On the coordination of motor output during visual flight control of flies

(1)

J Comp Physiol A (1991) 169:127-134

d o u r l m l o f

Neural, mid

l hy olow A

9 Springer-Verlag 1991

On the coordination of motor output during visual flight control of flies

Johannes M. Zanker, Martin Egelhaaf, and Anne-Kathrin Warzecha

Max-Planck-Institut fiir biologische Kybernetik, Spemannstrage 38, W-7400 Tiibingen, Federal Republic of Germany Accepted June 17, 1991

Summary. In tethered flying houseflies (Musca domes- tica), the yaw torque produced by the wings is accompa- nied by postural changes of the abdomen and hindlegs.

In free flight, these body movements would jointly lead to turning manoeuvres of the animal. By recording the yaw torque together with the lateral deflections of either the abdomen or the hindlegs, it is shown that these motor output systems act in a highly synergistic way during two types of visual orientation behavior, compensatory optomotor turning reactions and orientation turns elicited by moving objects. This high degree of coordination is particularly conspicuous for the pathway activated by moving objects. Here, orientation responses either may be induced or may fail to be generated always simultaneously in all three motor output systems. This suggests that the pathway mediating orientation turns towards objects is gated before it segregates into the respective motor control systems of the wings, the abdomen and the hindlegs.

Key words: Visual orientation - Optomotor response Motor control - Fly - Motion

Introduction

A flying animal easily may deviate from straight course in two situations. First, external disturbances such as turbulences of the air or internal asymmetries in the flight motor may force the animal to depart from its course, and the animal should try to compensate for it. Second, the animal actively may turn towards stationary or moving objects in its visual field. Both compensatory optomotor turns and turning reactions towards objects are at least partially under visual control. They are elicited by specific type of image flow on the retina. The strongest compensatory optomotor responses are induced by rotation of the complete retinal image, such as

experienced by the animal during deviations from its flight course. In contrast, orientation turns towards objects may be elicited by displacements of small parts of the retinal image. This retinal flow occurs when the animal passes a nearby stationary or moving object in front of a more distant background.

How these different retinal motion patterns are trans- formed by the fly in either compensatory optomotor turning responses or orientation responses towards objects has been analyzed in some detail in behavioral and neurophysiological experiments. There is now good evi- dence that two parallel control systems are involved which differ in their sensitivity to the size of the moving stimulus (Geiger and N/issel 1982; G6tz 1983a; Heisen- berg and Wolf 1984; Egelhaaf 1985a-c; Bausenwein et al. 1986; Egelhaaf et al. 1988; Egelhaaf 1989; Reichardt et al. 1989; Hausen and Wehrhahn 1990). In the housefly Musca and blowfly Calliphora, the control system that mediates compensatory optomotor turning reactions ("large-field system") responds best to extended binocu- lar stimulus patterns rotating around roughly the vertical axis of the animal. The HS-cells in the third visual ganglion have been concluded to be the corresponding output elements of the visual system (Hausen 1982a, b; for review, see Egelhaaf et al. 1988). The control system that mediates orientation turns towards objects ("small-field system") is tuned to retinal image displacements of small objects. The FD-cells that reside also in the third visual ganglion apparently represent the neuronal analogue of this control system at the output level of the visual system (Egelhaaf 1985b, c; Egelhaaf 1990). In addition to the size of the moving stimulus, the relative contribution of the large-field and small-field system to yaw torque is also influenced by the dynamic properties of stimulus motion.

A comparison of the HS- and FD-cell features with the behavioral responses suggests that high frequency modulations in the output signals of the HS-cells are attenu- ated somewhere between the lobula plate and the final motor output. In contrast, the FD-cells mediate yaw torque also at high oscillation frequencies (Egelhaaf

1987; Egelhaaf and Borst 1990, 1991).

(2)

128 J.M. Zanker et al. : Coordination of motor output in visual orientation of flies H o w a r e t h e s e r e p r e s e n t a t i o n s o f r e t i n a l m o t i o n p a t -

terns t r a n s f o r m e d i n t o the different t y p e s o f t u r n i n g r e a c t i o n s ? I n c o n t r a s t to l o c u s t s ( R o w e l l 1988), t h e r e is n o t m u c h k n o w n in this r e s p e c t a t the c e l l u l a r level, so far, in t h e fly. H o w e v e r , t h r e e m o t o r o u t p u t s y s t e m s h a v e b e e n s u g g e s t e d to c o n t r i b u t e to t u r n s a b o u t the s o - c a l l e d y a w axis o f flight c o n t r o l w h i c h is i n c l i n e d a b o u t 30 ~ r e l a t i v e to the v e r t i c a l b o d y axis o f the fly ( G 6 t z et al.

1979; Z a n k e r 1988): (i) D i f f e r e n c e s b e t w e e n the m e a n s t r o k e a m p l i t u d e s o f the t w o wings, (ii) a b d o m i n a l deflections, a n d (iii) d e f l e c t i o n s o f the h i n d legs. D o these v a r i o u s m o t o r o u t p u t s y s t e m s act i n d e p e n d e n t l y o r s y n e r g i s t i c a l l y in the different t u r n i n g m a n o e u v r e s ? T o w h a t e x t e n t are t h e y c o o r d i n a t e d ? F o r e x a m p l e , w h e n a p a t t e r n is m o v i n g f r o m the left to the r i g h t in f r o n t o f the fly, the a v e r a g e w i n g b e a t a m p l i t u d e is i n c r e a s e d o n the left a n d d e c r e a s e d o n the r i g h t side. I n free flight, the r e s u l t i n g y a w t o r q u e will be s u p p o r t e d b y u s i n g f r i c t i o n a l a n d g r a v i t a t i o n a l f o r c e c o m p o n e n t s w h e n the a b d o m e n a n d h i n d legs a r e b e n t s i m u l t a n e o u s l y to the right side.

I f these o u t p u t s y s t e m s w e r e n o t c o o r d i n a t e d , the t o r q u e elicited b y o n e o f t h e m w o u l d be a l t e r e d b y a c c i d e n t a l signal f l u c t u a t i o n s in the o t h e r s .

H e r e , we e x a m i n e the c o o r d i n a t i o n o f the 3 m o t o r o u t p u t s y s t e m s c o n t r i b u t i n g to c o m p e n s a t o r y t u r n i n g r e a c t i o n s a n d to o r i e n t a t i o n t u r n s t o w a r d s o b j e c t s in M u s c a . Since t h e large-field a n d s m a l l - f i e l d s y s t e m w h i c h c o n t r o l these t w o t y p e s o f t u r n i n g r e s p o n s e s differ m a i n l y in t h e i r d y n a m i c a l a n d s p a t i a l i n t e g r a t i o n p r o p e r t i e s , we t u n e d specific stimuli in size a n d d y n a m i c s to a d d r e s s e i t h e r o f the t w o c o n t r o l systems.

Materials and methods

Wild type female houseflies Musca domestica (L.) from laboratory stocks were prepared as described previously (Fermi and Reichardt 1963). The head of the fly was fixed to the thorax with a mixture of wax and colophonium under light carbon dioxide anesthesia.

A triangular piece of cardboard was glued to the wax just above the frontal part of the thorax. The ocelli were covered with the same mixture of wax and colophonium.

Three types of motor output were investigated. Torque: The flies were suspended from a torque compensator which prevented both rotatory and translatory movements of the animal and allowed direct measurement of the yaw torque generated by the animal (e.g.

Fermi and Reichardt 1963; G6tz 1964). Yaw torques measured under these conditions are mainly due to variations of the average wing beat, because in tethered flight in still air inertial or frictional forces of the body are not effective. This notion is supported by the observation that, under the stimulus conditions as used in the present study, essentially the same yaw torque responses are generated as in intact animals when the hindlegs were amputated and the abdomen was tightly fixed with wax to the thorax (Egelhaaf, unpublished observations). Abdominal deflections: The position of the abdomen tip in the equatorial plane was monitored by means of a light barrier (for details, see Zanker 1988). The light barrier was calibrated by displacing the entire fly over a known distance before and after the experiment. Le9 deflections." A linescan camera (TS- Opto 7120) was combined with an electronic device converting the position of a selected contour in the camera's field of view into an analogue signal. This recording technique proved to be sufficiently sensitive for monitoring the position of a single leg (in our experiments the right leg) during tethered flight.

The analogue output signals of the torque compensator, the light barrier or the linescan camera were AD-converted (DT2801) and fed into a computer (IBM-AT) and further processed as will be described in Results. For technical reasons, it was not possible to monitor simultaneously all 3 motor output variables described before. Therefore, the motor output variables were registered only in pairs, i.e. the yaw torque together with either the abdominal or the hind leg deflections.

The visual stimulation was the same as described in detail in an earlier paper (Egelhaaf 1989). In brief, the animal was positioned in the center of two concentric pattern cylinders. The outer cylinder ("ground") was opened in its rear to allow access to the abdomen and hind legs with the light barrier and linescan camera, respectively. This background pattern extended to 240 ~ in the horizontal plane of the fly's visual field. The inner stimulus pattern consisted of a cylinder segment ("figure") of 10 ~ angular width. Both, figure and ground had a vertical angular extent of 42 ~ They were covered with a vertical square-wave grating with a spatial wavelength of its fundamental of 10 ~ The fly was alternately stimulated by synchronous sinusoidal oscillations of figure and ground ("large-field motion") and by the figure oscillating alone while the background was kept stationary ("small-field motion"). While the large-field stimulus covered both eyes symmetrically, the vertical stripe mim- icking small-field motion was oscillated usually in front of the right eye about a mean position of 20 ~ with respect to the frontal midline of the animal. In the experiments shown here, the oscillation amplitude was _+ 10 ~ the oscillation frequency either 0.1 Hz or 1 Hz.

The experiments were carried out under open-loop conditions, i.e.

the responses of the fly did not affect the visual stimulus.

Results

I n a first set o f e x p e r i m e n t s , it w a s i n v e s t i g a t e d to w h a t e x t e n t the l a t e r a l d e f l e c t i o n s o f the a b d o m e n a n d the y a w t o r q u e g e n e r a t e d b y the w i n g s a r e c o o r d i n a t e d d u r i n g different t y p e s o f v i s u a l o r i e n t a t i o n r e s p o n s e s . F i g . 1 s h o w s the t i m e c o u r s e o f these r e s p o n s e s d u r i n g oscilla- t o r y p a t t e r n m o t i o n a t f r e q u e n c i e s o f 0.1 H z (left) a n d 1 H z (right). I n i t i a l l y , figure a n d g r o u n d were m o v e d s y n c h r o n o u s l y (large-field m o t i o n ) for 3 o s c i l l a t i o n cycles. T h e n the g r o u n d s t o p p e d m o v i n g , while the figure c o n t i n u e d o s c i l l a t i n g (small-field m o t i o n ) f o r a n o t h e r 3 cycles.

L e t us first c o n s i d e r the t o r q u e r e s p o n s e . D u r i n g large-field m o t i o n , w h e n the flies f o l l o w the p a t t e r n m o - tion, the y a w t o r q u e c a n be a s s u m e d to be s y m m e t r i c a l a r o u n d z e r o ( F i g . l ) . T h i s o p t o m o t o r r e s p o n s e w o u l d r e d u c e the r e l a t i v e a n g u l a r v e l o c i t y b e t w e e n the s t i m u l u s a n d the eyes, if the fly w a s n o t k e p t s t a t i o n a r y b y the t o r q u e m e t e r . D u r i n g s m a l l - f i e l d m o t i o n t h e t o r q u e signal n o l o n g e r o s c i l l a t e s a r o u n d zero. I n s t e a d , t o r q u e is c h a n g e d such t h a t , o n a v e r a g e , u n d e r c l o s e d - l o o p c o n - d i t i o n s the figure w o u l d be shifted to the f r o n t o f the eyes; this m e a n s t h a t it is p o s i t i v e w h e n the figure is p l a c e d in f r o n t o f the r i g h t eye. T h e r e s p o n s e a m p l i t u d e s differ c o n s i d e r a b l y f o r the t w o o s c i l l a t i o n frequencies. A t the low o s c i l l a t i o n f r e q u e n c y , the r e s p o n s e to large-field m o t i o n is m u c h s t r o n g e r t h a n to s m a l l - f i e l d m o t i o n . I n c o n t r a s t , at the high o s c i l l a t i o n f r e q u e n c y , the r e s p o n s e is m o r e p r o n o u n c e d to s m a l l - f i e l d m o t i o n t h a n to l a r g e - field m o t i o n . T h e s e results a r e i n d i s t i n g u i s h a b l e f r o m e a r l i e r e x p e r i m e n t s p e r f o r m e d u n d e r s i m i l a r s t i m u l u s c o n d i t i o n s ( E g e l h a a f 1987, 1989).

(3)

J.M. Zanker et al.: Coordination of motor output in visual orientation of flies 129

0.1 Hz 1 Hz

Torc

10-7 Nm

A b d o m e n

0.5 mm

10 -7 Nm

0.5 mm

Time

Fig. 1. Simultaneously recorded yaw torque responses (upper traces) and abdomen deflections (middle traces). The flies were stimulated by oscillatory motion of a cylindrical stripe pattern (the

"ground" G) and a vertical cylinder segment (the "figure" F). F was placed in front of the right eye at a mean position of 20 ~ with respect to the frontal midline of the animal. The patterns were oscillated with a frequency of either 0.1 Hz (left diagrams) or 1 Hz (right diagrams). The oscillation amplitude was 10 ~ Initially F and G were moved synchronously for 3 cycles ("large-field motion"). Then G stopped moving, while F continued oscillating for another two cycles (small-field motion). Deviations of F and G from their respective mean positions are plotted in the bottom traces. Upward and downward slopes indicate clockwise and counter-clockwise motion, respectively. In the yaw torque response traces, upward and downward deflections indicate intended clockwise and counter-clockwise turns, respectively. Upward and downward deflections in the abdomen response traces indicate counter-clockwise and clockwise turns, respectively. The data are averages obtained from 12 flies and a total of 135 (0.1 Hz) and 318 (1 Hz) stimulus presentations. (The

peak-to-peak amplitudes of the responses have the following S.E.M.s: Yaw torque (given in 10 v Nm): large-field motion, 0.1 Hz: 0.15, 1 Hz: 0.09; small-field motion, 0.1 Hz: 0.07, 1 Hz:

0.26. Abdomen deflections (given in mm): large-field motion, 0.1 Hz: 0.22, 1 Hz: 0.03; small-field motion, 0.1 Hz: 0.14, 1 Hz:

0.12). Note the different time scales in the left and right diagrams.

The kink in the middle of the stimulus trace does not correspond to a sudden jump of F, but rather is the consequence of the fact that no data were acquired while the computer triggered the motor control of the pattern to stop the ground moving. During oscillatory large- field motion both yaw torque and abdomen deflections oscillate synchronously with the pattern motion. During small-field motion and, in particular, at high oscillation frequencies the fly tries to turn towards the position of the figure. Response amplitudes are big during large-field motion at 0.1 Hz and small-field motion at 1 Hz.

Apart from minor differences (see text), the yaw torque and abdomen responses, under all stimulus conditions tested here, have rather similar time courses and relative amplitudes

T h e h o r i z o n t a l d e v i a t i o n o f the a b d o m e n tip f r o m its m e a n position is displayed in the middle traces o f Fig. 1.

D u r i n g b o t h , large-field a n d small-field m o t i o n , a n d at b o t h oscillation frequencies, the a b d o m e n is bent b a c k a n d forth periodically with the oscillation f r e q u e n c y o f the stimulus. F o r instance, the a b d o m e n turns c o u n t e r - clockwise d u r i n g a clockwise t u r n o f the animal, as ex- pected f o r an a e r o d y n a m i c rudder. U n d e r o u r stimulus c o n d i t i o n s the overall response profiles o f the y a w t o r q u e a n d the c o r r e s p o n d i n g lateral b e n d i n g o f the a b d o m e n l o o k very similar. T h e amplitudes are large d u r i n g small- field m o t i o n at high oscillation frequencies as well as d u r i n g large-field m o t i o n at low frequencies a n d c o m - paratively small f o r the o t h e r conditions. H o w e v e r , despite this overall similarity, there are m i n o r differences between the two m o t o r o u t p u t variables. T h e a m p l i t u d e o f the a b d o m i n a l response, as c o m p a r e d to torque, is s o m e w h a t larger f o r the low oscillation f r e q u e n c y t h a n f o r the high oscillation frequency. F u r t h e r m o r e , the characteristic response peaks o f the lateral a b d o m i n a l deflections at high oscillation frequencies are slightly

delayed with respect to the c o r r e s p o n d i n g t o r q u e responses (78 ms_+ 14.8 ms S E M ; 12 flies with 3 peaks each), a n d the response peaks seem to be s o m e w h a t broader. All these peculiarities o f the a b d o m i n a l response can be easily u n d e r s t o o d as a c o n s e q u e n c e o f the big inertial mass o f the a b d o m e n . This e x p l a n a t i o n is in a c c o r d a n c e with earlier c o n s i d e r a t i o n s on the d y n a m i c s o f a b d o m i n a l deflections in

Drosophila

( Z a n k e r 1988).

I n a s e c o n d set o f experiments, the hind leg deflections were investigated u n d e r the same stimulus c o n d i t i o n s (Fig. 2). T h e s i m u l t a n e o u s l y r e c o r d e d y a w t o r q u e traces are essentially the same as in the p r e v i o u s set o f data. T h e quantitative differences in the a m p l i t u d e s o f the y a w t o r q u e responses displayed in Figs. 1 a n d 2 are well within the range o f variability also f o u n d in earlier studies ( E g e l h a a f 1987, 1989). T h e middle traces o f Fig. 2 s h o w the m o d u l a t i o n s o f the h o r i z o n t a l hind leg position.

D u r i n g large-field a n d small-field m o t i o n a n d at b o t h oscillation frequencies, the hind legs are deflected periodically with the oscillation f r e q u e n c y o f the stimulus, again in the direction opposite to the t u r n which w o u l d be

(4)

130

0.1 Hz

[ l

T o r e

J.M. Zanker et al.: Coordination of motor output in visual orientation of flies 1 H Z

1 0 -7 Nm

e f l e c t i o ~

0.25

mm

Time

Fig. 2. Simultaneously recorded yaw torque responses (upper traces) and deflections of the right hindlcg (middle traces). Stimulus conditions as described in the legend of Fig. 1. In the leg response traces, upward and downward deflections indicate movements of the legs to the right and left, respectively. The data are averages obtained from 29 flies and a total of 312 (0.1 Hz) and 407 (1 Hz) stimulus presentations. (The peak-to-peak amplitudes of the responses have the following S.E.M.s: Yaw torque (given in

0.25

mm

10 -v Nm): large-field motion, 0.1 Hz: 0.17, 1 Hz: 0.10; small-field motion, 0.1 Hz: 0.08, 1 Hz: 0.09. Hind leg deflections (given in ram): large-field motion, 0.1 Hz: 0.08, 1 Hz: 0.04; small-field motion, 0.1 Hz: 0.04.1 Hz: 0.05). The response traces, within the usual range of variability, resemble the abdomen responses shown in Fig. 1. The hind legs are deflected synchronously with the yaw torque, and the response amplitudes vary in the same way with the stimulus conditions

generated by a freely flying fly. During large-field motion at the low oscillation frequency and during small-field motion at the high frequency the response amplitude is large, as compared to the other stimulus conditions.

Despite this overall similarity with the two other m o t o r output variables, one peculiarity o f the leg deflections should be noted. At the low oscillation frequency, the average leg position during small-field motion is closer to the body o f the fly (i.e. attains a smaller value, see Fig. 2) as during large-field motion. In contrast, the yaw torque response (and abdominal deflection, cf. Fig. 1) is shifted to higher values, on average. This difference may be less surprising when we assume that the average position o f the leg during large-field stimulation is not neces- sarily the same as during straight flight, as could be concluded for the abdominal and yaw responses based on symmetry properties. I f the zero level would be close to the body and, thus, deflections away from the body would be exaggerated, the average response during small-field motion could be smaller than during large- field motion. In fact, this explanation is supported by the observation that the leg position during straight flight is very close to the body, and that deflections away from the body are much more p r o n o u n c e d than towards it (unpublished observations).

Despite these differences in details of the response profiles o f the various m o t o r outputs, the high degree of correlation in their overall structure during visual stimulation suggests that they are controlled in the same way by visual input. So far, we have been concerned with

responses averaged over many stimulus presentations and several flies. However, the high degree o f coordination o f yaw torque response and postural changes o f the abdomen and legs can also be observed in single recordings, although in this case the visually induced responses may be superimposed by considerable spontaneous fluctuations which need not co-vary for the different output systems. The correlation o f the visually induced torque responses and the corresponding deflections o f the a b d o m e n and hind legs will be considered here only for the turning responses towards objects. Figures 3 and 4 show examples o f torque responses to large-field and small-field m o t i o n at an oscillation frequency o f 1 Hz, together with the corresponding abdominal (Fig. 3) or hind leg deflections (Fig. 4). In both figures, 3 original response traces are shown which were obtained each from a single fly, during a sequence o f consecutive stimulus presentations. The two flies differ with respect to the relative contributions o f spontaneous signal fluctuations to the overall response and, thus, illustrate the range of variability which can be found in a typical sample o f flies.

Nevertheless both examples have one interesting feature in common, namely some variability in the overall response pattern which covaries in the different m o t o r outputs. This is particularly obvious in Fig. 3 due to relatively small spontaneous signal fluctuations: Both the yaw torque and abdominal response shown in Fig. 3a resemble the averaged responses at the high oscillation frequency (Fig. 1), with small periodic modulations during large-field motion and large response peaks

(5)

J.M. Zanker et al. : Coordination of motor output in visual orientation of flies 131 1 H z

b

Torque

1 Hz

b

Torque

Leg Deft.

C

Torque 10-7Nm I

1 mm

Time

Fig. 3a-e. Single traces of pairwise recorded yaw torque and abdomen responses. The fly was alternately stimulated with large- field and small-field motion. The pattern was oscillated at a frequency of 1 Hz and with an amplitude of 10 ~ The upper and lower diagram of each response pair represent the yaw torque and abdomen deflection, respectively. The bottom trace of the figure represents the stimulus (conventions as explained in the legend of Fig. 1). During large-field motion the fly shows only weak optomotor turning responses, in agreement with the averaged data shown in Fig. 1. During small-field motion the characteristic response peaks towards the position of the figure may be generated:

(a) a peak is elicited during each stimulation cycle, (b) only the first peak is induced, (e) the second peak is missing. Hence the response peaks during small-field motion at high oscillation frequencies sometimes fail to be generated. They do this synchronously in both the yaw torque and abdomen response. As a consequence, this leads to an underestimation of the corresponding average response amplitudes

d u r i n g small-field m o t i o n . H o w e v e r , in Fig. 3 b ~ it is d e m o n s t r a t e d t h a t d u r i n g small-field m o t i o n single res- p o n s e p e a k s s o m e t i m e s m a y h a p p e n to be n o t g e n e r a t e d by the fly, a l t h o u g h the stimulus c o n d i t i o n s are always

C[ Torque 0.5.10-7Nm / ]

Leg Deft. 0.5 mm /

L

j ~ ~ / ~ V A P A F-,'N

v v l

Time

Fig. 4a--e. Single traces of pairwise recorded yaw torque and leg responses, Stimulus conditions as described in the legend of Fig. 3.

As for yaw torque and abdomen deflection, both behavioral response components are highly correlated. This is despite the fact that the visually induced responses are superimposed here by noise much more than in the example shown in Fig. 3. Nevertheless, it is obvious that in (a) only the first, in (b) only the the third and in (e) no response peak is elicited during small-field motion

identical. W h e r e a s in the example s h o w n in Fig. 3b o n l y the first response p e a k is p r o n o u n c e d , in the example s h o w n in Fig. 3c o n l y the s e c o n d p e a k is missing. T h e i m p o r t a n t p o i n t is t h a t w h e n e v e r a response p e a k is n o t generated, this is the case f o r b o t h the y a w t o r q u e a n d the a b d o m i n a l response. T h e e x t r a o r d i n a r y degree o f c o o r d i n a t i o n o f the different m o t o r o u t p u t systems d u r i n g the given type o f visual stimulation is also f o u n d f o r y a w t o r q u e a n d hind leg deflections, a l t h o u g h in the examples displayed in Fig. 4 it is c a m o u f l a g e d to s o m e extent by the m u c h larger s p o n t a n e o u s signal fluctuations.

(6)

132 J.M. Zanker et al.: Coordination of motor output in visual orientation of flies In conclusion, the various m o t o r o u t p u t systems in-

volved in mediating turning responses o f the fly, i.e. the modulations o f the average wing beat amplitude as re- flected in the yaw torque, as well as the postural changes of the abdomen and hind legs, were found to be highly correlated at least as far as their visually induced components are concerned. This is not only true on average but also for the time course o f single recordings.

Discussion

The virtuosic flight manoeuvres of insects require that the sensory input and m o t o r output systems are carefully matched to each other. To control the various behavioral routines, the available sensory inputs have to be trans- formed appropriately and distributed to the respective sets o f muscles which are involved in executing the movements. We studied the visual control o f three m o t o r output systems in the housefly Musca which appear to be involved in the control o f turning responses about a roughly vertical axis o f the animal. Turning responses are understood to a large extent as the consequence o f flight torque generated by the two wings though the underlying aerodynamic mechanisms are obscure, still (G6tz et al.

1979; G6tz 1983b; Zanker 1990; Zanker and G6tz 1990).

In the present study the wing movements were not monitored directly. Only their immediate result was measured by means o f a torque meter. In addition to modulations of the wing beat cycle, postural changes o f body appendages, such as lateral abdomen and hind leg deflections, have been concluded to be involved in the turning behavior o f flies (G6tz et al. 1979; Zanker 1988) and locusts (Camhi 1970; Arbas 1986; Baader 1990) by shifting the center o f gravity and by acting like an aerodynamic rudder.

In the present study, the coordination of yaw torque, abdominal and hind leg deflections was analyzed for two types o f visually induced flight manoeuvres in tethered flight, compensatory turning responses and turning responses towards objects. It was shown that the yaw torque generated by the wings, the lateral bending o f the abdomen and the hind leg deflections are elicited in a highly correlated manner during both types o f turning reactions. This is not only true for averaged responses but also for the time courses o f individual response traces. Hence, the different m o t o r output systems dis- cussed here do not appear to be specialized for mediating particular functional types o f turns. It should be noted that this is by no means self-evident. F o r instance, the different steering muscles o f the wings which are assumed to be involved in mediating yaw torque (for review, see Heide 1983) are functionally specialized: Whereas some direct flight muscles were concluded to mediate mainly orientation turns towards objects, another type o f steering muscle is also responsible for compensatory opto- m o t o r responses (Egelhaaf 1989). Hence, there is not only diversification between the muscular systems of the wings and the body appendages, but also within the wing beat steering muscles. On the other hand, the time course o f the various m o t o r output systems, such as postural

changes o f abdomen and hind legs and yaw torque, differ only in minor details from each other when the fly per- forms various types o f turning manoeuvres. What seems to be rather simple at the level o f the final m o t o r output, thus may be a complex problem for the underlying neuronal system which has to recruit the appropriate sets of muscles in a well organized and orderly fashion.

As mentioned in the Introduction, compensatory turning responses and orientation turns towards objects appear to be mediated by two parallel control systems, the "large-field" and the "small-field system" which can be attributed to two functional classes o f output cells o f the optic lobes, the HS- and the FD-cells. Before the specific information extracted by the HS- and FD-cells is distributed to the various m o t o r output systems, it is further processed in different ways. (i) A kind o f temporal low-pass filter was proposed (Egelhaaf 1987; Egel- h a a f and Borst 1990, 1991) for the pathway o f the large- field system which attenuates the high frequency components in the HS-cell signals. Although nothing is known so far about the neuronal nature o f the temporal low-pass filter, it should be located at some stage before the m o t o r commands segregate to the wings, the abdomen, and the hind legs, because all 3 have the same dynamical properties. (ii) The fly does not always respond to small-field motion with the m o t o r output systems investigated here. Often single response peaks are omitted. Although this may be sometimes camouflaged by signal components which appear to be independent o f visual stimulation, the response peaks seem to fail simultaneously in all the m o t o r output systems considered here. F r o m this observation two conclusions can be drawn. First, the responses are gated before the information on small-field motion segregates to the m o t o r control centers o f the wings, the a b d o m e n and the hind legs.

Second, the signals carried by the small-field system seem to be gated by some other determinant than visual information. Wind input may play a m o d u l a t o r y role since, at least in some flies, the responses to visual small-field motion occur more reliably during simultaneous wind stimulation o f the tethered flying animal (Egelhaaf, unpublished observations). However, other factors which are not related to sensory input and are hard to characterize experimentally, such as the internal state o f the animal, are likely to play an important role in gating the signals carried by the small-field system. In order to understand the neuronal basis o f this processing, the descending neurones conveying information from the optic lobes to the thoracic m o t o r centers are o f particular interest. In contrast to locusts (for review, see Rowell 1988), however, not much is known a b o u t their physiol- ogy in flies, despite an extensive anatomical description (e.g. Strausfeld 1989; Milde and Strausfeld 1990; Straus- feld and G r o n e n b e r g 1990; G r o n e n b e r g and Strausfeld 1990).

The gating in the pathway for small-field motion must not be confounded with two other gating phenomena which have been described previously. First, the gating of any visual input o f the wing steering muscles by signals of the flight motor. F o r instance, the wing steering muscles o f flies (Heide 1975), the abdominal reactions in

(7)

J.M. Zanker et al.: Coordination of motor output in visual orientation of flies 133 locusts (Camhi 1970), or the activity of flight m o t o n e u -

rons in locusts (Reichert et al. 1985) are only responsive to sensory input as long as the animal is flying. As a consequence o f this gating process, the various m o t o r output systems involved in course control are active only during flight. Second, the visual input has been found to gate wingbeat synchronous proprioceptive afferences which phasically influence the m o t o r control systems o f the wings, both in flies (Heide 1971, 1974, 1975, 1983) and in locusts (Reichert et al. 1985; Reichert and Rowell 1985). The gating o f the small-field system as described in the present study differs from these two other gating p h e n o m e n a in that the gating signal can be neither attributed to the activity o f the flight m o t o r per se n o r directly to sensory afferences.

In contrast to Musca, the lateral abdominal deflections in Drosophila in response to both large-field and small-field motion have the same dynamical properties (Zanker 1988; Zanker and Quenzer 1988). In Drosophila both behavioral response components are most sensitive to low oscillation frequencies, as has been found in Musca for the yaw torque, abdominal and hind leg responses to large-field but not to small-field motion (Egel- h a a f 1987, and present study). There are several possible explanations for this discrepancy. (i) There is no specific small-field system in Drosophila. (ii) The information a b o u t small-field motion is n o t transmitted to the abdominal m o t o r system in Drosophila. (iii) The small- field system could not be activated in Drosophila under the stimulus conditions used in these studies. (iv) The small-field system o f Drosophila has different temporal transfer properties than that o f Musca. If one o f the first three hypotheses were correct, the abdominal responses of Drosophila obtained during small-field motion are mediated by the large-field system alone. This could well be the case, since, at least in Calliphora, the HS-cells respond not only to large-field motion but also, though with smaller amplitudes, to small-field m o t i o n (Hausen 1982a, b). However, the first o f the above hypotheses can p r o b a b l y be discarded immediately. There is ample ev- idence that in Drosophila there is at least another control system, in addition to the one mediating c o m p e n s a t o r y o p t o m o t o r turning responses, which responds best to relatively small moving targets (e.g. G6tz 1983a; Heisen- berg and W o l f 1984; Bausenwein et al. 1986). However, in these studies this control system has been approached from a rather different perspective than in the attempts to characterize the small-field system in the larger flies.

This makes it hard to compare the visual input organization and, in particular, the dynamical response properties of the small-field system in Musca and Calliphora with its hypothetical counterpart in Drosophila.

In conclusion, our behavioral data reveal that, at least in Musca, the various m o t o r output systems which are involved in generating turning responses o f the animal a b o u t its yaw axis are coupled to a high degree. This was shown here for two types o f visually induced turning manoeuvres, namely for c o m p e n s a t o r y o p t o m o t o r responses and for orientation turns towards objects. It has to be analyzed whether the m o t o r output systems are coordinated to the same extent during other manoeuvres,

for example during spontaneous turns or turning responses induced by other sensory modalities.

Acknowledgements. We thank Silke Marcinowski for excellent technical assistance. We also thank Alexander Borst and Werner Reich- ardt for helpful discussions during the course of the experiments and for critically reading the manuscript. The figures are due to the skill of Anke Wildemann.

References

Arbas EA (1986) Control of hindlimb posture by wind-sensitive hairs and antennae during locust flight. J Comp Physiol A 159:849 857

Baader A (1990) The posture of the abdomen during locust flight:

Regulation by steering and ventilatory interneurones. J Exp Biol 151 : 109-131

Bausenwein B, Wolf R, Heisenberg M (1986) Genetic dissection of optomotor behavior in Drosophila melanogaster. Studies on wild-type and the mutant optomotor-blind H31. J Neuro- genetics 3:87-109

Camhi JM (1970) Yaw-correcting postural changes in locusts. J Exp Biol 52:51~531

Egelhaaf M (1985a) On the neuronal basis of figure-ground discrim- ination by relative motion in the visual system of the fly. I.

Behavioural constraints imposed on the neuronal network and the role of the optomotor system. Biol Cybern 52:123-140 Egelhaaf M (1985b) On the neuronal basis of figure-ground discrim-

ination by relative motion in the visual system of the fly. II.

Figure-Detection Cells, a new class of visual interneurones. Biol Cybern 52:195-209

Egelhaaf M (1985c) On the neuronal basis of figure-ground discrim- ination by relative motion in the visual system of the fly. III.

Possible input circuitries and behavioural significance of the FD-CelIs. Biol Cybern 52:267 280

Egelhaaf M (1987) Dynamic properties of two control systems underlying visually guided turning in house-flies. J Comp Physiol A 161:777-783

Egelhaaf M (1989) Visual afferences to flight steering muscles con- trolling optomotor response of the fly. J Comp Physiol A

165:719-730

Egelhaaf M (1990) Spatial interactions in the fly visual system leading to selectivity for small-field motion. Naturwissenschaf- ten 77:182-185

Egelhaaf M, Borst A (1990a) Bewegungswahrnehmung und visuelle Orientierung bei Fliegen. Naturwissenschaften 77:366-377 Egelhaaf M, Borst A (1991) Motion computation and visual

orientation in flies. In: Barnes WJP (ed) Sensory guidance in arthropod behaviour. Manchester University Press, Manches- ter, in press

Egelhaaf M, Hausen K, Reichardt W, Wehrhahn C (1988) Visual course control in flies relies on neuronal computation of object and background motion. Trends Neurosci 11:351-358 Fermi G, Reichardt W (1963) Optomotorische Reaktionen der

Fliege Musca domestica. Abhfi.ngigkeit der Reaktion yon der Wellenl/inge, der Geschwindigkeit, dem Kontrast und der mitt- leren Leuchtdichte bewegter periodischer Muster. Kybernetik 2:15 28

Geiger G, N/issel DR (1982) Visual processing of moving single objects and wide-field patterns in flies: Behavioural analysis after laser-surgical removal of interneurons. Biol Cybern 44:141-149

Geiger G, Poggio T (1977) On head and body movements of flying flies. Biol Cybern 25:177-180

Gronenberg W, Strausfeld NJ (1990) Descending neurons supply- ing the neck and flight motor of Diptera: Physiological and anatomical characteristics. J Comp Neurol 302:973-991

G6tz KG (1964) Optomotorisehe Untersuchung des visuellen Systems einiger Augenmutanten der Fruchtfliege Drosophila.

Kybernetik 2: 77-92

(8)

134 J.M. Zanker et al. : Coordination of motor output in visual orientation of flies GStz KG (1983a) Genetischer Abbau der visuellen Orientierung bei

Drosophila. Verh I)tsch Zool Ges 76:83-99

Grtz KG (1983b) Bewegungssehen und Flugsteuerung bei der Fliege Drosophila. In: Nachtigall W (ed) BIONA report.

Akademie Wissenschaften Literatur Mainz, G Fischer, Mainz Stuttgart New York, pp 21-34

Grtz KG, Hengstenberg B, Biesinger R (1979) Optomotor control of wing beat and body posture in Drosophila. Biol Cybern 35:101-112

Hausen K (1982a) Motion sensitive interneurons in the optomotor system of the fly..~. The Horizontal Cells: Structure and signals.

Biol Cybern 45:143-156

Hausen K (1982b) Motion sensitive interneurons in the optomotor system of (~he fly. II. The Horizontal Cells: Receptive field organization and response characteristics. Biol Cybern 46: 67-79

Hausen K, Wehrhalm ~C (1990) Neural circuits mediating visual flight in flies. II. ~$eparation of two control systems by micro- surgical brain lesions. J Neurosci 10:351-360

Heide G (1971) Die F~unktion der nicht-fibrill/iren Flugmuskeln von

Calliphora. Teil II : Muskul/ire Mechanismen der Flugsteuerung und ihre nervrse K.ontrolle. Zool Jb Physiol 76:99-137 Heide G (1974) The influence of wingbeat synchronous feedback on

the motor output systems in flies. Z Naturforsch 29c:739- 744

Heide G (1975) Properties of a motor output system involved in the optomotor response in flies. Biol Cybern 20:99-112

Heide G (1983) Neural mechanisms of flight control in Diptera. In:

Nachtigall W (ed) BIONA report. Akad Wissenschaften Litera- tur Mainz. G Fischer, Mainz Stuttgart New York, pp 35- 52

Heisenberg M, WolfR (1984) Vision in Drosophila. Springer, Berlin Heidelberg New York, Tokyo

Milde JJ, Strausfeld NJ (1990) Cluster organization and response characterictics of the giant fiber pathway of the blowfly Cal- liphora erythrocephala. J Comp Neurol 294:59-75

Reichardt W, Egelhaaf M, Guo A (1989) Processing of figure and background motion in the visual system of the fly. Biol Cybern 61 : 327-345

Reichert H, Rowell CHF (1985) Integration of nonphaselocked exteroceptive information in the control of rhythmic flight in the locust. J Neurophysiol 53:1201-1218

Reichert H, Rowell CHF, Griss C (1985) Course correction cir- cuitry translates feature detection into behavioural action in locusts. Nature 315 : 142-144

Rowell CHF (1988) Mechanisms of flight steering in locusts. Ex- perientia 44:389-395

Strausfeld NJ (1989) Beneath the compound eye: Neuroanatomicat analysis and physiological correlates in the study of insect vision. In: Stavenga DG, Hardie RC (eds) Facets of vision. Sprin- ger, Berlin Heidelberg New York, pp 317-359

Strausfeld NJ, Gronenberg W (1990) Descending neurons supply- ing the neck and flight motor of Diptera: Organization and neuroanatomical relationships with visual pathways. J Comp Neurol 302: 954-972

Zanker JM (1988) How does lateral abdomen deflection contribute to flight control of Drosophila melanogaster ? J Comp Physiol A 162:581-588

Zanker JM (1990) The wing beat of Drosophila melanogaster III.

Control. Phil Trans R Soc Lond B 327:45-64

Zanker JM, G6tz KG (1990) The wing beat of Drosophila melanogaster II. Dynamics. Phil Trans R Soc Lond B 327:19-44 Zanker JM, Quenzer T (1988) Abdominal deflections elicited by stripe movement. In: Elsner N, Barth FG (eds) Sense organs:

Interfaces between environment and behaviour. G Thieme, Stuttgart New York, pp 132