International Elsevier
PSYCHO
Journal of Psychophysiology.
10 (1990) 199-202
199
00313
Visuomotor tracking performance and task-induced modulation of alpha activity G. Ulrich and R. Kriebitzsch Loboratory of Clinical Psychophysiology,
Psychiatrische (Accepted
Klinik und Poliklinik, Freie UniuersitiitBerlin, Berlin (E R. G.) 1 June 1990)
This study is an attempt at clarifying the relationships between defined environmental demands, the efficiency of corresponding visuomotor behaviour and the EEG. We made use of a laboratory paradigm which was devised to closely represent a common man-machine system process, i.e., maintaining of a road vehicle on a predetermined track when squally wind is blowing from the flank. The deviation of the ideal track represents the input signal for the component system man, the compensating steering motion its output signal. This performance is modelled by a visuomotor tracking task which has been developed as a means of ascertaining subtle effects of psychotropic drugs and described elsewhere. We found that the higher the efficiency of test performance, the lower was the task-induced modulation of alpha-activity.
The subject is confronted with a video screen (distance nasion-centre of the screen: 1.35 m) displaying a target signal, a cross which moves continuously in a stochastic manner in a horizontal plane within a visual angle of 15 degrees, the time course of the target signal being programmed by a pseudo random noise generator. The subject’s task is to closely pursue this target signal with a response signal (an arrow) displayed below the target signal. The response can be controlled by a joy-stick which is fixed upon the right arm of the recording chair and can be moved easily by the finger tips. The mean efficiency of test performance can be determined exactly (bit/second) as the amount of transinformation between target and response signal according to Schweizer (1970)
Correspondence: G. Ulrich, Freie Universitlt Dept. of Clinical Psychiatry, Eschenallee (West) 19, F.R.G. 0167~8760/90/%03.50
Berlin UKRV/C, 3, D-1000 Berlin
0 1990 Elsevier Science Publishers
(see Fig. (1963).
IF=
l),
(f2-f&Id
using
the
formula
by
Fano
$f*&,df fi
//jl(s,, -H2.
IF
given
Sx)df
= transinformation between target and response f2 = upper frequency limit of the target signal (0.50 Hz) f, = lower frequency limit of the target signal (0.05 Hz) Id = dual logarithm df = differential; -for the other abbreviations we may refer to Fig. 1. For further methodological and technical details as well as for clinical applications we refer to Hayashi and Furusho (1966), Knebitzsch et al. (1978), Bonte et al. (1978), Ulrich and Kriebitzsch (1987) and Brau and Ulrich (1990).
B.V. (Biomedical
Division)
200 self-generated component2 visual input e 0) @T--- ee
___-_-_______
r
1
I
Spp=Skk-H
I
$e
manual response signal
I
I
transfer
I
Hz+-
1 t
function
CC3
S ee I
1 I
I I
k 0)
component Hi.See
L---_---_______:
e(t),k(t)
signals measured at@ auto spectrum of e (t) auto spectrum of k (t) cross spectrum of e (t)
S ee Skk
S ek Fig. 1. Assessment
of information
flow according
to Schweizer
Subjects were 12 unpaid healthy males with self-reported right-handedness, age 25-36 years (mean: 28.5 years) who were recruited from the staff of our department. The subjects were given the opportunity to become familiar with the laboratory setting and to practice the tracking task as long as it seemed necessary to them. Silver cup electrodes were placed according to the lo-20 scheme. We recorded bipolarly from 01-Czl and 02-Cz2. Czl and Cz2 represent the positions at a distance of 1 cm from Cz being situated upon a connecting line between Cz and 01 or 02, respectively. The electrode skin impedance was not allowed to exceed 5 KO and was checked before each run (time constant: 0.3 s. low pass filter at 70
TABLE
k
(1970) using the formula
(t) (below)
given by Fano (1963).
cps). The EEG was recorded and amplified by a Siemens-Elema Mingograph 21. To get the subjects accustomed to the recording situation we started with 3 min under resting conditions. Thereafter, the subjects had to open their eyes and to perform the tracking task which lasted exactly 50 s. The time functions of target and response signals as well as the simultaneously recorded EEG were both written on paper chart and stored on FM tape. An example is shown in Fig. 2. After visual checking the paper charts for artifacts (no subject had to be ommitted) the EEGs were subjected to analogue filtering (Krohn-Hite-filter 2 X 48 dB/octave). All spectral coefficients, except the 7-13 cps alpha-band,
I
Mean rrackq performance (TP) m hit/s time course of the RMS values Alpha:
and
7-13
cps. time resolution:
and wean spectral coherence (CoTA)
between the angular r~eloclty of the target .vgnul und the
0.5 s.
Subjects
TP CoTA
(bit/s)
1
2
3
4
5
6
7
8
9
10
11
12
x
SD.
3.43
2.59
3.44
3.67
4.11
3.54
3.07
2.93
2.22
2.72
3.65
2.97
3.20
0.54
01 -Czl
0.183
0.154
0.172
0.119
0.131
0.071
0.301
0.151
0.202
0.208
0.097
0.270
0.172
0.07
02-Cz2
0.162
0.208
0.173
0.140
0.122
0.073
0.345
0.230
0.297
0.147
0.176
0.227
0.192
0.08
201
Fig. 2. Sample
recording
of target indicates
and response signals and the simultaneously a movement to the right, downward deflection
were eliminated. After digitizing with a sampling frequency of 64 cps the root-mean-square values (RMS, pV) were calculated for successive 0.50 s epochs. To account for the task-induced modulation of ongoing alpha activity we determined the mean spectral coherence (Bendat and Piersol, 1971) within the spectral range of 0.05-0.50 cps (epoch length = 20 s: spectral resolution = 0.05 cps, %lax = + 10 s; sampling rate = 2/s: time resolution = 0.50 s, upper frequency limit = 1 cps), using the cross-correlation functions resulting from the angular velocity of the target signal (O/s) and
recorded EEG. Upward a movement to the left.
deflection
of the signals
the time course of the root-mean-square values of the 7-13 cps alpha activity. The mean efficiency of test performance was calculated for each subject as mentioned above. The possible relationship between test performance and task-induced modulation of alpha activity was determined using the rank correlations. The results showed significant negative correlations (TP and CoTA of lead 01-Czl: r, = -0.61, P = 0.04; TP and CoTA of lead 02-Cz2: c, = -0.64, P = 0.03; two-tailed). Thus, the higher the efficiency of test performance the lower was the
202
task-induced modulation of alpha-activity. As far as we survey the literature no comparable research has been done. Thus, until these results were replicated any attempt at an interpretation seems to be premature. Nevertheless, it may be allowed to point out that the present finding indicates that central nervous systems are the more efficient the less their functioning is governed by signals from their environment. Such an explanation does not easily fit to the generally accepted model of brains as input controlled information processors. It rather supports a conception of biological systems as autonomously functioning entities, which are structurally determined, i.e. environmental signals are acting as disturbances only (Maturana, 1975; Varela, 1979). For a thorough discussion of the epistemological difficulties we may refer to Werner (1988).
REFERENCES
Bendat, J.S. and Piersol, A.G. (1971) Random data: analysis and measurement procedures. Wiley-Interscience. New York. Bente, D., Chenchanna, P., Scheuler, W. and Sponagel, P. (1978) Psychophysiologische Studien zum Verhalten der
hirnelektrischen Wachaktivitlt bei definierter Vigilanzbeanspruchung. 3. Mitteilung: Zur Erfassung pharmakogener Effekte auf die himelektrische Aktivitat beim Fahrverhalten und die Optimiertmg des Systems FahrerrFahrzeug-Strasse. Z. EEG- EMG, 9: 61-73. Brlu, H. and Uhich, G. (1990) Electroencephalographic vigilance dynamics in multiple sclerosis during an acme episode and after remission. Eur. Arch. Psych&r. Neural. Ser., 239: 320-324. Fano, R.M. Transmission of Information (1963) M.I.T. Press, Cambridge/MA. Hayashi, M. and Furusho, H. (1966) The Response of Automobile Against a Gust. FISITA-Kongress, Miinchen, B?. Kriebitzsch, R., Bente, D. and Schemer, W. (1978) Ein verhaltensphysiologischer Messplatz zur Untersuchung des optomotorischen Folgeund Regelverhaltens. Biomed. Tech., 23: 147-148. Maturana, H. (1975) The organization of the Living: a theory of a living organization. Int. J. Man- Machine Studies. 7: 313-332. Schweizer, G. (1970) Probleme und Methoden zur Untersuchung des Regelverhahens des Menschen. In W. Oppelt and G. Vossius (Eds.), Der Mensch als Regler. VEB Verlag Technik, Berlin (Ost), pp. 159-238. Uhich, G. and Kriebitzsch, R. (1987) Ein rechnergestiitztes visuomotorisches Tracking-Verfahren znr trennscharfen Objektivierung zentralnervoser Pharmakoneffekte. Arzneim.-Forsch. 37: 472-475. Varela, F.J. (1979) Principles of Biological Autonomy, Elsevier. New York. Werner, G. (1988) The many faces of neuroreductionism. In E. Basar (Ed.), Dynamics of Sensory and Cognitive Processing by the Brain. Springer, Berlin, pp. 241-257.
PSYCHO
Journal of Psychophysiology.
10 (1990) 199-202
199
00313
Visuomotor tracking performance and task-induced modulation of alpha activity G. Ulrich and R. Kriebitzsch Loboratory of Clinical Psychophysiology,
Psychiatrische (Accepted
Klinik und Poliklinik, Freie UniuersitiitBerlin, Berlin (E R. G.) 1 June 1990)
This study is an attempt at clarifying the relationships between defined environmental demands, the efficiency of corresponding visuomotor behaviour and the EEG. We made use of a laboratory paradigm which was devised to closely represent a common man-machine system process, i.e., maintaining of a road vehicle on a predetermined track when squally wind is blowing from the flank. The deviation of the ideal track represents the input signal for the component system man, the compensating steering motion its output signal. This performance is modelled by a visuomotor tracking task which has been developed as a means of ascertaining subtle effects of psychotropic drugs and described elsewhere. We found that the higher the efficiency of test performance, the lower was the task-induced modulation of alpha-activity.
The subject is confronted with a video screen (distance nasion-centre of the screen: 1.35 m) displaying a target signal, a cross which moves continuously in a stochastic manner in a horizontal plane within a visual angle of 15 degrees, the time course of the target signal being programmed by a pseudo random noise generator. The subject’s task is to closely pursue this target signal with a response signal (an arrow) displayed below the target signal. The response can be controlled by a joy-stick which is fixed upon the right arm of the recording chair and can be moved easily by the finger tips. The mean efficiency of test performance can be determined exactly (bit/second) as the amount of transinformation between target and response signal according to Schweizer (1970)
Correspondence: G. Ulrich, Freie Universitlt Dept. of Clinical Psychiatry, Eschenallee (West) 19, F.R.G. 0167~8760/90/%03.50
Berlin UKRV/C, 3, D-1000 Berlin
0 1990 Elsevier Science Publishers
(see Fig. (1963).
IF=
l),
(f2-f&Id
using
the
formula
by
Fano
$f*&,df fi
//jl(s,, -H2.
IF
given
Sx)df
= transinformation between target and response f2 = upper frequency limit of the target signal (0.50 Hz) f, = lower frequency limit of the target signal (0.05 Hz) Id = dual logarithm df = differential; -for the other abbreviations we may refer to Fig. 1. For further methodological and technical details as well as for clinical applications we refer to Hayashi and Furusho (1966), Knebitzsch et al. (1978), Bonte et al. (1978), Ulrich and Kriebitzsch (1987) and Brau and Ulrich (1990).
B.V. (Biomedical
Division)
200 self-generated component2 visual input e 0) @T--- ee
___-_-_______
r
1
I
Spp=Skk-H
I
$e
manual response signal
I
I
transfer
I
Hz+-
1 t
function
CC3
S ee I
1 I
I I
k 0)
component Hi.See
L---_---_______:
e(t),k(t)
signals measured at@ auto spectrum of e (t) auto spectrum of k (t) cross spectrum of e (t)
S ee Skk
S ek Fig. 1. Assessment
of information
flow according
to Schweizer
Subjects were 12 unpaid healthy males with self-reported right-handedness, age 25-36 years (mean: 28.5 years) who were recruited from the staff of our department. The subjects were given the opportunity to become familiar with the laboratory setting and to practice the tracking task as long as it seemed necessary to them. Silver cup electrodes were placed according to the lo-20 scheme. We recorded bipolarly from 01-Czl and 02-Cz2. Czl and Cz2 represent the positions at a distance of 1 cm from Cz being situated upon a connecting line between Cz and 01 or 02, respectively. The electrode skin impedance was not allowed to exceed 5 KO and was checked before each run (time constant: 0.3 s. low pass filter at 70
TABLE
k
(1970) using the formula
(t) (below)
given by Fano (1963).
cps). The EEG was recorded and amplified by a Siemens-Elema Mingograph 21. To get the subjects accustomed to the recording situation we started with 3 min under resting conditions. Thereafter, the subjects had to open their eyes and to perform the tracking task which lasted exactly 50 s. The time functions of target and response signals as well as the simultaneously recorded EEG were both written on paper chart and stored on FM tape. An example is shown in Fig. 2. After visual checking the paper charts for artifacts (no subject had to be ommitted) the EEGs were subjected to analogue filtering (Krohn-Hite-filter 2 X 48 dB/octave). All spectral coefficients, except the 7-13 cps alpha-band,
I
Mean rrackq performance (TP) m hit/s time course of the RMS values Alpha:
and
7-13
cps. time resolution:
and wean spectral coherence (CoTA)
between the angular r~eloclty of the target .vgnul und the
0.5 s.
Subjects
TP CoTA
(bit/s)
1
2
3
4
5
6
7
8
9
10
11
12
x
SD.
3.43
2.59
3.44
3.67
4.11
3.54
3.07
2.93
2.22
2.72
3.65
2.97
3.20
0.54
01 -Czl
0.183
0.154
0.172
0.119
0.131
0.071
0.301
0.151
0.202
0.208
0.097
0.270
0.172
0.07
02-Cz2
0.162
0.208
0.173
0.140
0.122
0.073
0.345
0.230
0.297
0.147
0.176
0.227
0.192
0.08
201
Fig. 2. Sample
recording
of target indicates
and response signals and the simultaneously a movement to the right, downward deflection
were eliminated. After digitizing with a sampling frequency of 64 cps the root-mean-square values (RMS, pV) were calculated for successive 0.50 s epochs. To account for the task-induced modulation of ongoing alpha activity we determined the mean spectral coherence (Bendat and Piersol, 1971) within the spectral range of 0.05-0.50 cps (epoch length = 20 s: spectral resolution = 0.05 cps, %lax = + 10 s; sampling rate = 2/s: time resolution = 0.50 s, upper frequency limit = 1 cps), using the cross-correlation functions resulting from the angular velocity of the target signal (O/s) and
recorded EEG. Upward a movement to the left.
deflection
of the signals
the time course of the root-mean-square values of the 7-13 cps alpha activity. The mean efficiency of test performance was calculated for each subject as mentioned above. The possible relationship between test performance and task-induced modulation of alpha activity was determined using the rank correlations. The results showed significant negative correlations (TP and CoTA of lead 01-Czl: r, = -0.61, P = 0.04; TP and CoTA of lead 02-Cz2: c, = -0.64, P = 0.03; two-tailed). Thus, the higher the efficiency of test performance the lower was the
202
task-induced modulation of alpha-activity. As far as we survey the literature no comparable research has been done. Thus, until these results were replicated any attempt at an interpretation seems to be premature. Nevertheless, it may be allowed to point out that the present finding indicates that central nervous systems are the more efficient the less their functioning is governed by signals from their environment. Such an explanation does not easily fit to the generally accepted model of brains as input controlled information processors. It rather supports a conception of biological systems as autonomously functioning entities, which are structurally determined, i.e. environmental signals are acting as disturbances only (Maturana, 1975; Varela, 1979). For a thorough discussion of the epistemological difficulties we may refer to Werner (1988).
REFERENCES
Bendat, J.S. and Piersol, A.G. (1971) Random data: analysis and measurement procedures. Wiley-Interscience. New York. Bente, D., Chenchanna, P., Scheuler, W. and Sponagel, P. (1978) Psychophysiologische Studien zum Verhalten der
hirnelektrischen Wachaktivitlt bei definierter Vigilanzbeanspruchung. 3. Mitteilung: Zur Erfassung pharmakogener Effekte auf die himelektrische Aktivitat beim Fahrverhalten und die Optimiertmg des Systems FahrerrFahrzeug-Strasse. Z. EEG- EMG, 9: 61-73. Brlu, H. and Uhich, G. (1990) Electroencephalographic vigilance dynamics in multiple sclerosis during an acme episode and after remission. Eur. Arch. Psych&r. Neural. Ser., 239: 320-324. Fano, R.M. Transmission of Information (1963) M.I.T. Press, Cambridge/MA. Hayashi, M. and Furusho, H. (1966) The Response of Automobile Against a Gust. FISITA-Kongress, Miinchen, B?. Kriebitzsch, R., Bente, D. and Schemer, W. (1978) Ein verhaltensphysiologischer Messplatz zur Untersuchung des optomotorischen Folgeund Regelverhaltens. Biomed. Tech., 23: 147-148. Maturana, H. (1975) The organization of the Living: a theory of a living organization. Int. J. Man- Machine Studies. 7: 313-332. Schweizer, G. (1970) Probleme und Methoden zur Untersuchung des Regelverhahens des Menschen. In W. Oppelt and G. Vossius (Eds.), Der Mensch als Regler. VEB Verlag Technik, Berlin (Ost), pp. 159-238. Uhich, G. and Kriebitzsch, R. (1987) Ein rechnergestiitztes visuomotorisches Tracking-Verfahren znr trennscharfen Objektivierung zentralnervoser Pharmakoneffekte. Arzneim.-Forsch. 37: 472-475. Varela, F.J. (1979) Principles of Biological Autonomy, Elsevier. New York. Werner, G. (1988) The many faces of neuroreductionism. In E. Basar (Ed.), Dynamics of Sensory and Cognitive Processing by the Brain. Springer, Berlin, pp. 241-257.
