, Emotions, 3D architecture, reinforcing stimuli. Manuscript received November 9, 2006. Maringanti Hima Bindu is an Assistant Professor with the Indian Institute of Information Technology, Allahabad, India 211011 (phone: +91-532- 2922096,+91-9335070621; fax: +91-532-2430006; e-mail: mhimabindu@ iiita.ac.in). Priya Gupta was with Indian Institute of Information Technology, Allahabad, India 211011 as a post graduate student. Prof.U.S.Tiwary is with the Indian Institute of Information Technology, Allahabad, India 211011, also as a Dean of Academic Affairs. (e-mail: ust@iiita.ac.in). I. INTRODUCTION Cognitive Science is an interdisciplinary science which includes the mental states and processes such as thinking, remembering, perception, learning, consciousness, emotions etc. Of all the kinds of cognition, emotion holds the key for human social behavior. Identification and classification of emotions by computers has been a research area since Charles Darwin’s age. These actions could be performed with the help of facial images, blood pressure measurement, pupillary dilation, facial expressions and many more quantifiable attributes of humans [1]. Facial expression recognition is an area which poses great challenges for the researchers. It is an area where a lot has been done and a lot more can be done. Facial expression recognition is not a theoretical field but finds practical applications in many fields. Coupled with human psychology and neuroscience it can come up as an area which can bridge the divide between the more abstract area of psychology and the more crisp area of computations. The characteristic feature points [10] of a face are located at eyebrows, eyelids, cheeks, lips, chin and forehead. The feature points after being extracted from these regions help in recognizing the various expressions of a face. The first and the most important step in feature detection is to track the position of the eyes. Thereafter, the symmetry property of the face with respect to the eyes is used for tracking rest of the features like eyebrows, lips, chin, cheeks and forehead. Splitting face into two halves eases the process further. This paper uses Discrete Hopfield Networks as the basis for pre-processing of the Face-Signal and then Feature Extraction. A number of techniques have been proposed in this field and are being used which include Bayesian Classification [11],Gabor Wavelet Transform [12], Principle Component Analysis, HMM [13], Line-based Caricatures [14], Method of Optical Flow Analysis [15] etc. But they have an inherent complexity which makes them opaque and are computationally expensive. Apart from emotion classification and identification, the response from the computers is also being generated for making the human-computer interaction livelier. The field of affective computing has emerged for this kind of Cognitive Model – Based Emotion Recognition From Facial Expressions For Live Human Computer Interaction Maringanti Hima Bindu, Priya Gupta, and U.S.Tiwary, Senior Member, IEEE 351 Proceedings of the 2007 IEEE Symposium on Computational Intelligence in Image and Signal Processing (CIISP 2007) 1-4244-0707-9/07/$25.00 ©2007 IEEE