Donald F. Morrison

Cognitive and physiological substrates of impaired sentence processing in parkinson's disease

Sentence comprehension is a complex process involving at least a grammatical processor and a procedural component that supports language computations. One type of cerebral architecture that may underlie sentence processing is a network of distributed brain regions. We report two experiments designed to evaluate the cognitive and physiological substrate of sentence processing diaculties in nondemented patients with Parkinsons disease (PD). In the first experiment, patients answered simple questions about sentences that varied in their computational demands. Group and individual patient analyses indicated that PD patients are significantly compromised on this task, and that their difficulties become more prominent as the computational demands of the sentences increase. We manipulated the set of sentences to stress performance aspects of sentence processing. PD patients were compromised in their ability to detect errors in the presence and nature of a sentences grammatical morphemes, suggesting a deficit in selective attention, but their ability to answer questions about a sentence was not afFected by short-term memory factors. In the second experiment, positron emission tomography was used to correlate this pattern of sentence comprehension impairment with regional cerebral glucose metabolism (rCMRgl) obtained at rest in a representative subset of these PD patients. Grammatical comprehension and attention in sentence processing correlated significantly with mesial frontal rCMRgl. Regression analyses confirmed the central role of left mesial frontal cortex, and identified a subsidiary role for left caudate in overall sentence comprehension, for left dorsolateral frontal cortex in grammatical processing, and for bilateral dorsolateral frontal cortex in attending to the presence of grammatical features. We conclude that compromised mesial frontal functioning underlies in part the sentence processing deficit of these patients, and these data illustrate one method for mapping portions of a sentence processing mechanism onto a distributed cerebral architecture.

Expectations and Variances of Maximum Likelihood Estimates of the Multivariate Normal Distribution Parameters with Missing Data

Abstract Exact expectations and variances have been obtained for the maximum likelihood estimates of the elements of the mean vector and covariance matrix of the multivariate normal distribution when a subset of the variates does not have observations on some sampling units. The biases, variances, and mean square errors of the estimates are compared with those of the usual estimates computed from the complete observation vectors. When the correlations between the complete and incomplete sets of variates are small the multivariate missing value estimates are less efficient in the mean square error sense.

The Distribution of Linear Functions of Independent F Variates

Abstract The distribution of a linear function of two independent F variates with respective degrees of freedom n, N-n and m, N-m has been expressed as a quickly-convergent infinite series of incomplete beta functions. This exact distribution is used to determine the error of a simple F approximation to it obtained by equating the first two cumulants. By those results small sample percentage points can be obtained for some tests on the mean vector of a multivariate normal distribution whose covariance matrix has certain patterns.

Bivariate Normal Distribution

The bivariate normal distribution is defined by its density function. The regression function of one variable on the other is given by the linear conditional mean function. The conditional variance is expressed in terms of the correlation coefficient of the variables. References to tables of the bivariate normal distribution are included. Keywords: bivariate normal; regression function; normal and multivariate normal distributions

University of Pennsylvania Department of Statistics

It was one of the very first statistics departments in the United States. Established in 1931 as a spin-off from the Department of Economics, it was called the Department of Economic and Social Statistics, a name that remained until 1964. Along with the social science departments at the University of Pennsylvania, the Departments of Economics, Sociology, and Political Science, its home was in the Wharton School of Finance and Commerce. But unlike those departments, which eventually moved to the School of Arts and Sciences, it remained in Wharton and never moved.

Conference Abstracts, Anecdotes, and Appreciation

Abstracts of invited talks at the forthcoming Ames conference honoring H. A. David, and anecdotes and words of appreciation by some of the contributors and other professional colleagues are collected here.