Behavioural Brain Research, 45 (1991) 29-36
Elsevier
29
BBR 01218
Relationship between mental imagery and sporting performance C. Deschaumes-Molinaro i, A. Dittmar z and E. Vernet-Maury 1 ~Laboratoire de Physiologie neurosensorielle, CNRS-University Claude Bernard]Lyon, Villeurbanne (France) and ?Laboratoire de Thermor~gulation et Energ~tique de l'exercice, CNRS URA 1341, Facult~ de Mddecine, Lyon (France).
(Received 21 August 1990) (Revised version received 29 May 1991) (Accepted 22 July 1991) Key words: Mental imagery; Autonomic nervous system; Performance; Concentration
Simultaneous measurement of six autonomic nervous system (ANS) variable responses during mental rehearsal of an action, makes it possible to draw a parallel between mental imagery of a task and its actual execution. The experiment was carried out in the field during precision shooting competitions and in the laboratory for imagery activity, on 22 subjects. Results show that there is similarity of ANS response in the three situations: the period of concentration prior to shooting, actual shooting and mental representation of shooting. The ratio formed by ANS response during concentration and imagery of actual firing tends towards the value one, therefore towards some identity. All subjects may be classified around this value; it is worth noting that subject distribution around this value corresponds to performance value. It seems that the better the subject, the closer his concentration/shooting or imagery/actual shooting ratio is to the theoretical value one. These results show the utmost importance of the quality ofmental representation for performance improvement. It can be supposed that subject classification above the theoretical value one corresponds to overflowingemotional reactivity in one of the two phases and that this interferes with accuracy; a placing below the theoretical value one shows a lack of similarity between mental representation and the action.
INTRODUCTION Imagery falls within the framework o f repetitive mental practice or mental training and designates mental representation o f the performance o f a m o t o r pattern without concomitant production of the muscular activity normally required for the act 4. Imagery, extensively studied in its psychological aspects, is however more difficult to approach physiologically, as measurements can be indirect only. N u m e r o u s measurements have been carried out by the mental c h r o n o m e t r y m e t h o d z5 but these do not account for the processes brought into play. Recent research shows the interaction existing between imagery and perception 2"~6"24. The numerous hypotheses on the functional similarity between imagery and perception t9'2~ go as far as to assimilate mental and physical practice 3"13'!4'32. Autonomic nervous system ( A N S ) effectors are activated by imagery, as testified by extensive research
Correspondence: E. Vernet-Maury, Laboratoire de Physiologie neurosensorielle, CNRS-Universit6 Claude Bernard/Lyon, F-69622 Villeurbanne codex, France. Fax: (33)78949585.
carried out on the heart rate. But additionally, these effectors are thought to be liable to give a modified response in one way or another as the result o f the effects and contents o f imagery t9"2~ In most cases, individual differences, which are the result o f the subjects' ability to imagine and make images vivid and accurate, are evidenced 8'1~ Recent methodological and conceptual progress on A N S functioning as a specific system permits the thought o f its study to objectify brain activity 9'26m. At this time, brain ideography methodologies, promising though they may be, do not make such objectivation possible on account o f their temporal limitations. On the other hand, study o f the A N S response sheds new light on the central nervous system ( C N S ) functioning, on account o f their m a n y interrelations. Original indices and captors developed by the Vernet-Maury/Dittmar team zg, have m a d e it possible to apprehend A N S functioning in a reliable and reproductible manner. Continuous and simultaneous measurement o f six A N S effectors on the same group o f subjects in two situations: one bearing on physical practice and the other on imagination o f this practice will make it possible to bring out the impact o f imagery on overall autonomic responses. It has been possible to c o m p a r e these two
30 situations and performance. The hypothesis may be put forward that with these top-level marksmen, the concentration phase (immediately prior to shooting) based on mental rehearsal of the oncoming shot is very close to that of laboratory imagery from the view point of cognitive process implication. From the results obtained, it may be possible to discuss the ideomotor hypothesis which states that voluntary movement is mediated by the image which becomes the subject of this movement by inner representation ~7"2"28. Knowing that a subject responds through a preferential channel, it is important to measure six channels simultaneously in order to obtain maximal response from all subjects ~8"3~ These individual differences therefore also require that each subject's results are considered individually.
MATERIALS A N D M E T H O D S Six ANS variables were selected to be quantified: electrodermal response: skin potential and resistance; thermovascular variables: skin blood flow (original sensor) and skin temperature; cardiorespiratory variables: instantaneous heart rate and original instantaneous respiratory frequency.
A+
sec
AI 6'5
0 5
15
30
5
15
30
C+
Skin resistance response Cutaneous resistance (Kf~) was recorded using 30-mm 2 unpolarizable round Capsulex electrodes placed on the second phalanx of the forefinger and of the median finger of the non-dominant hand, held by an adhesive tape. Resistance is measured with ! 5 ttA d.c. current. The skin potential and resistance measurements we carried out by means of a high rate common rejection mode differential amplifier. In the same way, recorder inputs were of the differential mode and resistance circuit supply was of the floating type. Skin resistance measuring current passes between the forefinger and the middle finger while the skin potential was measured between the hypothenar eminence and the inner side of the forearm. Any interference between skin potential and resistance was therefore eliminated, as there was no influence in the measuring circuits, there was no possible influence in the biological tissues. This fact was verified by simultaneous and single recording of these two variables.
o
sec
5
Y
A_
B
60 sec 0 5
B+
(~
Skb~ potential responses Skin potential (mV) was recorded using Beckman self-adhesive 78-mm 2 electrodes. Electrode placement was in compliance with the recommendations of Fowles et alJ 2. The active electrode was stuck to the hypothenar eminence of the subject's non-dominant hand (after alcohol-ether cleaning of the skin). The reference electrode was stuck 10 cm from the wrist on the equidistant line of the median plane and the outer extremity of the forearm. Electrodermal potential variations were measured by the SYDER code 6 which permits classification of elementary responses, according to their form, sign ( + or - ) and duration. As far as potential variations .were concerned, only this index permitted their analysis, the other measurements being less satisfactory or even redundant. According to this code, 3 positive and 3 negative skin potential forms were considered (cf. Fig. I, A, B and C potential forms).
15
w
30
60 sec
C_
6'0 s e c 0
5
15
30
60 Sec
Fig. 1. SYDER code (decoding system for electrodermal responses). A, B and C skin potential forms. These are schemata ofcurve envelopes. Durations can vary except for the A form which is never longer than 5 s. Some responses correspond to a combination of 2 or 3 forms (example: response classically described as diphasic).
31 The response amplitude to a stimulation following the law of initial value ~3, another more reliable index was defined. A parallel observation ofthermovascular indices made it possible to define the time during which the subject 'responds' to stimuli without referring to the initial value time. This ohmic perturbation duration(s) or OPD index of skin resistance reflects the emotional load of the stimulation27.
Superficial skin blood flow The skin blood flow was measured using an original system: the HEMATRON (Dittmar, CNRS/ANVAR Patent n ~ 85 15932). The HEMATRON system continuously measures the thermal conductivity of material and of living tissues using the principle ofthermie clearance 7. The transducer consists of a disc 25 mm in diameter and 4-mm thick which is fixed on the skin by an adhesive tape. The measuring face in contact with the skin is made up of 2 parts: the reference area at the periphery of the disc and the measuring area at the center of the disc. The difference between these 2 areas is measured using 16 thermocouple jt~nctions. A very low thermal inertia flat heater is located in the central part of the disc. A proportional, integrative and derivative device controls the heating power to maintain a constant difference of 2 ~ between the central area and the periphery. The size and the shape of the heater are designed to induce a thermal field in the capillary network. The power necessary to maintain the constant difference depends on skin blood flow: heat is transferred through the skin and washed out by the blood flow. At all times the electric power is proportional to the heat evacuated by the tissue blood flow. Blood flow in skin capillaries is submitted to mierochanges reflecting the variations of emotional load. Skin blood flow variations were measured: (i) by the difference (positive or negative) between the pre- and post-stimulation values expressed in W" m - I. oC - ~; (ii) by the duration of the perturbation of the oscillations.
Skin temperature This was measured by a low inertia thermistor (I0 K3 MCD2 Betatherm). A 4-mm 2 sensor was placed in the middle of the palm of the non-dominant hand with non-caustic glue. A variation of about one-thousandth of a degree (~ can be detected under such conditions. The amplitude and duration ofresponse were measured.
lnstatttaneotts heart rate This was recorded from 3 silver electrodes in precordial position. The D2 derivation signal (the interval between 2 consecutive R waves) was processed and delivered in the form of instantaneous heart frequency (b.p.m.). The smallest appreciable variation was 0.5 of a beat per minute and the calibrated scale ranged from 0 to 200 beats per minute.
h2stantaneous respiratoryfrequenc)' This was estimated by an oblong thermistor, (length 10mm, diameter 3 mm) placed at the entrance of the left nostril with hypoallergic adhesive ribbon. The processed signal gave the instantaneous respiratory frequency (b.p.m.), on the basis of the difference of temperature between inhaled and exhaled air. The subject only 'feels' the presence of the sensor for a few minutes, then 'forgets' the appliance.
Recording apparatus This was made up of a YTSE 460 type BBC (Brown Boveri) 6-channel potentiometric d.c. recorder fitted with an event tracer,
and of an automatic synchronization appliance which cancelled out temporal differences between the 6 markers. (Paper width: 250 mm.)
Proced,lre Two groups were submitted to the experiment during a competition and at the laboratory. The different sensors were placed 30 rain before the beginning of the experiment in order to avoid disturbance and to record a stable base line. A first experimental situation was set ut5 in the field, during a competition. One group was composed of 15 marksmen (target at 10 m). The competition lasted 2 h 15 min (60 shots) for men (10) and I h 30 min (40 shots) for women (5). Time organization between successive shots was free. The other group was composed of 7 archers (4 men and 3 women). The competition consisted of 9 shots (30 rain). In both groups, two distinct phases were evidenced in each shot: a concentration phase prior to shooting and the shooting phase itself. All sportsmen tested belonged to a high level category: Marksman: international, subjects numbered 2, 3, 4, 5, 9, 10, 13; national, subjects numbered 1, 6, 7, 8, 11, 12, 14, 15. Archers: international, subjects numbered 1, 4, 7; national, subjects numbered 2, 3, 5, 6. A second experimental situation was set up at the laboratory for the two groups. Each subject's ANS responses were recorded during a mental recall of competition practice vs neutral imagery. Subjects had to be present 30 rain before the beginning of the experiment in order to maintain a good thermal balance (24 ~C). They were phonically isolated. Every two minutes, the subject received instructions to imagine all shooting parameters and successive stages and to picture all intero- and exteroceptive sensations usually associated with them. i.e. a landscape (neutral imagery), the outline of a leaf (neutral imagery), a shooting sequence (aiming and shooting, nine times). In each situation, ANS variables monitoring was continuous.
Anal)'sis of results From overall and individual observations, 9 shots and concentration periods (during which the competitor mentally anticipates the oncoming performance) were selected to be compared with 9 imagery situations. The marksman's typical concentration phase is based on mental rehearsal of his shot. The 9 shots chosen are representative of the competition duration. For male subjects and from the overall series of 60 shots, only those numbered 2, 9, 16, 23, 30, 37, 44, 51, 58 were selected. For female subjects, from the overall series of 40 shots, only those numbered 4, 8, 12, 16, 20, 24, 28, 32, 36 were selected. It is possible with this system to compare a competition situation with that of a laboratory, limiting information losses due to the drawn-out nature of the competition, male and female subjects being considered, for this task, to form the same population. Such an analysis is possible as other results from the same discipline have shown that performance evolution remains stable through time (Deschaumes-Molinaro et al., submitted for publication). The archery competition was made up of 9 shots per subject, i.e. 9 periods of concentration, 9 actual shots and 9 imagery situations. Each shot or rehearsal sequence is considered as a stimulation. Analysis of results is based on non-parametric statistics (Friedman analysis of variance to inter-compare the three situations,
32 Comparisons of ANS responses #t situations taken two-by-
and a Mann-Whitney test to compare groups of individuals with differing performances) and on the calculation of the ratio between two given situations for each parameter.
two
For subjects overall, comparisons of ANS responses in situations taken two-by-two makes up the ratio for each variable. The comparison of autonomic responses in the 3 situations (concentration, shooting and shooting imagery) taken 2-by-2 showed a tendency to similarity (ratio close to 1). Marksmen: concentration/shooting, median (semi-interquartile range) are 1.03 (0.07) imagery/shooting, median (semi-interquartile range) are 1.45 (0.44) imagery/concentration, median (semi-interquartile range) are 1.84 (0.86). Archers: concentration/shooting, median (semi-interquartile range) are 1.01 (0.08) imagery/shooting, median (semi-interquartile range) are 1.04 (0.29) imagery/concentration, median (semi-interquartile range) are 1.08 (0.33).
RESULTS
Example of a recording Shooting imagery induces a specific ANS pattern of responses. Fig. 2 shows a classical type of responses obtained in the laboratory, a decrease of skin blood flow, skin temperature, skin resistance and an increase of skin potential (especially in shots 1 and 2), instantaneous heart rate (shots 1 and 3) and instantaneous respiratory frequency (shots 1, 2 and 3). Comparison of the 3 situations: concentration, shoothlg and shoothlg hnagery by non-parametric statistical analysis Analysis of the 15 marksmen's results (Fig. 3) and that of the 7 archers (Fig. 4) during concentration phase, shooting and shooting imagery phases. Subjectby-subject analysis is more accurate owing to the subject's preferential response channel. Non-statistically different variables (we could not reject the null hypothesis) were evidenced. For each subject, at least one variable accounted for similarity up to 5. Only one marksman, out of 22 (P < 0.01, binomial test) did not show any similarity in the 3 situations: concentration, shooting and shooting imagery.
SHOOTING IMAGERY
hldividual subjects For each subject, in both marksmen (Fig. 5) and archers (Fig. 6), a variable-by-variable ratio is established between results obtained between concentration phases, shooting phases and imagery phases, respec-
~ m
SIIOOT
3
SHOOT 2
"'It
SHOOT
30 s:
I W.m't.*C-! 0.1~ I ! SKIN BLOODFLOW SKIN TEMPERATURE
10 mV SKIN POTENTIAL
10kQ I
I
I
SKIN RESISTANCE
10 bp.m. i-----I
10bp.m. I
!
INSTANTANEOUS HEART RATE
INSTANTANEOUS RESPIRATORy FREQUENCY
Fig. 2. Example of a shooting imagery recording. Each 'shot' is considered a stimulation and produces ANS responses: a decrease in skin blood flow, skin temperature, skin resistance; an increase in skin potential, instantaneous heart and respiratory frequencies are evidenced in this subject.
33
Fig. 3. Air rifle shooting: comparison of 3 situations (concentration, shooting, imagery) subject-by-subject (numbered from 1 to 15) by the Friedman analysis of variance, for each ANS parameter. SR, skin resistance; SP, skin potential; SBF, skin blood flow; ST, skin temperature; IHR, instantaneous heart rate; IRF, instantaneous respiratory frequency; (for each variable: PR, prestimulation level; VAR, amplitude of variation; DUR, perturbation duration). The non-significantly different variables are evidenced (threshold P > 0.05).
V a t l~r
Fig. 4. Archery: comparison of 3 situations (concentration, shooting, imagery) subject-by-subject (numbered from 1 to 7) by the Friedman analysis of variance, for each ANS parameter. SR, skin resistance; SP, skin potential; SBF, skin blood flow; ST, skin temperature; IHR, instantaneous heart rate; IRF, instantaneous respiratory frequency; (for each variable: PR, prestimulation level; VAR, amplitude of variation; DUR, perturbation duration). The non-significantly different variables are evidenced (threshold P > 0.05).
tively. F o r each subject, a median o f variable-byvariable ratios obtained is then established. T w o median ratios are thus obtained: concentration/ shooting and imagery/shooting (Fig. 5).
If these results were c o m p a r e d to those set up by Shooting Federations, we observed that: The most accurate competitors are close to ratio one. The distance between each subject's ratios and the
34 1.5
1.0
CzS 0.5 I
2
3
4
5
6
7
8
9
I0 I I
12 13 14 15 SL~JECTS
Fig. 5. Air rifle shooting: spread of individual median (concentration/shooting and imagery]shooting) around the theoretical ratio value one.
l~qTlO 1.51
mc,'~ 0.5 1
2
3
4
5
6
7 SUBJECTS
Fig. 6. Archery: spread of individual median (concentration] shooting and imagery/shooting) around the theoretical ratio value one.
theoretical value one, makes it possible to compare the most or the least efficient groups by the Mann-Whitney's test. Marksmen: subjects numbered 2, 3, 4, 5, 9, 10, 13 (more accurate) and 1, 6, 7, 8, 11, 12, 14, 15 (less accurate) are significantly different as far as the concentration/shooting ratio is concerned (z = -2.77, P < 0.002) and a tendency for the imagery/shooting ratio is shown (z = - 1.33, P < 0.09). Archers: subjects numbered 1, 4, 7 (more accurate) and 2, 3, 5, 6 (less accurate) are significantly different as far as the imagery/shooting ratio (z = - 1.76, P < 0.03) is concerned and a tendency for the concentration/shooting ratio is shown up (z = 1.06, P < 0.14).
DISCUSSION
Since Perky's work 23 in 1910, mental imagery has always interested psychologists, physiologists and psychophysiologists. However, research is still difficult as only indirect analyses are assessed. First, technical and conceptual improvements in the analysis of autonomic responses allowed us to compare imagery and actual
shots in real time. Second, the analysis ofcorresponding phenomena is reproducible, and third, shooting competitions are made up of alternating concentration and shooting phases. Moreover, marksmen's and archers' interviews point out that training is primarily based on visual and kinesthetic imagery of the shot. This training covers the concentration phase which can consequently be assimilated already to an imagery phase from a functional viewpoint. The similarity of the conceptual content of concentration and laboratory imagery phases enables us to assimilate them and therefore hypothesize concentration as a time ofmental rehearsal or in-the-field imagery. Experimental arrangements did not disturb the subject' performances, their results being in compliance with those usually achieved. Post-competition interviews did not evidence any disturbance connected with recording arrangements. Global analysis of results confirms that both these precision shooting events (firearm and archery) provoke the same kind of ANS response, this similarity being due to similar spatial and temporal preparation and action. Results from the overall test population (Deschaumes-Molinaro et al., submitted for publication) show that neutral imagery is never statistically related (7.2 test) with any of the phases under study, i.e. concentration, shooting, or shooting imagery. This shows that these situations draw specific response. Most of the subjects showed a response similarity between concentration, shooting and shooting imagery phases. Only one marksman showed no similar autonomic variable in these phases. On the one hand, we noted that this subject was not efficient; on the other he might be a low imagery subject's or poorly motivated. Numerous studies have successfully demonstrated that imagery ability has a functional significance as a mediator of behavioH The low number of statistically identical variables may surprise, but individual autonomic functioning expression is multiple, responding to individual logistics. Moreover, the now known and verified individuality of ANS response via a preferential channeP 8,3~ limits the statistical results ipso facto. It should be remembered that ANS intersubject correlations in face of conventional stimuli only reach 7% 29, whereas other research related to mental imagery and implying autonomic measurements (Deschaumes-Molinaro et al., submitted for publication) has shown that respectively 24~o and 31% of intersubject correlations exist in rifleshooting and archery. This much higher percentage indicates the reality of the described phenomenon. Thus, only multiehannel analysis can evidence identity
35 between concentration, shooting and imagery phases. For all 3 situations, all variables are in turn paired with action response, i.e. actual shooting. Thus, we may conclude from these results that ANS parameters show identical responses in concentration, shooting and shooting imagery situations; this relationship enables one to postulate the identity of the viscero-motor phenomena involved. This is in accordance with Decety and Michel's ~ results which showed in a chronometric study that mental movements closely mimic actual movements and are likely to involve the same planning program. In the same way, the A N S variable prestimulation level shows similarities between 2 phases; this result is in line with those obtained, and which are related to the estimation of different mental load levels and shows the importance of this preparatory period 5. The ratio of ANS responses in the concentration or imagery situation/shooting situation, generally speaking, is close to one. The ratio one could be defined as the ideal ratio existing between autonomic response to action representation and to actual action. In both events this ratio was studied; it varies from 1.01 to 1.08 in archery, and from 1.03 to 1.84 in rifle shooting, showing that the values that are closest to one for concentration/shooting, then for imagery/shooting and finally for imagery/concentration. It is consequently possible that imagery is not so close to concentration because, on the one hand, subjects are requested to imagine their shooting sequence, and on the other, concentration also comprises a correction and adaptation phase depending on the previous performance, and this is not taken into account by the experimental situation of imagery. Subject-by-subject study of this ratio shows an individual aptitude to produce a typical response inducing subidentical autonomic responses during concentration and shooting, and even during imagery. An important fact is here brought into perspective i.e. the comparison of autonomic measurements and the Federations' classification scores. The closer this ratio is to one, which would be 'perfect similarity', the more accurate the competitor. This tendency can be found in both events as concentration and imagery situations in their relationship to actual shooting classify individuals. This classification is significantly close to that obtained by the level of performance; this particularly important result shows that a correspondence must exist between mental representation of a motor action and execution of the action attempting to achieve a better performance. Subjects are spread above or below the theoretical value one. It can be supposed that the classification
'above' corresponds to overflowing emotional reactivity in one of the 2 phases and that this interferes with accuracy; the classification 'below' shows a lack of equivalence between mental representation and the action. There are numerous intermediaries between these two extremes; individual and adapted training with a view to limiting the impact of the emotional component or to enhancing mental representation could be envisioned. The ideamotor theory which states that voluntary movement is mediated by the image of this movement by inner representation ~7.2~.2s seems to be verified here through ANS response. This is in compliance with some recent results suggesting that there is some relationship between A N S and CNS activity 26. The better the mental representation of the action that brings autonomic response similar to actual action, the greater the accuracy. Imagery could be epiphenomenal but current theory 22 tends to show that the same neuromuscular networks are brought into play. Imagery that triggers or installs a viscero-motor program is an efficient help in action preparation. Moreover, mental rehearsal no doubt enables the subject to build up a better cognitive picture of the action 4. The concentration phase is itself based on the image the subject has of his or her own performance and the performance already depends on the quality of this phase. Accuracy could thus be enhanced with the accuracy of mental representation prior to the action. REFERENCES 1 Decety,J. and Michel, F., Comparative analysis of actual and mental movement times in two graphic tasks, Brain and Cognition, 11 (1989)87-97. 2 Deckert, G.H., Pursuit eye movements in the absence of visual stimulus, Science, 143 (1964) 1192-1193. 3 Denis, M., Visual imagery and the use of mental practice in the development of motor skills, Can. J. AppL Sport ScL, 10 (1985) 4S-16S. 4 Denis, M., Chevalier, N. and Eloi, S., Imagerie et r6p&ition mental dans racquisition d'habilet6s motrices. In A. Vom Hofe (Ed.), Tdches, Traitement de 17nformation et Comportements dans les Activit~s Physiques et Sportives, Issy-les Moulineaux, Editions EAP, 1989, pp. 11-37. 5 Deschaumes-Molinaro, C., Delhomme, G., Dittmar, A., Fouquet, R., Rougny,R. and Vernet-Maury,E., Characterization of mental load by monitoring of Autonomic Nervous System effectors, Ergonomics, in press. 6 Dinmar, A., Saumet, J.L. and Vernet-Maury, E., Apports des param6tres thermovasculaires dans i'analyse de la r~ponse 61ectrodermale, J. PhysioL (Paris), 80 (1985) 22. 7 Dinmar, A., Skin thermal conductivity.In J.L. L6v~que,(Eds.), Cutaneous blvestigation in Health and Disease, Marcel Dekker, New York, 1989, pp. 323-358. 8 Drummond, P., White, K. and Ashton, R., Imagery vividness affects habituation rate, Psychophysiology, 15 (1978) 193-195.
36 9 Ekman, P., Levenson, R.W. and Friesen, W.W., Autonomic nervous system activity distinguishes among emotions, Science, 221 (1983) 1208-1210. I0 Ernest, C.H. and Paivio, A., Imagery ability in paired-associate and incidental learning, Psychon. Sci., 15 (1969) 181-1829 11 Ernest, C.H., Imagery ability and cognition: a critical review, J. Ment. Imagery, 2 (1977) 181-216. 12 Fowles, D.C., Christie, M.J. and Edelberg, R., Publication recommendations for electrodermal measurements, Psychophysiology, 18 (1981) 232. 13 Furedy, J.J. and Klajner, F., Imaginational Pavlovian conditioning of large magnitude cardiac decelerations with tilt as unconditional stimulus, Psychophysiology, 15 (1978) 538-543. 14 Furedy, J.J., An experimental psychophysiological approach to human bradycardiac reflexes, Pay. J. Biol. Sci., 20 (1985) 88-96. 15 Ikeda, Y. and Hirai, H., Voluntary control of electrodermal activity in relation to imagery and internal perception scores, Psychophysiology, 13 (1976) 330-333. 16 Jaeobson, E., Electrical measurements of neuromuscular states during mental activities. I. Imagination of movement involving skeletal muscle, Am. J. Physiol. 91 (1930) 567-608. 17 James, W., Principles of Psychology, Vol. 2, New York, Dover, New Edn., 1950. 18 Lacey, J.I., Bateman, D.E. and Vanlehn, R., Autonomic response specificity: an experimental study, Psychosom. Med., 15 (1953) 8-21. 19 Lang, P.J., A bio-informational theory of emotional imagery, Psychophysiology, 16 (1979) 495-512. 20 Lang, P.J., Kosak, M.J., Miller, G.A., Levin, D.N. and McLean, A., Emotional imagery: conceptual structure and pattern of somato-visceral response, Psychophysiology, 17 (1980) 179-1929 21 Leslie, A.M., Pretense and representation: the origins of'Theory of mind', PsychoL Rev., 94 (1987) 412-426.
22 Lusebrink, V.B. and McGuigan, F.J., Psychophysiological components of imagery, Pay. J. BioL Sci., 24 (1989) 58-62. 23 Perky, C.W., An experimental study of imagination, Am. J. PsychoL, 21 (1910) 422-425. 24 Peronnet, F. and Farah, MJ., Mental rotation: an event-related potential study x~ith a validated mental rotation task, Brain and Cognition, 9 (1989) 279-288. 25 Posner, M.I., Chronomectric Explorations of Mind, Hillsdale, NJ, Erlbaum, 1978. 26 Rippon, G., Individual differenc6s in electrodermal and electroeneephalographic asymmetries, h~t. J. PsychophysioL, 8 (1990) 309-320. 27 Robin, O., Vernet-Maury, E., Dittmar, A. and Vinard, M., The ohmic perturbation duration (OPD index), an autonomic index to objectivate anxiety and pain response, Pain, 4 (1987) 428. 28 Shepard, R.N., The mental image, Am. Psychol., 2 (1978) 125-137. " 29 Vernet-Maury, E., Elaboration de Critdres Objectifs et Quantitatifs des Rdactions Subconscientes ou Emotionnelles, Contrat DRET 84111155 - Rapport de Synth~se, 1987. 30 Vernet-Maury, E., Sicard, G., Dittmar, A. and DeschaumesMolinaro, C., Autonomic nervous system preferential responses, Activ. Nerv. Super., 32 (1990) 37-38. 31 Wallin, B.G. and Fagius, J., The sympathetic nervous system in man. Aspects derived from microelectrode recordings, Trends Neurosci., 9 (1986) 63-66. 32 White, K.D., Ashton, R. and Lewis, S., Learning a complex skill: effects of mental practice, physical practice and imagery ability, hit. J. Sport PhysioL, I0 (1979) 71-'18. 33 Wilder, J., Basimetric approach (law ofinitial value) to biological rhythms, Ann. N.Y. Acad. Sci., 98 (1962) 1211-1228.
Elsevier
29
BBR 01218
Relationship between mental imagery and sporting performance C. Deschaumes-Molinaro i, A. Dittmar z and E. Vernet-Maury 1 ~Laboratoire de Physiologie neurosensorielle, CNRS-University Claude Bernard]Lyon, Villeurbanne (France) and ?Laboratoire de Thermor~gulation et Energ~tique de l'exercice, CNRS URA 1341, Facult~ de Mddecine, Lyon (France).
(Received 21 August 1990) (Revised version received 29 May 1991) (Accepted 22 July 1991) Key words: Mental imagery; Autonomic nervous system; Performance; Concentration
Simultaneous measurement of six autonomic nervous system (ANS) variable responses during mental rehearsal of an action, makes it possible to draw a parallel between mental imagery of a task and its actual execution. The experiment was carried out in the field during precision shooting competitions and in the laboratory for imagery activity, on 22 subjects. Results show that there is similarity of ANS response in the three situations: the period of concentration prior to shooting, actual shooting and mental representation of shooting. The ratio formed by ANS response during concentration and imagery of actual firing tends towards the value one, therefore towards some identity. All subjects may be classified around this value; it is worth noting that subject distribution around this value corresponds to performance value. It seems that the better the subject, the closer his concentration/shooting or imagery/actual shooting ratio is to the theoretical value one. These results show the utmost importance of the quality ofmental representation for performance improvement. It can be supposed that subject classification above the theoretical value one corresponds to overflowingemotional reactivity in one of the two phases and that this interferes with accuracy; a placing below the theoretical value one shows a lack of similarity between mental representation and the action.
INTRODUCTION Imagery falls within the framework o f repetitive mental practice or mental training and designates mental representation o f the performance o f a m o t o r pattern without concomitant production of the muscular activity normally required for the act 4. Imagery, extensively studied in its psychological aspects, is however more difficult to approach physiologically, as measurements can be indirect only. N u m e r o u s measurements have been carried out by the mental c h r o n o m e t r y m e t h o d z5 but these do not account for the processes brought into play. Recent research shows the interaction existing between imagery and perception 2"~6"24. The numerous hypotheses on the functional similarity between imagery and perception t9'2~ go as far as to assimilate mental and physical practice 3"13'!4'32. Autonomic nervous system ( A N S ) effectors are activated by imagery, as testified by extensive research
Correspondence: E. Vernet-Maury, Laboratoire de Physiologie neurosensorielle, CNRS-Universit6 Claude Bernard/Lyon, F-69622 Villeurbanne codex, France. Fax: (33)78949585.
carried out on the heart rate. But additionally, these effectors are thought to be liable to give a modified response in one way or another as the result o f the effects and contents o f imagery t9"2~ In most cases, individual differences, which are the result o f the subjects' ability to imagine and make images vivid and accurate, are evidenced 8'1~ Recent methodological and conceptual progress on A N S functioning as a specific system permits the thought o f its study to objectify brain activity 9'26m. At this time, brain ideography methodologies, promising though they may be, do not make such objectivation possible on account o f their temporal limitations. On the other hand, study o f the A N S response sheds new light on the central nervous system ( C N S ) functioning, on account o f their m a n y interrelations. Original indices and captors developed by the Vernet-Maury/Dittmar team zg, have m a d e it possible to apprehend A N S functioning in a reliable and reproductible manner. Continuous and simultaneous measurement o f six A N S effectors on the same group o f subjects in two situations: one bearing on physical practice and the other on imagination o f this practice will make it possible to bring out the impact o f imagery on overall autonomic responses. It has been possible to c o m p a r e these two
30 situations and performance. The hypothesis may be put forward that with these top-level marksmen, the concentration phase (immediately prior to shooting) based on mental rehearsal of the oncoming shot is very close to that of laboratory imagery from the view point of cognitive process implication. From the results obtained, it may be possible to discuss the ideomotor hypothesis which states that voluntary movement is mediated by the image which becomes the subject of this movement by inner representation ~7"2"28. Knowing that a subject responds through a preferential channel, it is important to measure six channels simultaneously in order to obtain maximal response from all subjects ~8"3~ These individual differences therefore also require that each subject's results are considered individually.
MATERIALS A N D M E T H O D S Six ANS variables were selected to be quantified: electrodermal response: skin potential and resistance; thermovascular variables: skin blood flow (original sensor) and skin temperature; cardiorespiratory variables: instantaneous heart rate and original instantaneous respiratory frequency.
A+
sec
AI 6'5
0 5
15
30
5
15
30
C+
Skin resistance response Cutaneous resistance (Kf~) was recorded using 30-mm 2 unpolarizable round Capsulex electrodes placed on the second phalanx of the forefinger and of the median finger of the non-dominant hand, held by an adhesive tape. Resistance is measured with ! 5 ttA d.c. current. The skin potential and resistance measurements we carried out by means of a high rate common rejection mode differential amplifier. In the same way, recorder inputs were of the differential mode and resistance circuit supply was of the floating type. Skin resistance measuring current passes between the forefinger and the middle finger while the skin potential was measured between the hypothenar eminence and the inner side of the forearm. Any interference between skin potential and resistance was therefore eliminated, as there was no influence in the measuring circuits, there was no possible influence in the biological tissues. This fact was verified by simultaneous and single recording of these two variables.
o
sec
5
Y
A_
B
60 sec 0 5
B+
(~
Skb~ potential responses Skin potential (mV) was recorded using Beckman self-adhesive 78-mm 2 electrodes. Electrode placement was in compliance with the recommendations of Fowles et alJ 2. The active electrode was stuck to the hypothenar eminence of the subject's non-dominant hand (after alcohol-ether cleaning of the skin). The reference electrode was stuck 10 cm from the wrist on the equidistant line of the median plane and the outer extremity of the forearm. Electrodermal potential variations were measured by the SYDER code 6 which permits classification of elementary responses, according to their form, sign ( + or - ) and duration. As far as potential variations .were concerned, only this index permitted their analysis, the other measurements being less satisfactory or even redundant. According to this code, 3 positive and 3 negative skin potential forms were considered (cf. Fig. I, A, B and C potential forms).
15
w
30
60 sec
C_
6'0 s e c 0
5
15
30
60 Sec
Fig. 1. SYDER code (decoding system for electrodermal responses). A, B and C skin potential forms. These are schemata ofcurve envelopes. Durations can vary except for the A form which is never longer than 5 s. Some responses correspond to a combination of 2 or 3 forms (example: response classically described as diphasic).
31 The response amplitude to a stimulation following the law of initial value ~3, another more reliable index was defined. A parallel observation ofthermovascular indices made it possible to define the time during which the subject 'responds' to stimuli without referring to the initial value time. This ohmic perturbation duration(s) or OPD index of skin resistance reflects the emotional load of the stimulation27.
Superficial skin blood flow The skin blood flow was measured using an original system: the HEMATRON (Dittmar, CNRS/ANVAR Patent n ~ 85 15932). The HEMATRON system continuously measures the thermal conductivity of material and of living tissues using the principle ofthermie clearance 7. The transducer consists of a disc 25 mm in diameter and 4-mm thick which is fixed on the skin by an adhesive tape. The measuring face in contact with the skin is made up of 2 parts: the reference area at the periphery of the disc and the measuring area at the center of the disc. The difference between these 2 areas is measured using 16 thermocouple jt~nctions. A very low thermal inertia flat heater is located in the central part of the disc. A proportional, integrative and derivative device controls the heating power to maintain a constant difference of 2 ~ between the central area and the periphery. The size and the shape of the heater are designed to induce a thermal field in the capillary network. The power necessary to maintain the constant difference depends on skin blood flow: heat is transferred through the skin and washed out by the blood flow. At all times the electric power is proportional to the heat evacuated by the tissue blood flow. Blood flow in skin capillaries is submitted to mierochanges reflecting the variations of emotional load. Skin blood flow variations were measured: (i) by the difference (positive or negative) between the pre- and post-stimulation values expressed in W" m - I. oC - ~; (ii) by the duration of the perturbation of the oscillations.
Skin temperature This was measured by a low inertia thermistor (I0 K3 MCD2 Betatherm). A 4-mm 2 sensor was placed in the middle of the palm of the non-dominant hand with non-caustic glue. A variation of about one-thousandth of a degree (~ can be detected under such conditions. The amplitude and duration ofresponse were measured.
lnstatttaneotts heart rate This was recorded from 3 silver electrodes in precordial position. The D2 derivation signal (the interval between 2 consecutive R waves) was processed and delivered in the form of instantaneous heart frequency (b.p.m.). The smallest appreciable variation was 0.5 of a beat per minute and the calibrated scale ranged from 0 to 200 beats per minute.
h2stantaneous respiratoryfrequenc)' This was estimated by an oblong thermistor, (length 10mm, diameter 3 mm) placed at the entrance of the left nostril with hypoallergic adhesive ribbon. The processed signal gave the instantaneous respiratory frequency (b.p.m.), on the basis of the difference of temperature between inhaled and exhaled air. The subject only 'feels' the presence of the sensor for a few minutes, then 'forgets' the appliance.
Recording apparatus This was made up of a YTSE 460 type BBC (Brown Boveri) 6-channel potentiometric d.c. recorder fitted with an event tracer,
and of an automatic synchronization appliance which cancelled out temporal differences between the 6 markers. (Paper width: 250 mm.)
Proced,lre Two groups were submitted to the experiment during a competition and at the laboratory. The different sensors were placed 30 rain before the beginning of the experiment in order to avoid disturbance and to record a stable base line. A first experimental situation was set ut5 in the field, during a competition. One group was composed of 15 marksmen (target at 10 m). The competition lasted 2 h 15 min (60 shots) for men (10) and I h 30 min (40 shots) for women (5). Time organization between successive shots was free. The other group was composed of 7 archers (4 men and 3 women). The competition consisted of 9 shots (30 rain). In both groups, two distinct phases were evidenced in each shot: a concentration phase prior to shooting and the shooting phase itself. All sportsmen tested belonged to a high level category: Marksman: international, subjects numbered 2, 3, 4, 5, 9, 10, 13; national, subjects numbered 1, 6, 7, 8, 11, 12, 14, 15. Archers: international, subjects numbered 1, 4, 7; national, subjects numbered 2, 3, 5, 6. A second experimental situation was set up at the laboratory for the two groups. Each subject's ANS responses were recorded during a mental recall of competition practice vs neutral imagery. Subjects had to be present 30 rain before the beginning of the experiment in order to maintain a good thermal balance (24 ~C). They were phonically isolated. Every two minutes, the subject received instructions to imagine all shooting parameters and successive stages and to picture all intero- and exteroceptive sensations usually associated with them. i.e. a landscape (neutral imagery), the outline of a leaf (neutral imagery), a shooting sequence (aiming and shooting, nine times). In each situation, ANS variables monitoring was continuous.
Anal)'sis of results From overall and individual observations, 9 shots and concentration periods (during which the competitor mentally anticipates the oncoming performance) were selected to be compared with 9 imagery situations. The marksman's typical concentration phase is based on mental rehearsal of his shot. The 9 shots chosen are representative of the competition duration. For male subjects and from the overall series of 60 shots, only those numbered 2, 9, 16, 23, 30, 37, 44, 51, 58 were selected. For female subjects, from the overall series of 40 shots, only those numbered 4, 8, 12, 16, 20, 24, 28, 32, 36 were selected. It is possible with this system to compare a competition situation with that of a laboratory, limiting information losses due to the drawn-out nature of the competition, male and female subjects being considered, for this task, to form the same population. Such an analysis is possible as other results from the same discipline have shown that performance evolution remains stable through time (Deschaumes-Molinaro et al., submitted for publication). The archery competition was made up of 9 shots per subject, i.e. 9 periods of concentration, 9 actual shots and 9 imagery situations. Each shot or rehearsal sequence is considered as a stimulation. Analysis of results is based on non-parametric statistics (Friedman analysis of variance to inter-compare the three situations,
32 Comparisons of ANS responses #t situations taken two-by-
and a Mann-Whitney test to compare groups of individuals with differing performances) and on the calculation of the ratio between two given situations for each parameter.
two
For subjects overall, comparisons of ANS responses in situations taken two-by-two makes up the ratio for each variable. The comparison of autonomic responses in the 3 situations (concentration, shooting and shooting imagery) taken 2-by-2 showed a tendency to similarity (ratio close to 1). Marksmen: concentration/shooting, median (semi-interquartile range) are 1.03 (0.07) imagery/shooting, median (semi-interquartile range) are 1.45 (0.44) imagery/concentration, median (semi-interquartile range) are 1.84 (0.86). Archers: concentration/shooting, median (semi-interquartile range) are 1.01 (0.08) imagery/shooting, median (semi-interquartile range) are 1.04 (0.29) imagery/concentration, median (semi-interquartile range) are 1.08 (0.33).
RESULTS
Example of a recording Shooting imagery induces a specific ANS pattern of responses. Fig. 2 shows a classical type of responses obtained in the laboratory, a decrease of skin blood flow, skin temperature, skin resistance and an increase of skin potential (especially in shots 1 and 2), instantaneous heart rate (shots 1 and 3) and instantaneous respiratory frequency (shots 1, 2 and 3). Comparison of the 3 situations: concentration, shoothlg and shoothlg hnagery by non-parametric statistical analysis Analysis of the 15 marksmen's results (Fig. 3) and that of the 7 archers (Fig. 4) during concentration phase, shooting and shooting imagery phases. Subjectby-subject analysis is more accurate owing to the subject's preferential response channel. Non-statistically different variables (we could not reject the null hypothesis) were evidenced. For each subject, at least one variable accounted for similarity up to 5. Only one marksman, out of 22 (P < 0.01, binomial test) did not show any similarity in the 3 situations: concentration, shooting and shooting imagery.
SHOOTING IMAGERY
hldividual subjects For each subject, in both marksmen (Fig. 5) and archers (Fig. 6), a variable-by-variable ratio is established between results obtained between concentration phases, shooting phases and imagery phases, respec-
~ m
SIIOOT
3
SHOOT 2
"'It
SHOOT
30 s:
I W.m't.*C-! 0.1~ I ! SKIN BLOODFLOW SKIN TEMPERATURE
10 mV SKIN POTENTIAL
10kQ I
I
I
SKIN RESISTANCE
10 bp.m. i-----I
10bp.m. I
!
INSTANTANEOUS HEART RATE
INSTANTANEOUS RESPIRATORy FREQUENCY
Fig. 2. Example of a shooting imagery recording. Each 'shot' is considered a stimulation and produces ANS responses: a decrease in skin blood flow, skin temperature, skin resistance; an increase in skin potential, instantaneous heart and respiratory frequencies are evidenced in this subject.
33
Fig. 3. Air rifle shooting: comparison of 3 situations (concentration, shooting, imagery) subject-by-subject (numbered from 1 to 15) by the Friedman analysis of variance, for each ANS parameter. SR, skin resistance; SP, skin potential; SBF, skin blood flow; ST, skin temperature; IHR, instantaneous heart rate; IRF, instantaneous respiratory frequency; (for each variable: PR, prestimulation level; VAR, amplitude of variation; DUR, perturbation duration). The non-significantly different variables are evidenced (threshold P > 0.05).
V a t l~r
Fig. 4. Archery: comparison of 3 situations (concentration, shooting, imagery) subject-by-subject (numbered from 1 to 7) by the Friedman analysis of variance, for each ANS parameter. SR, skin resistance; SP, skin potential; SBF, skin blood flow; ST, skin temperature; IHR, instantaneous heart rate; IRF, instantaneous respiratory frequency; (for each variable: PR, prestimulation level; VAR, amplitude of variation; DUR, perturbation duration). The non-significantly different variables are evidenced (threshold P > 0.05).
tively. F o r each subject, a median o f variable-byvariable ratios obtained is then established. T w o median ratios are thus obtained: concentration/ shooting and imagery/shooting (Fig. 5).
If these results were c o m p a r e d to those set up by Shooting Federations, we observed that: The most accurate competitors are close to ratio one. The distance between each subject's ratios and the
34 1.5
1.0
CzS 0.5 I
2
3
4
5
6
7
8
9
I0 I I
12 13 14 15 SL~JECTS
Fig. 5. Air rifle shooting: spread of individual median (concentration/shooting and imagery]shooting) around the theoretical ratio value one.
l~qTlO 1.51
mc,'~ 0.5 1
2
3
4
5
6
7 SUBJECTS
Fig. 6. Archery: spread of individual median (concentration] shooting and imagery/shooting) around the theoretical ratio value one.
theoretical value one, makes it possible to compare the most or the least efficient groups by the Mann-Whitney's test. Marksmen: subjects numbered 2, 3, 4, 5, 9, 10, 13 (more accurate) and 1, 6, 7, 8, 11, 12, 14, 15 (less accurate) are significantly different as far as the concentration/shooting ratio is concerned (z = -2.77, P < 0.002) and a tendency for the imagery/shooting ratio is shown (z = - 1.33, P < 0.09). Archers: subjects numbered 1, 4, 7 (more accurate) and 2, 3, 5, 6 (less accurate) are significantly different as far as the imagery/shooting ratio (z = - 1.76, P < 0.03) is concerned and a tendency for the concentration/shooting ratio is shown up (z = 1.06, P < 0.14).
DISCUSSION
Since Perky's work 23 in 1910, mental imagery has always interested psychologists, physiologists and psychophysiologists. However, research is still difficult as only indirect analyses are assessed. First, technical and conceptual improvements in the analysis of autonomic responses allowed us to compare imagery and actual
shots in real time. Second, the analysis ofcorresponding phenomena is reproducible, and third, shooting competitions are made up of alternating concentration and shooting phases. Moreover, marksmen's and archers' interviews point out that training is primarily based on visual and kinesthetic imagery of the shot. This training covers the concentration phase which can consequently be assimilated already to an imagery phase from a functional viewpoint. The similarity of the conceptual content of concentration and laboratory imagery phases enables us to assimilate them and therefore hypothesize concentration as a time ofmental rehearsal or in-the-field imagery. Experimental arrangements did not disturb the subject' performances, their results being in compliance with those usually achieved. Post-competition interviews did not evidence any disturbance connected with recording arrangements. Global analysis of results confirms that both these precision shooting events (firearm and archery) provoke the same kind of ANS response, this similarity being due to similar spatial and temporal preparation and action. Results from the overall test population (Deschaumes-Molinaro et al., submitted for publication) show that neutral imagery is never statistically related (7.2 test) with any of the phases under study, i.e. concentration, shooting, or shooting imagery. This shows that these situations draw specific response. Most of the subjects showed a response similarity between concentration, shooting and shooting imagery phases. Only one marksman showed no similar autonomic variable in these phases. On the one hand, we noted that this subject was not efficient; on the other he might be a low imagery subject's or poorly motivated. Numerous studies have successfully demonstrated that imagery ability has a functional significance as a mediator of behavioH The low number of statistically identical variables may surprise, but individual autonomic functioning expression is multiple, responding to individual logistics. Moreover, the now known and verified individuality of ANS response via a preferential channeP 8,3~ limits the statistical results ipso facto. It should be remembered that ANS intersubject correlations in face of conventional stimuli only reach 7% 29, whereas other research related to mental imagery and implying autonomic measurements (Deschaumes-Molinaro et al., submitted for publication) has shown that respectively 24~o and 31% of intersubject correlations exist in rifleshooting and archery. This much higher percentage indicates the reality of the described phenomenon. Thus, only multiehannel analysis can evidence identity
35 between concentration, shooting and imagery phases. For all 3 situations, all variables are in turn paired with action response, i.e. actual shooting. Thus, we may conclude from these results that ANS parameters show identical responses in concentration, shooting and shooting imagery situations; this relationship enables one to postulate the identity of the viscero-motor phenomena involved. This is in accordance with Decety and Michel's ~ results which showed in a chronometric study that mental movements closely mimic actual movements and are likely to involve the same planning program. In the same way, the A N S variable prestimulation level shows similarities between 2 phases; this result is in line with those obtained, and which are related to the estimation of different mental load levels and shows the importance of this preparatory period 5. The ratio of ANS responses in the concentration or imagery situation/shooting situation, generally speaking, is close to one. The ratio one could be defined as the ideal ratio existing between autonomic response to action representation and to actual action. In both events this ratio was studied; it varies from 1.01 to 1.08 in archery, and from 1.03 to 1.84 in rifle shooting, showing that the values that are closest to one for concentration/shooting, then for imagery/shooting and finally for imagery/concentration. It is consequently possible that imagery is not so close to concentration because, on the one hand, subjects are requested to imagine their shooting sequence, and on the other, concentration also comprises a correction and adaptation phase depending on the previous performance, and this is not taken into account by the experimental situation of imagery. Subject-by-subject study of this ratio shows an individual aptitude to produce a typical response inducing subidentical autonomic responses during concentration and shooting, and even during imagery. An important fact is here brought into perspective i.e. the comparison of autonomic measurements and the Federations' classification scores. The closer this ratio is to one, which would be 'perfect similarity', the more accurate the competitor. This tendency can be found in both events as concentration and imagery situations in their relationship to actual shooting classify individuals. This classification is significantly close to that obtained by the level of performance; this particularly important result shows that a correspondence must exist between mental representation of a motor action and execution of the action attempting to achieve a better performance. Subjects are spread above or below the theoretical value one. It can be supposed that the classification
'above' corresponds to overflowing emotional reactivity in one of the 2 phases and that this interferes with accuracy; the classification 'below' shows a lack of equivalence between mental representation and the action. There are numerous intermediaries between these two extremes; individual and adapted training with a view to limiting the impact of the emotional component or to enhancing mental representation could be envisioned. The ideamotor theory which states that voluntary movement is mediated by the image of this movement by inner representation ~7.2~.2s seems to be verified here through ANS response. This is in compliance with some recent results suggesting that there is some relationship between A N S and CNS activity 26. The better the mental representation of the action that brings autonomic response similar to actual action, the greater the accuracy. Imagery could be epiphenomenal but current theory 22 tends to show that the same neuromuscular networks are brought into play. Imagery that triggers or installs a viscero-motor program is an efficient help in action preparation. Moreover, mental rehearsal no doubt enables the subject to build up a better cognitive picture of the action 4. The concentration phase is itself based on the image the subject has of his or her own performance and the performance already depends on the quality of this phase. Accuracy could thus be enhanced with the accuracy of mental representation prior to the action. REFERENCES 1 Decety,J. and Michel, F., Comparative analysis of actual and mental movement times in two graphic tasks, Brain and Cognition, 11 (1989)87-97. 2 Deckert, G.H., Pursuit eye movements in the absence of visual stimulus, Science, 143 (1964) 1192-1193. 3 Denis, M., Visual imagery and the use of mental practice in the development of motor skills, Can. J. AppL Sport ScL, 10 (1985) 4S-16S. 4 Denis, M., Chevalier, N. and Eloi, S., Imagerie et r6p&ition mental dans racquisition d'habilet6s motrices. In A. Vom Hofe (Ed.), Tdches, Traitement de 17nformation et Comportements dans les Activit~s Physiques et Sportives, Issy-les Moulineaux, Editions EAP, 1989, pp. 11-37. 5 Deschaumes-Molinaro, C., Delhomme, G., Dittmar, A., Fouquet, R., Rougny,R. and Vernet-Maury,E., Characterization of mental load by monitoring of Autonomic Nervous System effectors, Ergonomics, in press. 6 Dinmar, A., Saumet, J.L. and Vernet-Maury, E., Apports des param6tres thermovasculaires dans i'analyse de la r~ponse 61ectrodermale, J. PhysioL (Paris), 80 (1985) 22. 7 Dinmar, A., Skin thermal conductivity.In J.L. L6v~que,(Eds.), Cutaneous blvestigation in Health and Disease, Marcel Dekker, New York, 1989, pp. 323-358. 8 Drummond, P., White, K. and Ashton, R., Imagery vividness affects habituation rate, Psychophysiology, 15 (1978) 193-195.
36 9 Ekman, P., Levenson, R.W. and Friesen, W.W., Autonomic nervous system activity distinguishes among emotions, Science, 221 (1983) 1208-1210. I0 Ernest, C.H. and Paivio, A., Imagery ability in paired-associate and incidental learning, Psychon. Sci., 15 (1969) 181-1829 11 Ernest, C.H., Imagery ability and cognition: a critical review, J. Ment. Imagery, 2 (1977) 181-216. 12 Fowles, D.C., Christie, M.J. and Edelberg, R., Publication recommendations for electrodermal measurements, Psychophysiology, 18 (1981) 232. 13 Furedy, J.J. and Klajner, F., Imaginational Pavlovian conditioning of large magnitude cardiac decelerations with tilt as unconditional stimulus, Psychophysiology, 15 (1978) 538-543. 14 Furedy, J.J., An experimental psychophysiological approach to human bradycardiac reflexes, Pay. J. Biol. Sci., 20 (1985) 88-96. 15 Ikeda, Y. and Hirai, H., Voluntary control of electrodermal activity in relation to imagery and internal perception scores, Psychophysiology, 13 (1976) 330-333. 16 Jaeobson, E., Electrical measurements of neuromuscular states during mental activities. I. Imagination of movement involving skeletal muscle, Am. J. Physiol. 91 (1930) 567-608. 17 James, W., Principles of Psychology, Vol. 2, New York, Dover, New Edn., 1950. 18 Lacey, J.I., Bateman, D.E. and Vanlehn, R., Autonomic response specificity: an experimental study, Psychosom. Med., 15 (1953) 8-21. 19 Lang, P.J., A bio-informational theory of emotional imagery, Psychophysiology, 16 (1979) 495-512. 20 Lang, P.J., Kosak, M.J., Miller, G.A., Levin, D.N. and McLean, A., Emotional imagery: conceptual structure and pattern of somato-visceral response, Psychophysiology, 17 (1980) 179-1929 21 Leslie, A.M., Pretense and representation: the origins of'Theory of mind', PsychoL Rev., 94 (1987) 412-426.
22 Lusebrink, V.B. and McGuigan, F.J., Psychophysiological components of imagery, Pay. J. BioL Sci., 24 (1989) 58-62. 23 Perky, C.W., An experimental study of imagination, Am. J. PsychoL, 21 (1910) 422-425. 24 Peronnet, F. and Farah, MJ., Mental rotation: an event-related potential study x~ith a validated mental rotation task, Brain and Cognition, 9 (1989) 279-288. 25 Posner, M.I., Chronomectric Explorations of Mind, Hillsdale, NJ, Erlbaum, 1978. 26 Rippon, G., Individual differenc6s in electrodermal and electroeneephalographic asymmetries, h~t. J. PsychophysioL, 8 (1990) 309-320. 27 Robin, O., Vernet-Maury, E., Dittmar, A. and Vinard, M., The ohmic perturbation duration (OPD index), an autonomic index to objectivate anxiety and pain response, Pain, 4 (1987) 428. 28 Shepard, R.N., The mental image, Am. Psychol., 2 (1978) 125-137. " 29 Vernet-Maury, E., Elaboration de Critdres Objectifs et Quantitatifs des Rdactions Subconscientes ou Emotionnelles, Contrat DRET 84111155 - Rapport de Synth~se, 1987. 30 Vernet-Maury, E., Sicard, G., Dittmar, A. and DeschaumesMolinaro, C., Autonomic nervous system preferential responses, Activ. Nerv. Super., 32 (1990) 37-38. 31 Wallin, B.G. and Fagius, J., The sympathetic nervous system in man. Aspects derived from microelectrode recordings, Trends Neurosci., 9 (1986) 63-66. 32 White, K.D., Ashton, R. and Lewis, S., Learning a complex skill: effects of mental practice, physical practice and imagery ability, hit. J. Sport PhysioL, I0 (1979) 71-'18. 33 Wilder, J., Basimetric approach (law ofinitial value) to biological rhythms, Ann. N.Y. Acad. Sci., 98 (1962) 1211-1228.
