A Laboratory Investigation of Positive and Negative Affect Within Individual Zones of Optimal Functioning Theory.

Russell, William D. ; Cox, Richard H.

Positive and Negative affect (PNA) were measured within Individual Zone of Optimal Functioning Theory to examine (a) IZOF position and motor task performance, (b) task and optimal PNA associated with performance, (c) actual and recalled optimal PNA, and (d) actual optimal PNA score correlations with recalled optimal PNA. Fifty-five college males were administered the Positive and Negative Affect Schedules (Watson, Clark, & Tellegen, 1988) and were assessed in basketball and football tasks. ANOVAs examining IZOF position and motor task performance indicated statistical support, however low effect sizes suggest IZOF limitations. No differences were found between actual and recalled optimal NA, however significant differences (p [less than].01) were found between actual and recalled PA for basketball. While some support was provided, weak effect sizes indicated IZOF limitations.

The anxiety-performance relationship has long been of great empirical interest and recently Hanin's (1980, 1986) individual zones of optimal functioning theory has posited that differences exist across athletes in precompetitive anxiety necessary for optimal performance. Optimal performance is likely to occur when one's anxiety falls within one's predetermined zone which can be measured directly or through retrospective recall (Hanin, 1986). The direct method involves empirical assessment until an optimal performance is identified, while the recall method involves obtaining retrospective recall of anxiety state just prior to an optimal performance.

Original IZOF work (Hanin, 1980, 1986) focused on pre-competitive state anxiety framework using Spielberger's (1983) State Trait Anxiety Inventory (STAI). Recent work, however, has incorporated a multidimensional approach (Gould, Tuffey, Hardy, & Lochbaum, 1993; Krane, 1993; Randle & Weinberg, 1997) and analysis of facilitating/debilitating emotions (Hanin & Syrja, 1995, 1996). Empirical results have supported the hypothesis that within-zone performances are better than out-of-zone performances (Hanin, 1986; Morgan, O'Connor, Sparling, & Pate, 1987). Further IZOF theory support has been found using mood states (Prapavessis & Grove, 1991) and multidimensional competitive anxiety (Gould et al., 1993). Finally, IZOF theory has been examined using positive and negative affect, as measured by the Positive and Negative Affect Scales (PANAS) (Watson, Clark, & Tellegen, 1988) and IZOF theory supported with respect to facilitative and debilitative anxiety and emotion (Hanin & Syrja, 1995). IZOF theory proposes that p erformances which fall within one's optimal zone will be better than performances which fall outside of (below or above) one's predetermined optimal zone which are determined by taking the athlete's optimal value associated with best performance (i.e. precompetitive positive/negative affect) plus-or-minus one-half standard deviation.

Retrospective assessment of individual zones has shown to be accurate in assessment of precompetitive affective states associated with optimal performance (Hanin, 1986; Harger & Raglin, 1994) and has been effective in predicting optimal pre-competitive anxiety (Raglin, Morgan, & Wise, 1990). However, Jokela and Hanin (1997) indicated that IZOF data was heterogeneous and 88% of the variance from retrospective recall was left unexplained. Some concern has been expressed over retrospective assessment due to weaker discrimination between high and low anxious individuals (Tenenbaum & Furst, 1984), changes in ratings of psychological states based on bogus feedback (Brewer Van Raalte, Linder, & Van Raalte, 1991), exaggerated differences in recall of best and worst performances (Krane, 1993), and reduced accuracy for longer recall intervals (Imlay, Carda, Stanbrough, Drieling, & O'Connor, 1993). The implications are that inaccurate recall may result in inaccurate zones. To date, no studies have compared these two as sessment methods in a controlled setting.

Recent studies using the Competitive State Anxiety Inventory-2 (CSAI-2) (Martens, R., Vealey, R., & Burton, D., 1990) and multidimensional anxiety predictions (Annesi, 1997; Randle & Weinberg, 1997; Woodman, Albinson, & Hardy, 1997) have been less supportive of IZOF theory. Randle and Weinberg (1997) examined IZOF theory using a multidimensional anxiety approach in collegiate softball players and found no differences in objective or subjective performance related to whether performance was inside/outside of zones. Woodman, Albinson, and Hardy (1997) crafted cognitive and somatic zones using the CSAI-2 with bowlers and concluded that multidimensional theory is insufficient in examining IZOF theory. Finally, Annesi (1997) recently found IZOF recall accuracy, using the CSAI-2, was weak enough to suggest the CSAI-2 may not be useful for IZOF research and practice when optimal affect is retrospectively assessed.

Emotions have become important in IZOF assessment, as optimal athlete performance is a function of both positive and negative affect (PNA). In addition, retrospective recall has remained important in crafting an individual's IZOF (Hanin, 1995; Hanin & Syrja, 1995, 1996) due to invasiveness (Harger & Raglin, 1994) and time investment (Hanin & Syrja, 1996) of actual assessment. While most IZOF study has focused on athletes' individual skill assessment, examination of athletes' zones across tasks has been largely unexamined. Hanin (1994) noted that PNA is triggered by one's appraisal that their performance goal will be achieved and PNA was assessed using athletes' self-generated items from the positive and negative affect scales (PANAS) (Watson, Clark, & Tellegen, 1988). Indeed, use of the PANAS may be more relevant to affective performance states in light of the CSAI-2 results.

IZOF theory has been examined primarily in the field where variables such as injuries, staleness, teammates' performance, and confidence based on prior performance efficacy are capable of altering performances. However, IZOF research has not been conducted in a laboratory setting where such variables may be more rigorously controlled. The current study established a controlled, direct examination of IZOF theory while maintaining a competitive framework in a contrived "competition" for external validity, thus addressing previous methodological concerns (Gould et al., 1993; Randle & Weinberg, 1997). The present design examined whether participants' actual optimal PNA was within their IZOF using a controlled setting, while also examining optimal PNA similarity across two similar sport tasks. The PANAS has not been previously used in direct IZOF assessment. It is minimally intrusive, a noted concern with IZOF assessment (Annesi, 1997), and may be more accurate for less skilled athletes who have a weaker sense fo r optimal PNA patterns than elite athletes (Hanin, 1994).

There were four purposes of this study. These were (a) to determine if a relationship exists between being within an individual zone of optimal functioning (IZOF), relative to PNA, and performance on two motor tasks; (b) to determine if a relationship exists between type of motor task and optimal PNA associated with best performance; (c) to determine if a difference fexists between actual and recalled optimal PNA scores, and (d) to determine if actual optimal PNA is correlated with recalled optimal PNA scores. The hypotheses tested were:

1. Best performance would occur when a participant's actualoptimal PNA was within his individual zone of optimal functioning.

2. No difference would be observed between a participant's actual optimal PNA for two different motor tasks.

3. No difference would be observed between a participant's actual and recalled optimal PNA.

4. There would be a significant correlation between actual and recalled optimal PNA and the variance accounted for would be greater than fifty percent.

The rationale for using actual measures of optimal affect in hypotheses one and two was that when it can be obtained, the actual measurement should be used since it is considered the criterion (Hanin, 1984).

Method

Participants

Fifty-five males (M = 23.82, SD = 4.07) from a large midwest university volunteered. All participants were moderately skilled in the tasks in this study: basketball shooting and football throwing. "Moderately skilled" for this study was defined as participants who had been involved in organized participation in the sports represented by the skills (e.g. high school, intramural, collegiate competition) and through responses to whether subjects felt they could make 4 out of 10 basketball shots from a location between the foul line and the key, and whether subjects felt they could accurately throw a football 20 yards to a stationary target. All selected participants endorsed the sport task competency statements. It was felt that if subjects met the experience criteria they had the minimal skill level necessary for inclusion to this study.

Questionnaire

Positive and negative affect were assessed with the positive and negative affect schedule (PANAS) (Watson, Clark, & Tellegen, 1988). The PANAS consists of two 10-item scales which measure affect using a five-point likert format. Briefly, Positive Affect (PA) reflects one's enthusiasm, activity, and alertness. High PA is a state of high energy and pleasurable engagement, whereas low PA is characterized by sadness and lethargy. Negative Affect (NA) is a general dimension of subjective distress that includes anger, contempt, disgust, guilt, fear, and nervousness, with low NA being a state of calmness. Scores on the PANAS are on a five-point likert scale, ranging from a low of 10 to a high of 50. Alpha reliabilities for the PANAS are all acceptably high ranging from .86 to .90 for PA and from .84 to .87 for NA. Examples of PA items are; interested, excited, and strong. Examples of NA items are distressed, upset, and scared (Watson et al., 1988). Zones were created by taking affect scores (positive and negative) associated with one's best performance scores and adding and subtracting one-half standard deviations from these two scores, respectively (Hanin, 1986). Standard deviations from which the positive affect IZOF and negative affect IZOF ranges were created were based upon sample means and standard deviations for affect scores across experimental trials. Zones for each sport task were then formed using intraindividualized Z-scores based upon 12 trials. These zones were participant-specific, as they were based upon each participants' individual mean and standard deviation for each task.

Sport Tasks

The sport tasks for this study were selected after carefully considering several alternative tasks. The basketball shooting task was selected because of scoring objectivity, it is a well-learned skill for many college-aged males, and performance feedback is immediate. A second task was selected that also exhibited similar characteristics. The football throwing task was also objective and similar to the basketball task in that a "hit or miss" criterion was used. Both tasks were given a time constraint so that activation levels would be elevated from a low baseline.

Timed basketball shooting task. Prior to each experimental trial, participants were given a standardized three minute warm-up shooting period. Performance was measured by the number of shots made from either the right or left side of the foul lane where foul line and lane meet. This option was provided so as not to "penalize" participants who may have been more comfortable shooting from one side or the other. A cart of balls was placed at the back of the foul line circle. Participants then took a ball, one at a time, and moved to shoot from either location. Optimal performance was most shots made in a given 60 sec trial.

Tuned football throwing task. Prior to each trial participants were given a standardized three minute throwing period. Performance was measured by the number of balls thrown which struck a target 9.14 m (10 yards) from the participant. For each ball thrown, subjects were required to start from a line situated 7.31 m from the target, take a football from a rack and drop back 1.83 m so that they were 9.14 m from the target when each throw was initiated. The rationale for this movement was to make the task similar to the shooting task as opposed to static throwing. A square target (.5 m X .5 m) was taped to a hanging mat with the bottom of the target 1.5 m from the floor. Successful throws were balls which made at least partial contact with the target. Optimal performance was the most balls accurately thrown at the target in a given 60 sec trial.

Procedures

Both tasks were performed by all participants so that participants' optimal PNA scores associated with optimal performance could be compared across sport task. To minimize learning effects and facilitate habituation, participants underwent two separate 10-min practice sessions on separate days prior to experimental trials, in which they performed both tasks under timed conditions. In any given trial, subjects performed both timed tasks. There were twelve separate experimental trials for each task and all sessions were separated by at least 24 hours. Each trial began with the standardized warm-up, after which individual participants were administered the PANAS immediately prior to their performance, with instructions asking how they felt at that moment about the upcoming trial. Participants were tested in triads to facilitate competitiveness and hence external validity.

Affect Manipulation

To obtain a greater range of affect and further external validity, an attempt was made to manipulate participants' positive and negative affect by assigning varying performance goals across participants' experimental trials to create greater dispersal of precompetitive affect scores. Hanin (1994) has indicated that PNA is triggered by a person's appraisal of goal achievement probability. Goal difficulty has also been noted to effect motivation relative to a sport task which, in turn, impacts anxiety and emotions (Thill & Brunel, 1995). Therefore, the rationale for affect manipulation was to increase external validity by varying the participants' appraisal of goal-achievement probability based upon the perceived difficulty of the goal.

Using each participant's performance score average from the first three trials (baseline average), three goal-setting conditions were randomly distributed throughout the remaining nine trials, in which participants were told to equal their baseline score (easy condition), to attain 25% above their baseline condition (moderate condition) or to attain 50% above their baseline score (difficult condition). For goal-setting conditions, participants were informed only of the minimal score needed and performance order was counterbalanced across trials.

Data Analysis

Creation of individualized zones of optimal functioning (IZOF). Individual zones were formed by locating the precompetitive PNA which was linked to each individual's best performance score in each sport task. Each individual's IZOF was their obtained positive/negative affect score plus-and-minus their individualized one-half standard deviation. Thus, for each task, the single resultant best performance score was selected as optimal and performance scores for this trial were used to craft the zone for that individual framed by one half of each participant's individualized standard deviation.

Ipsative z-scores. Both performance scores and IZOFs were intraindividualized to examine the idiographic relationship between precompetitive PNA and performance. Performance scores were intraindividualized by transforming raw performance scores into ipsative z-scores based on each participant's own twelve-trial performance mean and standard deviations. Positive and negative affect scores were not intraindividualized but the IZOFs were, in that once an affect score was identified with a best performance score, the participant's own standard deviation was used to create their IZOF.

Analysis of data and research hypotheses. Hypothesis one (1) was analyzed using simple one-factor ANOVAs. The three levels of the factor were "in", "above", or "below" the IZOF. The hypothesis that precompetitive PNA associated with best performance would fall within one's IZOF was supported if overall F-ratios were significant and groups means indicated significantly higher performance scores for the "in" group compared to the other two groups. Hypothesis two (2) was analyzed using one-way ANOVA procedures with sport task serving as the independent variable and actual optimal positive/negative affect serving as the dependent variable. The hypothesis that no difference would be observed between a participant's actual optimal PNA for two sport tasks was supported if overall F-ratios were not significant. In Hypothesis three (3), recall and actual assessment of optimal positive/negative affect served as the independent variable (time of measurement), while affect score served as the dependent variable. Separat e repeated measures ANOVA procedures were used in the analysis for basketball (contrasting actual and recall) and football (contrasting actual and recall). The hypothesis of no difference between a participant's actual and recalled optimal PNA was supported if overall F-ratios were not significant. In Hypothesis four (4), correlation techniques were used to study the relationship between actual optimal PNA and recalled optimal PNA. A significant and meaningful relationship would mean actual optimal affect is associated with recalled optimal affect even though the two scores may differ significantly from one another.

Results

Affect Manipulation

Affect manipulation through goal-setting conditions was only partially effective. In the shooting task, a significant main effect resulted for condition, with PA, F (3,162) = 5.03, p =.002, and NA, F(3,162) = 9.71, p =.000l. In football, results revealed a significant main effect for condition with NA, F(3,162) = 5.45, p =.001. However, in all analyses, follow-up tests revealed significant differences between baseline and experimental conditions, but no differences among experimental conditions. Means and standard deviations for the four conditions are shown in Table 1. However, while the manipulation strategy did not result in differences among the experimental conditions, a fluctuation in affect did occur. For example, the PA basketball mean and standard deviation for the difficult condition was 33.99 and 8.84 respectively, indicating that 68% of the scores ranged from approximately 25 to 43, a fairly large spread.

Performance and the IZOF Hypothesis

This analysis tested the hypothesis that performance would be best when PNA scores fall within one's pre-determined IZOF and intraindividualized to more accurately examine the idiographic IZOF approach (Hanin, 1995).

Basketball shooting. A summary of omnibus and planned comparisons for testing the IZOF hypothesis is shown in Table 2. In addition, mean intraindividualized performance scores and standard deviations associated with groups (below, in, above IZOF) are shown in Table 3.

For PA, a one-way ANOVA on the three in or out-of-zone groups resulted in a significant group effect, F(2,657) = 7.58, p =.0006. The planned comparison for in versus out-of-zone performance was also significant indicating performance was significantly higher when PA scores fell within one's IZOF (p [less than].01, [Eta.sup.2] = .02). The planned comparison for above versus below-IZOF performance was not significant. For NA, a one-way ANOVA on the three in or out-of-zone groups also reached significance, F(2,657) = 3.15, p =.04. The planned comparison for in versus out-of-zone performance was also significant indicating performance was significantly higher when negative affect scores fell within one's IZOF (p [less than].05, [Eta.sup.2] = .009). The planned comparison for above versus below-IZOF performance was not significant.

Football throwing. For PA, a one-way ANOVA on the three in or out-of-zone groups resulted in a significant group effect, F(2,657) = 7.67, p =.0005. The planned comparison for in versus out-of-zone performance was also significant, indicating that performance was significantly higher when PA scores fell within one's IZOF (p [less than].01, [Eta.sup.2] = .02). The planned comparison for above versus below-IZOF performance was not significant. For NA, ANOVA results on the in or out-of-zone groups were not significant, F(2,657)=2.51, p=.08. Planned comparisons for in versus out-of-zone performance, as well as for above versus below-IZOF performance were also not significant.

Comparison of Participants' Actual Optimal PNA for Two Sport Tasks

ANOVA results indicated no significant difference between basketball and football, F (1,54)=.00, p=.95 for PA. Similarly for NA, ANOVA results indicated no significant differences between basketball and football, F(1,54)=.2l,p=.65. Thus, hypothesis two was supported in that no differences resulted between one's actual optimal PNA for the two tasks. The means and standard deviations for actual optimal PA were M = 33.21, SD 8.79 (basketball) and M =33.25, SD = 8.61 (football). The means and standard deviations for actual optimal NA were M =13.71, SD = 5.28 (basketball) and M = 13.50, SD =4.33 (football).

Comparison of Actual and Recalled Optimal PNA

Table 4 shows positive and negative affect means and standard deviations for sport tasks and assessment method. Separate repeated measures ANOVAs were performed for basketball and football (comparing actual and recall methods). In basketball there was a significant difference between one's actual and recalled PA, F( 1,54) =7.50, p =.008, suggesting that one's optimal PA recall score was significantly different from one's actual optimal PA score. In the football task, a repeated measures ANOVA revealed a significant difference between one's actual and recalled PA, F(1,54) = 12.53, p =.0008, suggesting that one's recall of optimal PA for the throwing task was significantly different (higher) than actual optimal PA.

Separate repeated measures ANOVAs for NA in basketball and football resulted in nonsignificant differences between actual and recalled optimal actual scores, F(1,54)=1.56, p=.22 (basketball); F(1,54)=2.16, p=.15 (football). These results suggest that for NA, recall is similar to actual affect associated with optimal performance, regardless of sport task.

Relationship Between Actual and Recalled Optimal PNA

Hypothesis four hypothesized that the correlation between actual and recalled optimal PNA would be significant, and the variance accounted for would be greater than fifty percent. Because differences between recall and actual positive affect were in the same direction for sport tasks, and no differences were found for the assessment methods for negative affect in either task, actual and recalled positive and negative affect were collapsed across sport task for this analysis. The correlation between actual optimal PA and recalled optimal PA was r = .76 ([r.sup.2]=.58), p=.0001, while the correlation between actual optimal NA and recalled NA was r=.74, ([r.sup.2]=.55), p=.0001, suggesting a moderate relationship between actual optimal affect and recalled optimal affect.

Discussion

Utilizing a laboratory-like setting, this study attempted to: (a) determine if a relationship existed between being within one's zone of optimal functioning relative to PNA, and performance on two motor tasks; (b) determine whether a relationship existed between motor task and optimal PNA associated with best performance, (c) determine if a difference existed between actual and recalled optimal PNA scores, and (d) determine if actual optimal PNA is correlated with recalled optimal PNA scores. While PNA manipulation through task goal setting was unsuccessful, standard deviations from goal conditions indicated there was sufficient fluctuation of PNA under the somewhat artificial conditions of the lab tasks.

Performance and the Test of the IZOF Hypothesis

While contrasts for in versus out of zone affect scores indicated statistically significant differences in three out of the four conditions (basketball PA, p[less than].O00l, basketball NA, p[less than].0152, and football PA, p[less than].0001), the corresponding effect sizes ranged from very small to minimal (basketball PA, [ETA.sup.2] =.02, basketball NA, [ETA.sup.2] =009, and football PA, [ETA.sup.2] =.02) (Rosenthal & Rosenow, 1991). These statistically significant differences were due in part in the large number of observations recorded (660 observations for each task). However, it may also be argued that these results indicate a lack of support for IZOF hypothesis due to several reasons.

With the large number of observations, stronger support for IZOF theory would have been indicated in larger effect sizes. Because of the large number of observations and low effect sizes, it may be argued that it did not make much difference whether or not participants were in or out of their IZOF with respect to PA and NA. This finding is not unlike recent studies which have not found support for IZOF theory within a multidimensional anxiety context (Annesi, 1997; Dennis, Bartsokas, Lewthwaite, & Palin, 1993; Randle & Weinberg, 1997; Thelwell & Maynard, 1998; & Woodman, Albinson & Hardy, 1997). Specifically, studies which have used cognitive and somatic anxiety (CSAI-2) to formulate IZOFs have found no differences in performance across zone position (Dennis et al., 1993), or interactions between cognitive and somatic anxiety which show more support for catastrophe theory (Hardy, 1990) than IZOF theory (Krane, 1993; Randle & Weinberg, 1997; Thelwell & Maynard, 1998; Woodman, Albinson, and Hardy, 1997). Randl e and Weinberg (1997) examined objective and subjective performance and found that when examining players' somatic and cognitive anxiety, objective performance was significantly better outside of players' IZOF than inside. Thelwell and Maynard (1998) addressed directionality of anxiety within IZOF theory, and found low standardized performance scores when athletes were within their cognitive and somatic zones, offering little support for IZOF theory. Thus, as has been indicated within the anxiety literature (Jones & Swain, 1992), it may be necessary to examine athletes' functional interpretation of PNA direction as well as intensity.

Another reason for the results for this hypothesis, similar to Thelwell and Maynard (1998), may have been that participants were unable to experience an optimal zone which was accompanied by an optimal performance over their 12 trials. The contrived competitive setting may have been too artificial to detect meaningful performance differences between in and out of zone conditions. In addition, since these were nonelite athletes, they may not have been able to recognize and regulate optimal preperformance emotions. This would make participants unaware of optimal zones changing depending upon task difficulty, resulting in good scores occurring across varying levels of positive and negative affect. Results from this hypothesis point out the importance for investigators to examine effect size as well as tests of significance within IZOF theory. Significant differences were found between in and out of zone conditions, yet the amount of variance in performance accounted for zone position was small. Larger effect siz es are needed to warrant more confidence in this hypothesis within IZOF theory.

Comparison of Actual Optimal PNA for Two Sport Tasks

The hypothesis that no difference would be observed across a participant's optimal PNA for two tasks was supported. Because these two tasks were deemed similar in motor skill requirements, it was hypothesized that optimal PNA would remain constant across both tasks. This element of one's IZOF determination may be due to one's functional interpretation of content, intensity, and content of affect associated with the sport task (Hanin & Syrja, 1997) rather than sport task itself. Thus, it appears that one's optimal PNA is stable within an idiographic context, and does not vary as a function of two similar motor tasks.

Comparison of Actual and Recalled IZOF Assessment

This hypothesis was supported for negative affect and reinforces the notion that actual and recalled precompetition anxiety are similar (Harger & Raglin, 1994). For positive affect, however, actual optimal PA was significantly higher than recalled optimal PA, regardless of the task. Optimal PA differences may have been due to the differential interpretation with which participants viewed the experimental paradigm, in that some participants may have assessed precompetitive affect with the competitive paradigm in mind, yet recalled optimal precompetitive affect with the laboratory paradigm in mind. The greater accuracy of PNA stimulus lists (Hanin, 1997), may explain differences between actual and recalled optimal PA, in that participants' emotion content was not present in the PANAS. The variability of recall optimal PA may have also been due to the fact that these participants were non-elite athletes and had not developed a keen sense for determining their IZOF with respect to positive affect.

Relationship Between Actual and Recalled Optimal PNA

The hypothesis that moderately high and significant correlations exist between actual and recalled optimal PNA was supported. This occurred despite the observation that for optimal PA, recall scores were significantly larger than actual scores, for both tasks. The practical value of this observation is that actual optimal affect can be predicted from recalled optimal affect, even though recalled affect may overestimate actual affect, and does so in the case of PA. Thus, even though differences may be apparent, recall of optimal affect may be useful in predicting actual precompetitive PNA.

General Conclusions

Most IZOF theory research has been conducted on collegiate (Gould et al., 1993; Raglin & Turner, 1993; Turner & Raglin, 1990) or elite athletes (Hanin, 1986; Prapevessis & Grove, 1991; Hanin & Syrja, 1995), while participants in this study were, for the most part, recreational athletes. While it has been suggested that less-skilled athletes may be potentially unable to identify or regulate their optimal zones (Thelwell & Maynard, 1998), some support for IZOF theory was provided and actual optimal PNA and recalled optimal PNA were correlated with one another even though they were different. Findings from these less-elite athletes indicate the importance of examining effect size as well as statistical significance within IZOF theory. For these participants, there was not much meaningful difference in terms of performance whether they were in or out of their IZOF with respect to PA and NA. IZOF theory's hypothesis that athletes perform better when inside one's predetermined IZOF may be limited to highly-skilled athletes who have refined sense of emotional state associated with optimal performance.

Gould et al. (1993) have noted a limitation of previous studies was their omission of sport-specific measures. Since the PANAS is a general affect measure, it may have been less sensitive to pertinent affect in an athletic context, thus, a limitation of this study. This was supported in that some participants indicated that certain PANAS items were irrelevant to individual appraisal of athletic performance (i.e. guilty, proud) and these items may have contributed to the lack of affect manipulation. In addition, affect direction was not assessed and the impact of felt interpretation (Jones & Hanton, 1996) was unknown. This distinction may bear direct relevance on IZOF formation and appears to support Hanin's (1992) idiographic IZOF profile method. Future researchers should incorporate subjective measures of performance in examining the IZOF hypothesis, as they may be more relevant to idiographic IZOF assessment (Randle & Weinberg, 1997; Salminen, Liukkon, Hanin, & Hyvonen, 1995).

Future researchers examining IZOF theory with less-skilled athletes will need to examine both effect size as well as tests of significance in order to determine how meaningful IZOF theory is with these populations. Since it has been proposed that IZOF tenets are applicable to any naturalistic performance setting, (Hanin, 1994) further study of IZOF theory with less-skilled athletes might provide support for this contention. In summary, examination of IZOF tenets in a controlled setting, while limited in their generalizability, are nonetheless important because constructs may be rigorously examined while preserving a competitive environment and directly comparing such important aspects as actual and retrospective recall methods. Some statistical support was found for IZOF replication within a controlled setting, however the low amount of variance accounted for suggests limitations for IZOF theory.

References

Annesi, J.J. (1997). Three-dimensional state anxiety recall implications for individual zone of optimal functioning research and application. The Sport Psychologist, 11, 43-52.

Brewer, B.W., Van Raalte, J.L, Linder, D.E., & Van Raalte, N.S. (1991). Peak performance and the perils of retrospective introspection. Journal of Sport and Exercise Psychology, 8, 227-238.

Dennis, K.M., Bartsokas, T., Lewthwaite, R., & Palin, D. (1993). Relationship between CSAI-2 subscales and performance in youth athletics: The zone of optimal functioning. Medicine and Science in Sports and Exercise, 25, Suppl., S155.

Gould, D., Tuffey, S., Hardy, L., & Lochbaum, M. (1993). Multidimensional state anxiety and middle distance running performance: An exploratory examination of Hanin's (1980) Zone of Optimal Functioning Hypothesis. Journal of Applied Sport Psychology, 5, 85-95.

Hanin, Y.L., (1980). A study of anxiety in sports. In W.F. Straub (Ed.), Sport psychology: An analysis of athlete behavior. (236-249). New York: Movement Publications.

Hanin, Y.L. (1986). State-trait anxiety research on sports in the U.S.S.R. In C.D. Spielberger & R. Dias-Guerrero (Eds.), Cross cultural anxiety. (45-64). Washington D.C.: Hemisphere.

Hanin, Y.L. (1994, July). Performance affect in top sports: Prediction and monitoring. A paper presented at the 23rd Congress of Applied Psychology, Madrid, Spain.

Hanin, Y.L. (1995). Assessment of performance emotions in sport. In J. Vitasalo & U. Kujala (Eds.). The way to win. Proceedings from the International Congress on Applied Research in Sports held in Helsinki, Finland on August 9-11, 1994 (pp. 311-316). Helsinki, Finland: The Finnish Society for Research in Sport and Physical Education.

Hanin, Y.L. (1997). Emotions and athletic performance: Individual zones of optimal functioning model. European Yearbook of Sport Psychology J, 29-72.

Hanin, Y.L., & Syrja, P. (1995). Performance affect in soccer players: An application of the IZOF model. International Journal of Sports Medicine, 16, 260-265.

Hanin, Y.L., & Syrja, P. (1996). Predicted, actual, and recalled affect in Olympic-level soccer players: Idiographic assessments on individualized scales. Journal of Sport and Exercise Psychology, 18, 325-335.

Hanin, Y.L., & Syrja, P. (1997). Optimal emotions in elite cross-country skiers. In E. Muller, H. Schwameder, E. Kornexl, & C. Raschner (Eds.), Science and skiing (pp. 408-419). London: E. & F.N. Spon.

Hardy, L. (1990). A catastrophe model of anxiety and performance. In J.G. Jones, & L. Hardy (Eds.), Stress and performance in sport. Chichester, UK. Wiley. (81-106).

Harger, G.J., & Raglin, J.S. (1994). Correspondence between actual and recalled precompetitive anxiety in collegiate track and field athletes. Journal of Sport and Exercise Psychology, 16, 206-211.

Imlay, G.J., Carda, R.D., Stanbrough, M.E., Drieling, A.M., O'Connor, P.J. (1995). Anxiety and athletic performance: A test of zone of optimal functioning theory. International Journal of Sport Psychology, 26,295-306.

Jokela, M., & Hanin, Y.L. (1997). Does the IZOF model discriminate between successful and less successful athletes? A meta-analysis. A paper presented at the annual congress of European College of Sport Science in Copenhagen, Denmark, August 20-23.

Jones, G., & Hanton, S. (1996). Interpretation of competitive anxiety symptoms and goal attainment expectancies. Journal of Sport and Exercise Psychology, 18, 144-157.

Jones, G., & Swain, A. (1992). Intensity and direction as dimensions of competitive state anxiety and relationships with competitiveness. Perceptual and Motor Skills, 74,467-472.

Krane, V. (1993). A practical application of the anxiety-athletic performance relationship: The zone of optimal functioning hypothesis. The Sport Psychologist, 7, 113-126.

Martens, R., Vealey, R.S.,& Burton, D. (1990). Competitive anxiety in sport. Champaign, IL: Human Kinetics.

Morgan, W.P., O'Connor, P.J., Sparling, P.B., & Pate, R.R. (1987). Psychological characterization of the elite female runner. International Journal of Sports Medicine, 8, 124-131.

Prapevessis, H., & Grove, J.R. (1991). Precompetitve emotions and shooting performance: The mental health and zone of optimal functioning models. The Sport Psychologist, 5, 223-234.

Raglin, J.S., Morgan, W.P., & Wise, K.J. (1990). Pre-competition anxiety and performance in female high school swimmers: A test of optimal functioning theory. International Journal of Sports Medicine, 11, 171-178.

Raglin, J.S., & Turner, P.E. (1993). Anxiety and performance in track and field athletes: A comparison of the inverted-U hypothesis with zone of optimal functioning theory. Personality and Individual Differences, 14, 163-171.

Randle, S., & Weinberg, R. (1997). Multidimensional anxiety and performance: An exploratory examination of the zone of optimal functioning hypothesis. The Sport Psychologist, 11, 160-174.

Rosethal, R., & Rosenow, R.L. (1991). Essentials of behavioral research: Methods and data analysis. New York: McGraw-Hill.

Salminen, S., Liukkonen, J., & Hyvsnen A. (1995). Anxiety and athletic performance of Finnish athletes: Application of the zone of optimal functioning model. Personality and Individual Differences, 19, 725-729.

Spielberger, C.D. (1983). Manual for the State-Trait Anxiety Inventory (Form Y). Palo Alto, CA: Consulting Psychologists Press.

Tenebaum, G., & Furst, D.M. (1984). Similarities between retrospective and actual anxiety states. The Journal of Psychology, 119, 185-191.

Thelwell, R.C. & Maynard, I.W. (1998). Anxiety-performance relationships in criketeers: Testing the zone of optimal functioning hypothesis. Perceptual and Motor Skills, 87, 675-689.

Thill, E.E., & Brunel, P. (1995). Cognitive theories of motivation in sport. In S.J.H. Biddle (Ed.) European perspectives on exercise and sport psychology. (198-217). Champaign, IL: Human Kinetics.

Turner, P.E., & Raglin, J.S. (1996). Variability in precompetition anxiety and performance in college track and field athletes. Medicine and Science in Sports and Exercise, 28, 378-385.

Watson, D., Clark, L.A., & Tellegen, A. (1988). Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology, 54, 1063-1070.

Woodman, T., Albinson, J.G., & Hardy, L. (1997). An investigation of the zones of optimal functioning hypothesis within a multidimensional framework. Journal of Sport and Exercise Psychology 19, 131-141.

 Affect Means and Standard Deviations Across
 Goal Manipulations For Sport Task
Sport Baseline Easy Moderate Difficult
Basketball
 PA 35.33 33.71 34.03 33.99
 (7.50) [*] (8.39) (8.34) (8.84)
 NA 15.03 13.60 13.82 13.43
 (5.39) (5.45) (4.73) (4.73)
Football
 PA 34.31 33.53 33.81 33.86
 (7.50) (8.36) (7.52) (8.45)
 NA 14.80 13.67 13.88 14.14
 (5.25) (5.34) (5.21) (5.47)
(*.)Standard deviations are in parentheses.
 Summary Table For The Omnibus Test of The
 ZOF Hypothesis and Planned Comparisons
 Omnibus Contrast 1 [#]
 (In Versus Out)
Sport PNA F p [ETA.sup.2] F p
Basketball PA 7.58 .0006 [**] .0200 14.94 .0001 [**]
 NA 3.15 .0400 [*] .0090 5.93 .0152 [*]
Football PA 7.67 .0005 [**] .0200 14.78 .0001 [**]
 NA 2.51 .0800 .0080 1.75 .1864
 Contrast 2
 (Above Versus Below)
Sport [ETA.sup.2] F p [ETA.sup.2]
Basketball .0200 .02 .8937 .0000
 .0090 .14 .7040 .0002
Football .0200 .18 .6710 .0003
 .0030 2.84 .0926 .0040
(*.)p[less than].05,
(**.)p[less than].01
(#.)Differences reflect higher performance scores for in IZOF group.
 Number of Observations, Means, and Standard, Deviations
 Associated with Intraindividualized Performance Scores
Sport Task ENA Position N [*] M SD
 Below 214 8.72 4.14
 PA In 281 9.62 4.65
 Above 165 8.04 3.46
Basketball
 Below 132 7.79 3.80
 NA In 358 9.73 4.54
 Above 170 8.16 3.61
 Below 211 9.68 3.29
 PA In 286 10.03 3.91
 Above 163 9.10 3.63
Football
 Below 130 9.60 3.65
 NA In 367 9.88 3.77
 Above 163 9.34 3.43
(*.)Represents the number of performance scores associated
with PNA that fell below, in, and above the ZOF.
 Positive and Negative Affect Means and
 Standard Deviations For Sport Skills and
 Assesment Methods
 Positive Affect Negative Affect
Sport Assessment Mothod M SD M SD
Basketball Recall [35.36.sub.a] 8.24 [13.13.sub.a] 4.21
 Actual [33.21.sub.b] 8.79 [13.71.sub.a] 5.28
Football Recall [35.93.sub.a] 7.17 [12.87.sub.a] 4.06
 Actual [33.25.sub.b] 8.61 [13.50.sub.a] 4.33
Note: Within sport and affect cells, different subscripts
denote significant differences between means.