Search Results

Hardy and Fazey’s (1987) cusp catastrophe model of anxiety and performance has been criticized for being overly complex and difficult to test. The present paper attempts to clarify the model for researchers who are less familiar with its more subtle nuances; it then differentiates between the characteristics of cusp catastrophe models in general and the specific predictions of Hardy and Fazey’s cusp catastrophe model of anxiety and performance. For each prediction, methodological and statistical procedures are suggested whereby the prediction can be tested, and the available evidence that has used these procedures is then briefly reviewed. Some of the practical implications of the cusp catastrophe model for best practice are also discussed.

This paper examines Hardy’s (1990, 1996a) proposition that self-confidence might act as the bias factor in a butterfly catastrophe model of stress and performance. Male golfers (N = 8) participated in a golf tournament and reported their cognitive anxiety, somatic anxiety, and self-confidence prior to their tee shot on each hole. All anxiety, self-confidence, and performance scores were standardized within participants in order to control for individual differences. The data were then collapsed across participants and categorized into a high self-confidence condition and a low self-confidence condition by means of a median split. A series of two-way (Cognitive Anxiety × Somatic Anxiety) ANOVAs was conducted on each self-confidence condition in order to fag where the maximum Cognitive Anxiety × Somatic Anxiety interaction effect size lay along the somatic anxiety axis. These ANOVAs revealed that the maximum interaction effect size between cognitive and somatic anxiety was at a higher level of somatic anxiety for the high self-confidence condition than for the low self-confidence condition, thus supporting the moderating role of self-confidence in a catastrophe model framework.

We (Hardy, Woodman, & Carrington, 2004) recently proposed an innovative segmental quadrant analysis for exploring the role of self-confidence within a higher-order catastrophe model framework. Using this exploratory analysis, we found initial support for the main hypothesis, namely that the maximum interaction effect size between cognitive anxiety and somatic anxiety would be located at a lower level along the somatic anxiety continuum for conditions of low self-Cong compared to conditions of high self-confidence. In the present issue of this journal, Tenenbaum and Becker (2005) offer a critique of this study. In formulating their critique they have employed four principal approaches: (a) a largely indiscriminate critique of catastrophe model research as a whole; (b) a more specific critique of the method and analysis employed in our study; (c) a misrepresentation of our own work and that of previous authors; and (d) abundant confusion and irrelevancy. We address each of these issues in turn.

The current paper criticizes the concept, research methodology, data analyses, and validity of the conclusions made in Hardy, Woodman, and Carrington’s (2004) article published in this journal. In their repeated-measures analysis of data from the performances of 7 golfers, they did not examine changes in performance scores on successive holes. Instead, Hardy et al. used several ANOVA models to examine how performance varied with respect to somatic and cognitive anxiety level and self-confidence interaction. By doing so, their findings produced effects which we argue to be conceptually and empirically limited. We also address problems associated with dichotomization of continuous variables, measurement errors when splitting data, eradication of random significant effects, cell sizes in segmental quadrant analysis, and correlation between somatic and cognitive anxiety. We believe these difficulties prevent any reliable conclusions and/or generalizations from being made.

This study examines intensity and direction of competitive state anxiety symptoms, and the interactive influence of anxiety subcomponents upon netball performance. Netball players (N = 45) completed the modified Competitive State Anxiety Inventory-2 (CSAI-2) and a retrospective performance measure over a season, utilizing an intraindividual design. The modified CSAI-2 includes a direction scale assessing the facilitative or debilitative interpretation of the original intensity symptoms. Although the facilitative influence of anxiety upon performance did not emerge directly through the direction scale, a significant interaction emerged from the two-factor Cognitive Anxiety × Physiological Arousal quadrant analyses, suggesting that anxiety may enhance performance, as proposed by catastrophe model predictions. Findings also highlighted the importance of self-confidence for possible inclusion in higher order catastrophe models.

Hanin’s (1980) zones of optimal functioning (ZOF) hypothesis suggests that a person is most likely to attain peak performance within an individual, specific bandwidth of state anxiety. The present study investigated Hanin’s ZOF hypothesis within a multidimensional framework, whereby zones of optimal functioning were computed for cognitive and somatic anxiety. Participants (N = 25) were members of a competitive bowling league; they completed the CSAI- 2 prior to each league match over a period of 20 weeks. Performance was operationalized as each participant’s score in the first game of each match, and these scores were standardized within subjects. The analysis revealed a significant main effect for somatic anxiety zone level and a significant interaction between cognitive and somatic anxiety zone levels (below, in, and above zone) and subsequent performance. Results are discussed in terms of the theoretical implications for future researchers, specifically in relation to the cusp catastrophe model.

Structured in-depth interviews explored the catastrophic experiences of eight elite performers. Participants responded to questions concerning an event in which they felt they had experienced an uncharacteristic but very noticeable drop in their performance, a “performance catastrophe.” Inductive and deductive analyses were employed to provide a clear representation of the data. This paper reports on how the dimensions emerging from the hierarchical content analysis changed from prior to the catastrophic drop in performance, during the drop, and after the drop (in terms of any recovery). Two emerging higher order dimensions, “sudden, substantial drop in performance” and “performance continued to deteriorate” provide support for one of the fundamental underpinnings of the catastrophe model (Hardy, 1990, 1996a, 1996b); that is, performance decrements do not follow a smooth and continuous path. The paper examines the implications of the findings with respect to applied practice and future research.