Learning first-pass structural attachment preferences with dynamic grammars and recursive neural networks

One of the central problems in the study of human language processing is ambiguity resolution: how do people resolve the extremely pervasive ambiguity of the language they encounter? One possible answer to this question is suggested by experience-based models, which claim that people typically resolve ambiguities in a way which has been successful in the past. In order to determine the course of action that has been "successful in the past" when faced with some ambiguity, it is necessary to generalize over past experience. In this paper, we will present a computational experience-based model, which learns to generalize over linguistic experience from exposure to syntactic structures in a corpus. The model is a hybrid system, which uses symbolic grammars to build and represent syntactic structures, and neural networks to rank these structures on the basis of its experience. We use a dynamic grammar, which provides a very tight correspondence between grammatical derivations and incremental processing, and recursive neural networks, which are able to deal with the complex hierarchical structures produced by the grammar. We demonstrate that the model reproduces a number of the structural preferences found in the experimental psycholinguistics literature, and also performs well on unrestricted text.     

Deleterious effects of alcohol intoxication: diminished cognitive control and its behavioral consequences

The authors tested the hypothesis that impaired behavioral performance during intoxication results partly from alcohol's deleterious effects on cognitive control. The impact of alcohol on perseverative behavior was examined with an n-back working memory task that included manipulations of task complexity and prepotency of inclinations to respond or withhold responding. Thirty-two social drinkers (16 men) participated in either an alcohol (.075g/100ml) or a no-alcohol condition. Alcohol increased perseveration of prepotent, task-inappropriate response patterns only under cognitively demanding (heavy memory load) conditions. This effect was evident for both commission errors (response persistence despite contingencies altered to require restraint) and omission errors (failure to respond when contingencies were revised to encourage action). Findings suggested that alcohol-induced perseveration arises from impairments in cognitive control.     

Team learning: collectively connecting the dots

This article tests the degree to which personal and situational variables impact the acquisition of knowledge and skill within interactive project teams. On the basis of the literature regarding attentional capacity, constructive controversy, and truth-supported wins, the authors examined the effects of cognitive ability, workload distribution, Agreeableness, Openness to Experience, and structure on team learning. Results from 109 four-person project teams working on an interdependent command and control simulator indicated that teams learned more when composed of individuals who were high in cognitive ability and when the workload was distributed evenly. Conversely, team learning was negatively affected when teams were composed of individuals who were high in Agreeableness. Finally, teams using a paired structure learned more than teams structured either functionally or divisionally. Theoretical and practical implications are discussed, as well as possible limitations and directions for future research.     

The Tripartite Model of Neuroticism and the suppression of depression and anxiety within an escalation of commitment dilemma

We found evidence of a mutual suppression effect between anxiety and depression on an individual's level of commitment within escalation dilemmas. On the one hand, our results demonstrate a positive relationship between anxiety and level of commitment; on the other, our results demonstrate a negative relationship between depression and level of commitment. Based on the opposing relationships between anxiety and depression and commitment, the broad factor of neuroticism does not demonstrate any relationship with level of commitment, and the significant effects of anxiety and depression on commitment is contingent upon partialling the effect of the other facet of neuroticism. Thus, we contend that applied psychologists, who have focused on neuroticism as a broad construct, should consider the large body of work among clinical psychologists, who argue that anxiety and depression have unique variance associated with them. We conclude by addressing organizational implications of measuring the broad trait of neuroticism more narrowly.     

What Front-Line CBO Staff Can Tell Us About Culturally Anchored Theories of Behavior Change in HIV Prevention for Asian/Pacific Islanders

Few rigorously tested primary prevention programs have been developed to prevent HIV infection among immigrant communities in the United States. This is in part because of the lack of culturally specific behavioral theories that can inform HIV prevention for immigrant communities in the United States. This article aims to develop such theories for a population—Asian/Pacific Islanders (A/PIs) immigrant communities—who have been overlooked in theory development and program evaluation. Frontline community-based organization (CBO) peer educators, an underutilized source of expertise regarding cultural factors specific to HIV infection among A/PI communities, are the sample of study Asian/Pacific Islander peer educators working at an urban AIDS service organization devoted to health promotion for this population; (N=35). They were interviewed to examine (1) detailed narratives describing instances of behavior change and (2) culturally anchored theories of behavior change which the narratives imply. Theories of the influence of positive cultural symbols on the taboo of HIV/AIDS, moderators of the effectiveness of social network influences on behavior change, and setting- and community-level processes predicting HIV risk behavior were implicit in the peer educators' narratives. Implications for future research, methodology and prevention practice are discussed.     

Lateralized effects of subthalamic nucleus stimulation on different aspects of speech in Parkinson's disease

In this report, we have analyzed the effects of left and right STN stimulation separately on different aspects of speech. Significant differences were found between left and right stimulation. It appears that selective left-sided stimulation has a profoundly negative effect on prosody, articulation and hence intelligibility. Right-sided stimulation does not display this side-effect. There is no significant difference in speech characteristics between bilateral stimulation on and off. We suggest that a balanced tuning of bilateral basal ganglia networks is necessary for speech, and that the left circuit is probably dominant.     

Evidence-based treatment of stuttering: II. Clinical significance of behavioral stuttering treatments

An evidence-based framework can be described as an empirically-driven, measurement-based, client-sensitive approach for selecting treatments. It is believed that using such a framework is more likely to result in a clinically significant outcome. For this paper, a clinically significant outcome was defined as a meaningful treatment change. It was suggested that there are at least three groups for whom a treatment's outcome is meaningful. These groups include clinicians/clinical researchers, the clients, and relevant others who have some interest in the outcome (e.g., parents of a child who stutters). The meaning and measurement of clinical significance was discussed for each of these three groups, based on research from the behavioral stuttering treatment literature. EDUCATIONAL OBJECTIVES: The reader will learn about and be able to (1) broadly define a clinically significant outcome and identify some of the groups who are interested in such an outcome and (2) describe how clinical significance has been evaluated in stuttering treatment within an evidence-based framework.     

Towards a functional neural systems model of developmental stuttering

This paper overviews recent developments in an ongoing program of brain imaging research on developmental stuttering that is being conducted at the University of Texas Health Science Center, San Antonio. This program has primarily used H₂¹⁵O PET imaging of different speaking tasks by right-handed adult male and female persistent stutterers, recovered stutterers and controls in order to isolate the neural regions that are functionally associated with stuttered speech. The principal findings have emerged from studies using condition contrasts and performance correlation techniques. The emerging findings from these studies are reviewed and referenced to a neural model of normal speech production recently proposed by Jürgens [Neurosci. Biobehav. Rev. 26 (2002) 235]. This paper will report (1) the reconfiguration of previous findings within the Jürgens Model; (2) preliminary findings of an investigation with late recovered stutterers; (3) an investigation of neural activations during a treatment procedure designed to produce a sustained improvement in fluency; and (4) an across-studies comparison that seeks to isolate neural regions within the Jürgens Model that are consistently associated with stuttering. Two regions appear to meet this criterion: right anterior insula (activated) and anterior middle and superior temporal gyri (deactivated) mainly in right hemisphere. The implications of these findings and the direction of future imaging investigations are discussed.

Educational objectives: The reader will learn about (1) recent uses of H₂¹⁵O PET imaging in stuttering research; (2) the use of a new neurological model of speech production in imaging research on stuttering; and (3) initial findings from PET imaging investigations of treated and recovered stutterers.     

Parental assessment and treatment of food selectivity in natural settings

This study evaluated the effects of a parent-conducted functional analysis and treatment consisting of differential reinforcement of an alternative behavior, escape extinction, and demand fading on food selectivity in a young child with autism. Increases in food acceptance at home and in a restaurant were obtained.     

Lateralization of two-transient stimuli

In this study of the precedence effect in binaural hearing, subjects adjusted the interaural delay of a wideband acoustic pointer to match the perceived intracranial position of transient test stimuli presented over headphones. The test stimuli had leading and lagging components (either brief noise bursts or clicks), each with its own interaural delay. In some test conditions, the leading and lagging stimuli were coherent copies of one another, whereas in others, they were independent samples of noise. The duration of the stimuli and the delay from the leading component to the lagging component were also varied. All the stimulus conditions showed a moderate or strong precedence effect (i.e., covariation of perceived lateral position of the composite two-transient stimulus with the interaural delay of the leading component). Predictions of the lateralization data are presented for variants of models based on temporal weighting and/or bandpass correlation. In one model variant, the binaural stimuli are temporally weighted to emphasize the onset and then subjected to bandpass correlation analysis. In another variant, it is assumed that the onset mechanism provides a rough estimate of the initial interaural delay that guides a slower and more focused bandpass correlation analysis. The accuracies of these two model's predictions were equivalent and superior to those of models that either represent leading and lagging cues equally (bandpass correlation with no onset effect) or do not represent lagging cues at all (a complete precedence effect). The results of these analyses show the need for both a strong onset effect and for bandpass correlation analysis and suggest two modeling approaches for achieving that goal.