Vector Approach to Integral Evaluating of Sportsmen Functional States

Available online at www.sciencedirect.com

ScienceDirect Procedia - Social and Behavioral Sciences 86 (2013) 610 – 614

V Congress of Russian Psychological Society

Vector Approach to Integral Evaluating of Sportsmen Functional States Vyacheslav V. Lebedeva*, Sergey A. Isaycheva, Alexandr M. Chernorizova, Yuriy P. Zinchenkoa a

Faculty of Psychology, Lomonosov Moscow State University,Mokhovaya str. 11-5, Moscow,GSP-1, 125009, Russia

Abstract One of the urgent problems of modern sports psychology is constructing the methods for objective diagnosis of the functional state of athletes (FSS). Monitoring the FSS allows to track and control any deviations in FSS from the individual optimal FSS of the athlete and to predict the reduction or increase the efficiency of his/her activity. For the organization of such kind of control it's necessary to develop an integrated assessment of FSS based on the complex psychophysiological data. In this study, an original algorithm for integral assessment of FSS was elaborated, which was based on the method of kk means clustering. The results showed that the proposed algorithm can differentiate resting functional state, the state for the optimal performance of the test task, and the functional state of physiological and psychological stress. © 2013 2013 The Published by Elsevier Ltd. Ltd. TheAuthors. Authors. Published by Elsevier Selection and/or under responsibility of Russian Psychological Society Society. Selection and/orpeer-review peer-review under responsibility of Russian Psychological Keywords: functional state of the athlete, complex evaluation functional state, modeling of functional states, algorithm for the clustering kmeans, stress.

1.

Introduction

In modern sports psychology great attention is paid to the objective assessment of the functional state of athletes (FSS) [1], [2]. The dynamics of the FSS is caused by many external and internal factors reflecting effects of physical and social environment, individual characteristics of the subject, the specific conditions and the implementation of a specific sport [3]. Among the continuum of FSS one can select the optimal FS of an athlete (OFSS), in which the athlete reaches the maximal efficiency with a relative minimum of physiological costs. In general, OFSS can be viewed as a dynamic system of physiological, psychological and behavioral indicators of the activity of the morphological structures of different levels of the organization, implementing

* Corresponding author. Tel.: +7 495 629 37 23; fax: +7 495 629 37 23. E-mail address: [email protected]

1877-0428 © 2013 The Authors. Published by Elsevier Ltd. Selection and/or peer-review under responsibility of Russian Psychological Society doi:10.1016/j.sbspro.2013.08.622

Vyacheslav V. Lebedev et al. / Procedia - Social and Behavioral Sciences 86 (2013) 610 – 614

human behavior in this kind of sports activity [3]. In the Euro-American sports psychology there the OFSS is on the problem of the relationship between the emotional sphere, stress athlete and his professional efficiency [4], [5]. The main problem, remaining to be open, concerns the development of FSS integral evaluation on the base of the complex psychophysiological data for objective differentiation of FSS. Mathematically, this integrated assessment can be represented as a single vector with elements that include a set of three sets of indicators: physiological, psychophysiological, and psychological. Selection of indicators for each block is defined by the specific sports direction and reflects a set of professionally important qualities needed to facilitate a successful competition in this sport. Physiological and psychological characteristics of the athletes are quite stable and constant [5] - [7], and depending on the degree of importance they are attributed to necessary or compensatory professional qualities (PQ) [8], [9]. Their impact on the effectiveness and success of the sport activity is detected in the comparative group studies. The most dynamic indicators are indicators of psychophysiological block, as they reflect the value of functional system formed during training and competition and realizing a specific sports activity. The main purpose of this study was to develop a comprehensive integrated assessment OFSS based on the measurement of the complex psychophysiological indicators, registered in various experimentally simulated functional states. 2. Method 1.1. Structure The method of objective diagnosis the OFSS of athletes was developed and experimentally tested. The technique involves two steps: the first stage of a pilot study included the modeling FSS, registration and selection of the most important psychophysiological indicators of FSS; at the second stage the development and testing of trainee algorithm of diagnosis and differentiation of FSS based on a set of psychophysiological indicators were realized.

2.2. Test procedures that simulate the FS One of the most important PQ in the sport is to be able to concentrate at the right time and be able to mobilize a relatively long time [10], [11]. To test this PQ various parameters of attention are used [12], [13]. To create experimental situations, simulating various FSS a special computer program based on signal detection theory was developed and used. This program has great potential for configuring the presentation of stimulus items and allows creating different complexity tasks that require maximum concentration and mobilization of human attention. By changing the conditions of performance tests the different functional states (normal, optimal condition, psychological stress) can be modeled, and subjects were asked to perform four test tasks: 1) training to differentiate the allocation of targets from the background, 2) the adjustment test designed to determine the optimal level of the test for detection of target stimuli, 3) hold the optimal level of FSS for execution of the preceding tests designed the allocation of optimal functional state (OFS), 4) stress test aimed to provocate a nonoptimal functional state (NSF). In the latter case, the level of stress was modeled by increase in the rate of appearance of test objects and/or their similarity with the background objects. The program also allows evaluating the results of the test in each FSS. Key performance indicators were the probabilities of right detection of test stimulus and false alarms.

611

612

Vyacheslav V. Lebedev et al. / Procedia - Social and Behavioral Sciences 86 (2013) 610 – 614

In addition to the simulated FSS we added one more test, namely a significant exercise on a bicycle ergometer. The load on a bicycle ergometer was created individually for each subject separately, taking into account his age and body weight. That test made possible to distinguish one more FSS - a state of physical stress (FSSF). Therefore we studied the dynamics of the psychophysiological parameters for modeling the five functional states: QW - functional state of quiet wakefulness; - optimal functional state 1; - optimal functional state 2; - functional state of psychological stress; - functional state of physical stress. 2.3. Registration and selection of psychophysiological indicators During the execution of tests the following psychophysiological parameters were registered: electroencephalogram (EEG), electrooculogram (EOG), electrocardiogram (ECG), photoplethysmogram (FIG), electromyogram (EMG) and breathing. After the electrophysiological recordings were performed primary data were processed. The artifacts in all records were removed automatically using the special algorithm based on the data obtained through two channels EOG, ECG channel and the channel EMG. Thus, to estimate the FS we used data on the effectiveness of the test and psychophysiological indicators reflecting the physiological performance, i.e. the degree of stress body systems necessary for solving test problems. For individual or group diagnosis and differentiation of various FSS the most appropriate indicators were selected: 1) the parameters of muscle activity and autonomic nervous system - EOG BLINK, ECG RATE, RD RATE, RD RQ; 2) the parameters of the EEG: ALPHA / BETA, ALPHA%, BETA % (see Table 1). Table 1. A list of physiological parameters used for the classification of functional states. Signal

Parameters (features)

EOG BLINK

Averaged value of winking / min

ECG RATE

Frequency of cardiac beat

RD RATE

Number of breathing cycles /min

RD RQ

Relative portion of inhalation in breathing cycle

Alpha/Beta

Ratio of alfa-rhythm to beta-rhythm in EEG

Alpha%

Portion of the time when alfa-rhythm dominates in EEG

Beta%

Portion of the time when bets-rhythm dominates in EEG

2.4. Development of algorithms for processing and analysis of psychophysiological data The main objective of the study was to develop a trainee mathematical algorithm that can automatically on a set of physiological parameters differentiate optimal FSS from "non-optimal functional state," which may reflect a state of psychological or physiological stress, the state of fatigue or affect. To develop the trainee diagnostic algorithm FSS data were used to conduct the series of experiments with modeling FSS of athletes-fighters. Each FSS was estimated by the criteria of productivity assignments and characterized by specific pattern of the selected physiological parameters. The algorithm for differentiation of FSS was based on the method of clustering, which defines the rules for the classification of the FSS on the base of registered physiological parameters. To monitor changes of subject FSS over time the overall record, which ranged from 5 to 10 minutes for each of the simulated FSS, was divided into two-minute segments with a minute overlap. In each of these

Vyacheslav V. Lebedev et al. / Procedia - Social and Behavioral Sciences 86 (2013) 610 – 614

segments a parameter vector was constructed. Furthermore, the vectors of parameters were normalized and clustered using k-means algorithm. The normalization is necessary in order each parameter has identical mean and variance. We only used the normalization on average value because of the large number of peaks that distort the variance. For each component of the vector of parameters we calculated the median value and divided the corresponding components of vector by obtained value. Algorithm for clustering k-means produces the division N vectors in K {v j } clusters, where each vector corresponds to the cluster with the nearest mean. We used K = 5 clusters. K-means algorithm consists of the following steps: the choice of the initial cluster centers {mi } . We used random values; assigning each vector cluster number with the nearest center:

Si(t )

{v j : v j

mi(t )

mi(*t ) for all i* 1,..., K} .

vj

2. Calculation of the new average values

mi(t

1)

1 vj | Si(t ) | v j Si( t ) ( t 1)

. (t )

If for all i we have mi mi or if the number of iterations exceeded a threshold, then stop, otherwise go to step 2. A characteristic property of clustering is the selection of feature vectors with similar parameters in a single cluster. Thus, if in the performance of different tasks the subject's functional status changes, then it will be accounted for as a change in the cluster to which both the feature vectors belong to. 3. Results The study was performed at Moscow State University; it was approved by the Ethic Committee of the MSU Faculty of Psychology. Changes of functional states were attained using special scenarios for a computer system developed by the staff members of the MSU Faculty of Psychology. Data segmentation and identification of a functional state for each segment were performed manually by experts of the MSU Faculty of Psychology basing on a complex analysis of peripheral physiological data, EEG data and task performance parameters. The results suggest the adequacy of the experimental procedure designed to simulate different kinds of functional states. For a more subtle differentiation of FSS we divided time period in the OFS into two phases: OFS1 (a subject fulfills an easy task) and OFS2 (a subject fulfills a difficult task, because at this stage he/she needs to stay with a minimum of errors during 5 minutes). To assess the significance of differences between the studied FSS we used -Friedman criterion (Friedman test), which showed statistical significance of distinctions. Clustering results for all the subjects of the control group showed that k-means algorithm for the 7-component vector containing all the parameters mentioned above can clearly differentiate functional states FSQW, FSSP and FSSP in 70% of subjects. In 80% of the subjects we were able to differentiate only functional states FSQW and FSSF. As the number of used parameters was diminished from 7 to 4 (without EEG - ALPHA / BETA, ALPHA%, BETA%) the accuracy of distinguishing FSS significantly reduced. Using the 4 parameters for muscles and autonomic

613

614

Vyacheslav V. Lebedev et al. / Procedia - Social and Behavioral Sciences 86 (2013) 610 – 614

nervous system allows distinguishing only a dramatic change in the human FS (for example, to distinguish the state of physiological stress, FSSF, from state of rest, FSQW). 4. Conclusion Using a vector approach to the diagnosis and differentiation of athletes-wrestlers FS showed its effectiveness to detect negative functional states, in particular stress. For the construction of an integrated assessment of individual optimal functional state of athlete train k-means algorithm for the 7 component vectors in situations of relaxation (norm), optimal FSS and FSS of physical and psychological stress. 5. Data availability All data and the detail of the study are available upon request from the authors. Acknowledgements The research was supported by the grant RSSF -06-00218. The research used the equipment purchased in frames of the MSU Development Program. The authors thank the staff members of the MSU Faculty of Psychology for the valuable comments and discussions. References [1] Taylor J, Wilson GS. Applying sport psychology: four perspectives. Human Kinetics Publishers, 2005. [2] Hartley S. Peak Performance Every Time. Routledge; 2012. [3] Isaychev SA, Chernorizov AM, Korolev AD, Isaychev ES, Dubynin IA, Zakharov IM. The Psychophysiological Diagnostics of the Functional State of the Athlete. Preliminary Data. Psychology in Russia: State of the Art 2012, p. 244-268. [4] Hackfort, D. I. E. T. E. R. A conceptual framework and fundamental issues for investigating the development of peak performance in sports. Essential processes for attaining peak performance; 2006, p. 10-25. [5] Krane V, Williams JM. Psychological characteristics of peak performance. Applied sport psychology: Personal growth to peak performance 5; 2006, p. 207-227. [6] Johnson U, Ivarsson I. Psychological predictors of sport injuries among junior soccer players. Scandinavian journal of medicine & science in sports 21.1; 2011, p. 129-136. [7] Sheldon JP, Eccles J.S. Physical and psychological predictors of perceived ability in adult male and female tennis players. Journal of Applied Sport Psychology 17.1; 2005, p. 48-63. [8] Thomas JR, Nelson JK, Silverman S, Silverman SJ. Research methods in physical activity. Human Kinetics Publishers, 2010. [9] , Cooke C. Anthropometric, physiological and selection characteristics in high performance UK junior Rugby League players. Talent Development and Excellence 2; 2010, p. 193-207. [10] Kremer J, Moran AP. Pure sport: Practical sport psychology. Routledge, 2012. [11] Cusimano K, DeBoom N, Hartman C, Stratton RK. Focus. In J. Taylor & G. Wilson. (Eds.), Applying sport psychology: Four perspectives. United States of America: Human Kinetics; 2005, p. 51-63. [12] Fan J, McCandliss BD, Sommer T, Raz A, Posner MI. Testing the efficiency and independence of attentional networks. Journal of cognitive neuroscience, 14(3); 2002, p. 340-347. [13] Posner MI., Fan J. Attention as an organ system. Topics in integrative neuroscience; 2008, p. 31-61.