In 1996, Roussy et al. reported a study which failed to demonstrate the predictability of REM dream content from presleep ideation. The current study was designed to further test this predictability by using a larger sample of REM dream content and waking ideation. Thirteen young-adult female university students reported three REM dreams on each of four laboratory nights spaced at weekly intervals. On two of the study days, waking Thought Samples (TS) were recorded by participants at five randomly selected times. On the two other days, participants recorded their five most Significant Concerns (SC) prior to going to sleep at the laboratory. In a within-participant ranking task, two student peers had to read each item which consisted of the waking ideation sample for one night (WI) for one participant and the four dream sets (DS) for this same participant. Their task for each of these items was to rank the 4 DS from 1 to 4 as they believed them to be most likely related to the WI (rank 1) or least likely related to the WI (rank 4). For this task, there were two items for each of 12 participants. It was hypothesized that a dream set would receive a rank closer to 1 when it was the target than when it served as a foil. In a within-participant matching task, ten student peers were given the 4 dream sets and 4 waking ideation samples separately for each participant. Judges were asked to match each DS with one of the WI, thus making 4 pairs. In the between-participant matching task, these same ten judges were given 12 items. For each item, they were asked to match 4 dream sets from different participants with their corresponding waking ideation samples. For the ranking task, a three-way mixed ANOVA (2 judges - between, 2 waking ideation conditions: TS & SC - within, 2 dream conditions: target & nontarget - within) revealed that there were no significant differences between ranks given to target or foil dreams, and this in both the TS and SC conditions, as well as by both judges. Chi-square values for both the within- and between-participant matching task were not significant (.047 and .096 respectively, p < .99). The observed number of matches was therefore not significantly different from what was expected by chance. These results are consistent with those of our earlier study and suggest that it is not possible for non-clinically trained judges to match REM dream content with waking ideation. Further analyses conducted with a clinical psychologist serving as judge demonstrated that she was also unable to complete the tasks at greater than chance levels. It therefore seems that the relationship between waking and dreaming is not as obvious and predictable as has been believed. Rather, this study supports the cognitive position that dreams appear to randomly integrate elements from waking life and do not seem to give priority to immediately preceding experiences and concerns.

The purpose of this thesis research was to investigate the decision making of rugby referees. Based on past research, concepts felt to be important in this task were anticipation, and knowledge base. To investigate these features for referees, the expertise paradigm was incorporated. Thus, 12 high and 12 low experience referees were asked to evaluate clips of videotaped rugby footage for infractions. Participants were asked to anticipate upcoming actions and calls, and their ability to detect infractions was measured along with voice reaction time (VRT). In addition, referees were asked questions related to their knowledge bases, and the information sources from which they drew in performing this task. The responses to the videotaped plays were analysed using signal detection theory (Macmillan & Creelman, 1991) and revealed no significant group differences. Further, a two-way analysis of variance performed using the information sources data revealed significant differences between the high and low experience referee groups. Further sections discuss data collected on anticipation, VRT, and another measure of knowledge, that were eventually eliminated from the study. This and other related features led to a final section describing common characteristics and defining characteristics for the general population of sport monitors.

This program of research sought an answer to the following question: What is the nature of the mechanism underlying French readers' ability to identify syllabic boundaries in print? Two complementary approaches were taken to shed some light on this problem: first, we examined how the structural properties of syllables influence syllabification decisions; and second, we explored the relationship between syllabic structure and the issue of syllabic mediation in visual word recognition (VWR). In the first study, university students were asked to syllabify a list of 248 words (Experiment 1) or pseudowords (Experiment 2), in a paper-and-pencil syllabification task. The data were coded and then analyzed according to a framework representing the realm of all possible environments in which syllabic boundaries can be identified, within any given letter string. In the second study, participants were asked to verify the proper location for a syllabic boundary marker, which was inserted between the letters of isolated words displayed at the center of a computer screen (e.g., 'ca/bane' [shack]). In the syllable boundary verification (SBV) task, trisyllabic stimuli that contained a single medial consonant or medial consonant pair were contrasted either at the first or second syllabic boundary (i.e., Experiments 1 and 3: CV.C&barbelow;V.CV vs. CV C.CV.CV; Experiment 2: CV.CV.C&barbelow;V vs. CV.CV C.CV). The results showed that lexicality was not a factor in the untimed syllabification task (Study 1), where participants relied on simple heuristics to parse letter strings (i.e., spelling-to-sound translation, prescribed rule for separating geminate consonants). In the timed verification task (Study 2), a location-specific effect of syllabic structure was observed in the critical marker position for all three experiments, in addition to a word-frequency effect. Moreover, a reliable effect of marker serial position revealed an increase in response times towards the end of words. The pattern of results for this series of experiments is most consistent with the idea of a lexically-derived syllabic structure effect and inconsistent with one form of the syllabic mediation hypothesis (i.e., syllable-as-access-code hypothesis). Overall, the findings of both studies converge towards the following conclusion: The mechanism underlying the ability to identify syllabic boundaries in French print relies on multiple sources of constraint that vary according to task demands.

Expectations for performance, such as an expectation set by within team order in gymnastics (Scheer & Ansorge, 1975; 1980), have been shown to create a non-performance bias in terms of sport performance evaluation. Plessner (1999) found such a bias to influence both the encoding and evaluation phases of gymnastics judgement. In the sport of figure skating, the encoding phase is considered the period during which a judge perceives elements and any errors that may occur. After the performance is complete, judges evaluate the merit of the performance by providing two marks that are combined to create an ordinal rank, which is considered the evaluation phase of judging. The current study investigated whether or not expectations created by the reputation, or name, of an athlete occurs in the sport of figure skating, and if so, in what phase of the process, encoding or evaluation. Fourteen female figure skaters' short programs were viewed by judges to whom the athletes were either known or unknown. Judges were asked to provide technical base marks, technical merit, and artistic impression marks (all indicative of the evaluation phase), as well as to identify the elements and associated deductions (the encoding phase variables). Post-experiment, the ordinal placements of the skaters were calculated and used to define whether biases do exist at a basic level. Indeed, the skaters' average ordinal placement was higher when they were known by the judge as compared to unknown (t = 2.43, p < .05). A more in-depth analysis revealed that skaters received higher technical merit and artistic impression marks, as well as technical base marks when known to the judges. No significant differences, however, were found for deductions allocated for errors or for the judges' percent correctness in perceiving the elements performed. Thus, the findings suggest that a reputation bias exists at the level of the evaluation phase, but not during the encoding phase, of figure skating judging.

The relation between speed of auditory discrimination and intelligence was investigated. Thirty-six females in experiment 1 and 24 in experiment 2 completed the Multidimensional Aptitude Battery (MAB; Jackson, 1984). Subjects were divided into higher (HA) and lower (LA) ability groups. An auditory oddball paradigm was employed, with the addition of a masking stimulus (1000 Hz, 55ms duration) following the standard (600 Hz, 25 ms) and deviant (700 Hz, 25 ms) tones. The interval between standard or deviant offset and masking tone onset, the stimulus onset asynchrony (SOA), was varied at 150, 50 and 25 ms. The auditory discrimination task was presented in a passive condition in which participants ignored the tones while reading a book and in an active condition wherein they detected the deviant tone. The passive condition generated the mismatch negativity (MMN), an event-related potential measure that indexes the operation of echoic memory. The active condition produced behavioural data and P3, an ERP measure associated with stimulus evaluation and classification. In both experiments, HA participants detected more deviant tones, and had faster and less variable reaction times. Overall, P3 amplitude was significantly larger for HA subjects, with some differences in P3 latency observed. The HA group bad larger MMN amplitude in the first study only; shorter MMN latencies for the HA group were observed in study 2 only. An intensity manipulation in experiment 2, intended to manipulate task difficulty, produced no graded changes in any of the variables. Results for both experiments were combined to increase power and facilitate interpretation. Overall, results indicate that more intelligent individuals process information in short-term memory more rapidly than those of lower-ability; results also suggest that speed of processing in echoic memory may underlie ability-related differences in performance.

L'objectif de la présente recherche consiste à étudier le rôle des structures dépressogènes et des événements de vie dans la rechute dépressive au sein d'une population âgée. L'échantillon est composé de 41 personnes, âgées en moyenne de 75 ans, qui vivent dans un centre pour soins de longue durée ou qui reçoivent des services offerts par un service de psychogériatrie. L'utilisation d'un devis de rémission a permis d'évaluer l'hypothèse de congruence auprès d'une population âgée qui était en rémission d'un épisode dépressif au cours d'un suivi de 6 mois. Les participant(e)s ont complété une fiche signalétique, l'Échelle de statut mental modifié (3MS), deux modules de l'Entrevue clinique structurée pour le DSM-IV, l'Échelle de dépression gériatrique, l'Échelle des attitudes dysfonctionnelles de 24 items et le Questionnaire des changements de vie récents. Les résultats des analyses ont indiqué que la congruence entre la structure de dépendance et l'impact subjectif de l'événement de vie sur le réseau social prédit la rechute dépressive. De même, la congruence entre la structure d'accomplissement et l'impact subjectif de l'événement de vie sur l'autonomie prédit la symptomatologie dépressive. La symptomatologie dépressive n'a pas été prédite par l'hypothèse de non congruence entre la structure d'accomplissement et l'impact subjectif de l'événement de vie dans le domaine social, mais de très peu. Ces résultats appuient le modèle de diathèse du stress dans la prédiction de la rechute dépressive au sein de la population âgée. Cette étude montre également que l'événement de vie doit être évalué en fonction des conséquences de l'événement pour l'individu dans son domaine de vulnérabilité et non pas seulement en fonction d'une catégorisation a priori. De plus, la présente étude souligne l'importance de la structure dépressogène d'accomplissement qui est peut-être activée par des menaces autant dans le domaine social que dans la sphère de l'accomplissement. Les implications cliniques et de la recherche sont abordées dans la discussion.

In 4 experiments, university students played video games in which one action or cause covaried with an outcome. Judgments on sufficient and necessary causes were observed. On the basis of the obtained judgments, different computational models, Cheng and Novick's (1990a, 1992) probabilistic contrast ($\Delta$P rule) and the Rescorla-Wagner (1972) model were evaluated. In Experiments 1 and 2, for the positive contingencies, the participants judged sufficient and necessary causes differently; they also showed judgment deviations from the real contingencies. The $\Delta$P rule could not account for these data. An alternative weighted $\Delta$P rule was proposed and, along with the Rescorla-Wagner model, it successfully explained these results. In Experiment 3, negative contingencies were included. The pattern of judgements among the negative sufficient and necessary causes mirrored that of the positive contingencies but did not reach statistical significance. The $\Delta$P rule could not account for the judgments in Experiment 3, the adjusted $\Delta$P rule did not either. However, the Rescorla-Wagner model accounted for the results very well. In Experiment 4, the predictive power of these different models was compared. In general, the Rescorla-Wagner model remains the best descriptive model for explaining and predicting the patterns of contingency judgments.

Eight dysphasics read two 29 word lists. Both lists consisted of long nouns of high: concreteness, imagery, and meaningfulness. Words on one list were all of a high frequency (100 or more occurrences per million). Words on the other list occurred only from one to ten times per million on the Thorndike-Lorge word frequency list. Rates of verbal perseveration, number of correct responses and response latencies were examined on the two lists. The only significant difference was found on the number of correct responses. Subjects made more errors on the low frequency list. This conforms with prior results noted in the literature. The lack of significant difference for response latencies was contrary to prior findings. These results may have been due to a higher number of "no responses" on the low frequency list, and the quickness with which misperceptions of words were given. Only two subjects perseverated at all on either list. The low abstraction of the words and the lack of switching from one level of word difficulty to another were possible explanations for the results. Subjects were given the Boston Diagnostic Aphasia Examination and the Wide Range Achievement Test - Reading. The occurrence of perseveration on these tests was examined and classified. Two broad types of perseveration were noted (as well as 15 sub-categories). One type was of an intrusive nature whereby past responses interfered with or impinged on new responses, whether they be verbal or motor. The other was a form of modified responding which served an adaptive purpose. Correlations were calculated on the number of perseverative errors made on seven sub-tests of the Boston and the Wide Range Achievement Test - Reading. Significant positive correlations were found between three tests involving naming tasks. This was probably due to their similarity in nature. A significant negative correlation existed between the Wide Range Achievement Test - Reading and one of the naming tests. Perseveration tests have usually been found to have low positive correlations, sometimes reaching significance.

Abstract not available.

Abstract not available.