Continuous and discrete latent structure models for item response data

Relations are examined between latent trait and latent class models for item response data. Conditions are given for the two-latent class and two-parameter normal ogive models to agree, and relations between their item parameters are presented. Generalizationss are then made to continuous models with more than one latent trait and discrete models with more than two latent classes, and methods are presented for relating latent class models to factor models for dichotomized variables. Results are illustrated using data from the Law School Admission Test, previously analyzed by several authors.     

Some hypotheses concerning the evolution of polysemous words

The evolution of polysemous words is examined in three studies. Study 1 indicates that older words are more polysemous than recent words, and that frequently used words are more polysemous than infrequently used words. These effects were replicated, controlling for word concreteness (Study 2) and evaluation (Study 3), using two other samples of words. If words evolve additional meanings through metaphor (MacCormac, 1985), then variables (such as concreteness and evaluation) that play a role in metaphor may mediate the evolution of polysemy. The correlation between concreteness and polysemy was negligible (Study 2). As hypothesized, evaluation is a significant predictor of polysemy: Negative words are more polysemous than positive words, and relatively neutral words are more polysemous than evaluatively extreme words. These findings are consistent with the notion that polysemous words evolve, in part, through metaphor.     

Finding relations between systems of entities: a multivariate approach

A procedure to find out the relationship between systems of entities is described. The crux of the approach depends on projecting the systems to be compared on the same arbitrary hyperspace so that any two systems of entities can be related. The procedure described applies to data presented in square or rectangular non-singular matrices and provides a procedure to determine multiple regression coefficients, multiple correlations, canonical correlations and the relationships between the inner product of the row vectors and of the column vectors of the matrices involved. The operations described apply to raw values, deviation or standard scores, and may help in interdisciplinary research, though the risks involved in the blind application of a mathematical abstract formulation should not be ignored.     

The role of compliance in hypnotic and nonhypnotic analgesia

We modified the standard procedure for administering finger pressure pain in order to assess the extent to which subjects biased their reports of pain reduction when exposed to compliance-inducing instructions. Experiment 1 used only highly hypnotizable subjects and found that compliance-induced reductions in reported pain were about half as large as the reported reductions produced by hypnotic analgesia suggestions. Experiment 2 used low as well as high hypnotizables and administered hypnotic or nonhypnotic analgesia instructions and compliance instructions to the same subjects on separate pain trials. Reported pain reductions produced by hypnotic and nonhypnotic analgesia suggestions and those produced by compliance instructions were substantially and significantly correlated. Furthermore, hypnotizability correlated with compliance-induced reported pain reductions as highly as it correlated with the reported reductions produced by hypnotic analgesia suggestions. Among high hypnotizables (but not low hypnotizables) compliant responding was predicted by social desirability. The findings of both studies indicate that compliance strongly influences the reports of hypnotic and nonhypnotic analgesia proffered by high hypnotizables. The role of compliance in the pain reductions reported by low hypnotizables is less clear-cut.     

Syntactic comprehension in Japanese aphasics: effects of category and thematic role order.

Syntactic comprehension disturbances in Japanese aphasics were investigated focusing on the role of category order and thematic role order. The data from Japanese aphasics show that the canonicity of thematic role order determines the ease of interpretation of different sentence types. This finding implies that language-specific factors, rather than language-universal factors, influence the sentence comprehension mechanisms that are retained by aphasic patients.     

The predictive validity of teacher nominations: A five-year followup of at-risk youth

The purpose of this study was to examine the predictive validity of a teacher nomination procedure for identifying at- risk children. Two hundred and twenty-five children were nominated by their teachers as well- adjusted (n= 75), socially withdrawn (n=76), or socially aggressive (n=74) during the fourth grade. Five years later, 198 of these children (88%) were located and their adjustment evaluated. Significant differences among the nominated children were found on a variety of measures, including academic grades, sociometric status, and social behavior. In addition, differences in school dropout and delinquent offenses were noted. Results are discussed in terms of the validity of teacher nomination procedures and their utility in identifying at- risk youth.     

Rate variables and automatized naming in developmental dyslexia

The rate variable in rapid automatized naming (RAN) was investigated in 50 adolescent and 40 adult students with developmental dyslexia, in matched normal controls, and in learning-disabled students without reading difficulties. Visual stimuli depicting familiar colors and common objects were presented in isolation at three film speeds and three exposure times. Film speed and exposure time contributed as independent variables to error rate; and dyslexic subjects of both age groups made significantly more naming errors than controls. Dyslexic subjects also responded with longer naming latencies than controls when the same RAN stimuli were presented in a continuous sequential mode as a matrix of rows and columns. Naming latencies in the sequential presentation were highly correlated with naming errors in the film version. The implications of reduced naming rates for nongraphological stimuli in developmental dyslexia are discussed.     

Stimulus repetition effects and the dimension-feature distinction in alternative targets

The negative repetition effect (NRE) refers to a poorer perception of the target in noise-same-as-target than in noise-alternative-target displays. Using the parenthesis as target and noise that are either identical or different in orientation, we tested predictions derived from a feature-specific inhibition model (Bjork & Murray, 1977) that explains the NRE as arising from inhibitory interactions among channels handling identical inputs. Contrary to our expectation of evidence for orientation-specific inhibition, a positive repetition effect (PRE) was observed. We also found a PRE for the dimension target sets whose stimuli differ only in orientation and an NRE for the feature target sets whose stimuli differ in the number of elements. Further experiments that systematically varied relationships between alternative targets indicate that these basic findings, termed the stimulus attributes effects, do not arise from pattern masking, perceptual strategies, or response bias.     

The effects of right hemisphere damage on the pragmatic interpretation of conversational remarks

The ability to interpret conversational utterances was assessed in a group of 12 male patients with unilateral right hemisphere brain damage (RHD) and 12 non-brain-damaged, age-matched male control subjects. Subjects listened to short vignettes which described both the affective tone of the relationship between a speaker and an actor, and the actor's performance on a task. Each vignette concluded with the speaker characterizing the actor's performance. In half of the items, the speaker's utterance was literally true; in the other half, the utterance was literally false and invited a nonliteral interpretation. Results showed no appreciable differences in the performance of control subjects and RHD patients when interpreting literally true utterances. In contrast, the two groups differed reliably when interpreting the pragmatic intent of nonliteral utterances: Control subjects used information about both the actor's performance and the speaker-actor relationship, while RHD patients demonstrated difficulty in using the information about the speaker-actor relationship. Results have implications for patients' understanding of essential elements of conversations, such as characters' internal states and their intentions in employing different forms of literal and nonliteral language.     

The ontogeny of hemispheric specialization: some old hypotheses revisited

This paper re-examines two hypotheses on the ontogeny of hemispheric specialization that long ago fell into disfavor in neuropsychology. It is argued that these hypotheses, equipotentiality and progressive lateralization, may have suffered from misinterpretation that fostered premature acceptance of the developmental invariance view of hemispheric specialization. An alternative view is presented that incorporates both hypotheses within a neurodevelopmental context using anatomical (postmortem) and childhood aphasia data.