Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.

Pattern Recognition and Machine Learning

Co-Planar Stereotaxic Atlas of the Human Brain—3-Dimensional Proportional System: An Approach to Cerebral Imaging, J. Talairach, P. Tournoux. Georg Thieme Verlag, New York (1988), 122 pp., 130 figs. DM 268

Ethnologue: Languages of the World

A MUCH debated question is whether sex differences exist in the functional organization of the brain for language1–4. A long-held hypothesis posits that language functions are more likely to be highly lateralized in males and to be represented in both cerebral hemispheres in females5,6, but attempts to demonstrate this have been inconclusive7–17. Here we use echo-planar functional magnetic resonance imaging18–21 to study 38 right-handed subjects (19 males and 19 females) during orthographic (letter recognition), phonological (rhyme) and semantic (semantic category) tasks. During phonological tasks, brain activation in males is lateralized to the left inferior frontal gyrus regions; in females the pattern of activation is very different, engaging more diffuse neural systems that involve both the left and right inferior frontal gyrus. Our data provide clear evidence for a sex difference in the functional organization of the brain for language and indicate that these variations exist at the level of phonological processing.

Sex differences in the functional organization of the brain for language

1. Word frequency and vocabulary size Paul Nation and Rob Waring 2. Variation in spoken and written language Michael McCarthy and Ron Carter 3. Vocabulary connections Rosamund Moon 4. Vocabulary and context William Nagy 5. Productive vs. receptive aspects of vocabulary Francine Melka 6. Models of lexical acquisition Paul Meara 7. Cognitive constraints on vocabulary acquisition Nick Ellis 8. Intralexical factors affecting the difficulty of vocabulary acquisition Batia Laufer 9. L1 influences on L2 vocabulary acquisition 10. Acquisition of the formal aspects of vocabulary Michael Swan 11. Vocabulary learning strategies Ann Ryan 12. Vocabulary and the syllabus Norbert Schmitt 13. Teaching vocabulary: most recent trends Felicity O'Dell 14. Vocabulary and testing John Read 15. Dictionaries, thesauruses, lexical resources Phil Schofield.

詞彙:描述、習得與教學=Vocabulary : description, acquisition and pedagogy

It is well known that there exist preferred landing positions for eye fixations in visual word recognition. However, the existence of preferred landing positions in face recognition is less well established. It is also unknown how many fixations are required to recognize a face. To investigate these questions, we recorded eye movements during face recognition. During an otherwise standard face-recognition task, subjects were allowed a variable number of fixations before the stimulus was masked. We found that optimal recognition performance is achieved with two fixations; performance does not improve with additional fixations. The distribution of the first fixation is just to the left of the center of the nose, and that of the second fixation is around the center of the nose. Thus, these appear to be the preferred landing positions for face recognition. Furthermore, the fixations made during face learning differ in location from those made during face recognition and are also more variable in duration; thi...

/pdf/two-fixations-suffice-in-face-recognition-3gy7xxvas7.pdf

Two Fixations Suffice in Face Recognition

We use a hidden Markov model (HMM) based approach to analyze eye movement data in face recognition. HMMs are statistical models that are specialized in handling time-series data. We conducted a face recognition task with Asian participants, and model each participant's eye movement pattern with an HMM, which summarized the participant's scan paths in face recognition with both regions of interest and the transition probabilities among them. By clustering these HMMs, we showed that participants' eye movements could be categorized into holistic or analytic patterns, demonstrating significant individual differences even within the same culture. Participants with the analytic pattern had longer response times, but did not differ significantly in recognition accuracy from those with the holistic pattern. We also found that correct and wrong recognitions were associated with distinctive eye movement patterns; the difference between the two patterns lies in the transitions rather than locations of the fixations alone.

/pdf/understanding-eye-movements-in-face-recognition-using-hidden-8dtffhzhs0.pdf

Understanding eye movements in face recognition using hidden Markov models

The complexity of Chinese orthography has hindered the progress of research in Chinese to the same level of sophistication of that in alphabetic languages such as English. Also, there has been no publicly available resource concerning the decomposition of Chinese characters, which is essential in any attempt to model the cognitive processes of Chinese character recognition. Here we report our construction and analysis of a Chinese lexical database containing the most frequent phonetic compounds decomposed into semantic and phonetic radicals according to Chinese etymology. Each radical was further decomposed into basic stroke patterns according to a Chinese transcription system, Cangjie (Chu, 1979 Laboratory of chu Bong-Foo Retrieved August 25, 2004, from http://www.cbflabs.com/). Other information such as pronunciation and character frequency were also incorporated. We examine the distribution of different types of character, the information skew in phonetic compounds, the relations between subcharacter orthographic units and the pronunciation of the entire character, and the processing implications of these phenomena in terms of universal psycholinguistic principles.

Analysis of a Chinese phonetic compound database : Implications for orthographic processing

We examined whether two purportedly face-specific effects, holistic processing and the left-side bias, can also be observed in expert-level processing of Chinese characters, which are logographic and share many properties with faces. Non-Chinese readers (novices) perceived these characters more holistically than Chinese readers (experts). Chinese readers had a better awareness of the components of characters, which were not clearly separable to novices. This finding suggests that holistic processing is not a marker of general visual expertise; rather, holistic processing depends on the features of the stimuli and the tasks typically performed on them. In contrast, results for the left-side bias were similar to those obtained in studies of face perception. Chinese readers exhibited a left-side bias in the perception of mirror-symmetric characters, whereas novices did not; this effect was also reflected in eye fixations. Thus, the left-side bias may be a marker of visual expertise.

/pdf/not-all-visual-expertise-is-holistic-but-it-may-be-leftist-58pgmx678f.pdf

Not All Visual Expertise Is Holistic, but It May Be Leftist The Case of Chinese Character Recognition

How people look at visual information reveals fundamental information about them; their interests and their states of mind. Previous studies showed that scanpath, i.e., the sequence of eye movements made by an observer exploring a visual stimulus, can be used to infer observer-related (e.g., task at hand) and stimuli-related (e.g., image semantic category) information. However, eye movements are complex signals and many of these studies rely on limited gaze descriptors and bespoke datasets. Here, we provide a turnkey method for scanpath modeling and classification. This method relies on variational hidden Markov models (HMMs) and discriminant analysis (DA). HMMs encapsulate the dynamic and individualistic dimensions of gaze behavior, allowing DA to capture systematic patterns diagnostic of a given class of observers and/or stimuli. We test our approach on two very different datasets. Firstly, we use fixations recorded while viewing 800 static natural scene images, and infer an observer-related characteristic: the task at hand. We achieve an average of 55.9% correct classification rate (chance = 33%). We show that correct classification rates positively correlate with the number of salient regions present in the stimuli. Secondly, we use eye positions recorded while viewing 15 conversational videos, and infer a stimulus-related characteristic: the presence or absence of original soundtrack. We achieve an average 81.2% correct classification rate (chance = 50%). HMMs allow to integrate bottom-up, top-down, and oculomotor influences into a single model of gaze behavior. This synergistic approach between behavior and machine learning will open new avenues for simple quantification of gazing behavior. We release SMAC with HMM, a Matlab toolbox freely available to the community under an open-source license agreement.

/pdf/scanpath-modeling-and-classification-with-hidden-markov-zgd6dy3osy.pdf

Janet H. Hsiao

Papers

Two Fixations Suffice in Face Recognition

Understanding eye movements in face recognition using hidden Markov models

Analysis of a Chinese phonetic compound database : Implications for orthographic processing

Not All Visual Expertise Is Holistic, but It May Be Leftist The Case of Chinese Character Recognition

Scanpath modeling and classification with Hidden Markov Models