Treating connectionism properly: Reflections on Smolensky

Summary In this essay, I undertake to examine the principal theses of Paul Smolensky's 1988Behavioral and Brain Sciences target article, On the proper treatment of connectionism, from the point of view of the methodology and epistemology of science, that is, the philosophical theory of theories in general. After exploring the instrumentalist and realist views of the relationships between micro- and macrotheories on their home ground in the natural sciences, the procedures by which a phenomenally described cognitive task is prepared for symbolic or subsymbolic modeling, and the contrast between the deliberate conscious reasoning processes of a novice and intuitive behavior of an expert in solving a given family of cognitive problems, I argue that although Smolensky is right about what it would take for connectionist subsymbolic models to relate to symbolic models as micro- to macrotheories, he is wrong in concluding that they do. On the contrary, it would be something of a miracle if the idealized nomological structure of the behavior of stochastic patterns of activity over large numbers of subsymbolic units in a connectionist machine corresponded even approximately to the nomological structure of the conceptual level behavior of a Von Neumann computer running off a program whose syntax had been explicitly designed to structurally operationalize a determinate fragment of intentional semantics — unless, of course, the connectionist machine had been deliberately constructed to implement the symbol processor in the first place.I conclude that the proper treatment of connectionism is to be found among the blandly ecumenical proposals for irenic cooperation and division of labor that Smolensky considers and rejects.     

Homeless men and women: Commonalities and a service gender gap

Gender differences among homeless persons on a host of variables were examined. A stratified random sample of 248 homeless persons staying in shelters in St. Louis provided data for the study. More similarities than differences between men and women were found. However, men tended to be homeless for longer periods and were more likely than women to sleep on the streets. Men were also more likely than women to have a drinking problem and to have been convicted of a crime. Men were less likely than women to ever have been married and to be caring for dependent children. The most striking gender differences occurred on service utilization variables; women were much more likely than men to have received social services. Comparison of our data with previous studies and potential explanations for the gender gap in service utilization are offered.     

SIX CHARACTERISTICS OF A POSTPATRIARCHAL CHRISTIANITY

Abstract. Christianity is best understood not as a set of timeless doctrines, but as a historical movement capable of change and growth. In this respect, Christianity is like a science. Heretofore, most instances of Christianity have exhibited certain ways of thinking that, taken as a whole, have led to the subordination of women (and the Earth and animals as well) to men in power. This article describes these ways of thinking, then contrasts six ways of thinking and acting that can inform postpatriarchal Christianity and science.     

Evolution, Traits, and the Stages of Human Courtship: Qualifying the Parental Investment Model

ABSTRACT Individual differences are explicitly connected to social interaction in Darwin's notion of sexual selection Traits that increase the probability of successful reproduction will tend to increase in frequency This process operates partly through differential choice, by one sex, of certain traits in the other According to the parental investment model, females frequently have more stringent criteria for the traits they will accept in a mate because they have a relatively larger investment in each offspring Because human mating arrangements often involve a substantial commitment of resources by the male, it is necessary to invoke a distinction between the selectivity involved during casual mating opportunities and the selectivity exercised when choosing a long-term partner Ninety-three undergraduate men and women rated their minimum criteria on 24 partner characteristics at four levels of commitment In line with an unqualified parental investment model, females were more selective overall, particularly on status-linked variables In line with a qualified parental investment model, males' trait preferences depended upon the anticipated investment in the relationship Males had lower requirements for a sexual partner than did females, but were nearly as selective as females when considenng requirements for a long-term partner     

Cues to obsessive-compulsive symptoms: Relationships with other patient characteristics

The waxing and waning of symptoms in obsessive-compulsive disorder (OCD) suggests that environmental cues may impact on exacerbations of this disorder. Eighty-one Ss with OCD completed the Cues Checklist (CCL; Mackenzie, Ristvedt, Christenson, Lebow & Mitchell, 1992), a 339-item checklist of rationally-derived cues and circumstances that might be expected to elicit or worsen symptoms. Principal components analysis revealed four components: household order and organization, contamination and cleaning, negative affect, and prevention of harm and checking. Total number of cues endorsed and component scores were correlated with other characteristics of the disorder, and with the presence of other Axis I and II disorders. Patterns of cue endorsement related to standard measures of obsessive-compulsive content but not to symptom severity. Early-onset Ss endorsed a greater number of cues. History of depression, anxiety disorders and Axis II pathology related most strongly to scores on the negative affect component. The values and limitations of this approach are discussed.     

Are morphological structures computed during word recognition?

In the recent literature in generative morphology, it has been assumed that multimorphemic words must be characterized as having a hierarchical internal structure which is expressible in the form of morphological trees. This paper reports on an experiment which addresses the following question: Are these structures relevant to complex word recognition? In a naming latency experiment, subjects were presented with prefixed and suffixed nonsense roots. The use of stimuli such as these allowed for the control of real word effects such as frequency and semantic plausibility and made it possible to systematically vary the configuration of the morphological trees. Significant response time differences were found between the morphologically illegal forms and legal configurations. This was taken as evidence that subjects do compute morphological representations. Because no differences were found between legal left-branching structures and legal right-branching structures, it was concluded that morphological computation is not sensitive to the serial ordering of morphemes within a complex word.     

Assessing violent offending in serious juvenile offenders

The convergent validity of the two most frequently used methods for assessing violent offending in juveniles (i.e., selfreports and arrests) was evaluated. Participants were 87 serious juvenile offenders and their maternal figures, primarily from disadvantaged families. Validation measures tapped established behavioral, family, and peer correlates of delinquency. Results failed to support the ability of either arrests for violent crimes or selfreported violent offenses to index violent criminal behavior accurately. Several methodological features of the study support our hypothesis that the findings were not spurious. Procedural and conceptual implications of the findings are discussed.The Family and Neighborhood Services Project was funded by Grant 2H87MH43557-04 to the South Carolina Department of Mental Health from the National Institute of Mental Health. The authors extend their most sincere appreciation to Commissioner Joseph Bevilacqua, Commissioner Richard McLawhorn, Joe James, Dale Chandler, Madeline Blackwell, Kristen Rowden, Pam King, Nancy King, Nancy Yue, and Jerry DeLoye.     

Edgar Schein's Process Versus Content Consultation Models

Edgar Schein (1969, 1978, 1989, 1990) has proposed three models of consultation based on assumptions inherent in different helping styles. The first two models, purchase-of-expertise and doctor-patient, focus on the content of organizational problems. The client gives the problem to the consultant to find and implement solutions. The process consultation model focuses on how organizational problems are solved. The client and consultant collaborate throughout the problem-solving effort to find workable solutions. Because of the nature of organizations' cultures and the underlying assumptions that determine how they operate, Schein has suggested that it is most efficacious for consultants to begin in the process consultation mode and involve the client in the investigation of the problem.     

IF ATTITUDES AFFECT HOW STIMULI ARE PROCESSED, SHOULD THEY NOT AFFECT THE EVENT-RELATED BRAIN POTENTIAL?

Abstract— In Experiment 1, subjects completed an attitude survey to identify items toward which they held positive and negative attitudes. Subsequently, subjects were instructed to count the number of positive (or negative) stimuli in a series. Each series contained six attitude stimuli from a given semantic category (e.g., fruits), and the structure of the series was varied so that positive and negative stimuli, as indexed by subjects' idiosyncratic attitudes, were evaluatively consistent or inconsistent within the series. In Experiment 2, subjects were exposed to personality traits that were positive or negative in series of six. Again, the structure of the series was varied so that positive and negative traits were evaluatively consistent or inconsistent within the series. Results of Experiments 1 and 2 indicated that although the event-related brain potential did not differ as a function of stimulus valence per se, evaluatively inconsistent, in contrast to consistent, stimuli evoked a larger amplitude late P300-like positive component that was maximal over the centroparietal region.