Frog Barrier Avoidance

Frogs and toads will go over (top view in figure 1) or around barriers (bottom view in figure 1) in order to acquire food or some other desirable object. As David Ingle (1970) describes it:

“A toad or frog confronted with a grid barrier will usually not snap directly at the worm moving within the usual snap zone. If the barrier is placed laterally the frog will merely orient toward the object. On the other hand, if the barrier is near the midline, the frog will make a sidestep or hop beyond the nearest side of the obstacle, placing himself in position for a direct strike. If one end of the worm protrudes only slightly, a hungry frog will strike directly at the exposed tip, and will usually bump the side of the barrier in his enthusiasm.”

Figure 1
Barrier Avoidance. (Ingle - 1970)

A clue on how the frog brain accomplishes barrier avoidance is suggested by the effect a cut has on orienting behavior (figure 2). Following a knife cut near the caudo-lateral margin of one tectum which cuts the output fibers an object moving in the affected visual field still produces a turn but one that overshoots the target. As shown in the upper left illustration the location signals must be spread out in the tectum for this to occur.

David Ingle (1970) described it like this:

“In order to fit this observation into our scheme of rigid topographical relationships between tectal input and output, we suggest an intra-tectal displacement of worm elicited excitation...The rising excitation, blocked in this case, might ‘spill over’ via lateral intra-tectal connections to activate an adjacent region whose output would turn the frog toward that region of space adjacent to the window.”

Presumably the retinal class 1 edge detectors are the source responsible for the detection of the barriers since a line or edge stopping in their visual field will continue to activate the neuron.

Figure 2
Tectal Cuts Mimic Barriers. (Ingle - 1970)

And yet the tectum does not seem to be required for barrier avoidance as shown in figure 3. In this experiment, the frog Rana pipiens is prodded to jump by touching its rear. The right eyelid sown shut and the right tectum has been removed. The remaining left tectum only views the right and forward visual field. The barrier is extended to the left visual field, first to 45 and then to 90 where the radial lines represent barrier avoidances Previous experiments have shown that the frog cannot see through its eyelid. Such a frog will avoid barriers even when they are not seen by any tectum (figure 3). The conclusion is that the tectum is only needed to identify spatially defined goal targets (moving prey) and to generate a spatially broad neural signal subject to inhibitory constraints.

Figure 3
Frogs Can Avoid Barriers Without the Use of One Tectum. (Ingle - 1973)

Another piece of evidence that the tectum is not required for barrier avoidance is shown in figure 4. Here a frog with both tectums destroyed will escape a confined area either on its own or when prodded by a rear end touch. It will even do so when depth perception is the only variable. The color of the window has now effect for it will escape towards a white windows contrasted to black box sides or towards a black window contrasted to white box sides. In the experiment shown in figure 4 the frog escapes using only depth cues towards white bars contrasted with white box side

Figure 4
Frog (*Rana pipiens* Can Escape Even After Removal of Both Tectums. (Ingle - 1977)

More evidence indicating that the tectum is not necessary to visually identify barriers is the observation that after six to eight months after one tectum is destroyed the retinal fibers which normally connect to the destroyed tectum will regenerate and connect to the remaining tectum on the opposite side. In this situation the frog will turn and snap at a location on the side opposite to the presented prey. The response is a mirror image to the prey locations. Yet this same frog will avoid barriers normally.

David Ingle (1973) noticed that frogs with the caudal thalamus - pretectal region removed will, like the toad, lack inhibition towards visual objects and they will usually be unable to sidestep stationary barriers.

The torus semicircularis is the region responsible for generating tactile actions and the greater its damage the poorer the aim. This gradual degradation process suggests that it also generates a broad spatial signal just like is found in the tectum.

Characteristics of Barrier Avoidance

Barrier avoidance in frogs and toads is not perfect as shown in the top two illustrations in figure 5. Some times they will head straight towards a barrier but most of the time they will make a detour with varying degrees of probability depending on the situation. Thw hammer shows toad initial position and orientation. Dots are the barrier. Arrows give the percentage of approaches in that direction. Rectangles contain food (mealworms). Fences in Collett’s experiments were made from dowel or wire, 30 cm high and 0.5 to 0.25 cm in diameter. The experimental arena was painted black.

Figure 5
A Second Barrier Affects Toad *Bufo viridis* Detour Behavior. (Collett - 1982)

Toads tend to prefer going through a gap even if another barrier is in sight behind that gap as shown in the lower illustrations of figure 5. As shown if figure 6 the optimum conditions for this are a fence separation greater than 10 cm and a gap width greater than 7 cm (size of the toad not given). The vertical dimension in left graph "a" gives the percentage of toad approaches to the gap. Barriers are 20 cm long so that when gap is 20 cm the front fence is eliminated. Fence separation is 10 cm, prey lies 12 cm behind rear fence, toad is located 10 cm ahead of gap. The vertical dimension in right graph "b" gives the gap approach percentage vs. fence separation with 6 cm wide gap. Connecting the fences to make a cage has little change on the responses.

Figure 6
Effect of Gap Width and Barrier Separation on Toad *Bufo viridis*. (Collett - 1982)

The toad has very fine depth discrimination abilities for it can discern the distance of prey relative to barriers as shown in figure 7. Bottom test (c) has prey 12 cm behind fence. The farther the prey the more likely the toad is to go around the barriers.

Figure 7
Effect of Vertical Prey Location Relative to Second Fence with Toad *Bufo viridis* (Collett - 1982).

Toads tend to go around the barrier end which is nearest to the prey as shown in figure 8. Left (a) experiment has toad facing middle of fence and prey offset while right (b) experiment has toad offset and prey in the middle

Figure 8
Effect of Prey Horizontal Location on Detour Preference with the toad *Bufo viridis* (Collett - 1982).

Toads tend to prefer wider gaps as shown in figure 9. In this experiment the prey located 12 cm behind lower fence in line with the toad (illustration "c"). Fences are 60 cm long. The charts in illustration "b" show the percent gap approach vs. gap width ("illustration a") in cm and degrees. Dotted Line: d = 10 cm and c = 8.5 cm. Left Solid Line: d = 15 cm, and c = 4.5 cm, Right Solid Line d = 7.5 cm, c = 4.5 cm.

Figure 9
Effect of Offset Gaps in Double Fences with the Toad *Bufo viridis*. (Collett - 1982).

Toads tend to prefer in-line gaps so as to pass easily through double barriers as shown in figures 10 and 11. In figure 10 the prey is in-line with the toad 12 cm behind rear fence for illustrations "a" and "b". In Illustration "c" they are 15 cm apart with prey just ahead of the second fence. In figure 11 the prey is in line with toad 12 cm behind rear fence.

Figure 10
Effect of Double Symmetric Forward Gaps With Single Rear Gap with Toad *Bufo viridis*. (Collett - 1982).

Figure 11
Effect of Offset Double Gaps with Toad *Bufo viridis*. (Collett - 1982).

Toads tend to avoid shallow pits (figure 12) yet they will walk into one if that is the shortest route. Toads will jump over deeper pits adjusting their jump length to match the length of the pit (Lock and Collett - 1979). This again shows their fine ability at visual depth perception.

Figure 12
Effect of a Chiasm (Pit) on Toad *Bufo viridis*. (Lock and Collett - 1979)

The further away a barrier is the less effective it is in inhibiting orientation as shown in figure 13. In this experiment the prey in line with toad 10 cm behind rear fence.

Figure 13
Effect of Fence Depth on Toad *Bufo viridis*. (Collett - 1982)

Finally, barrier avoidance behaviors proceed in increments. The frog will move to the edge of a barrier and then pause to re-evaluate the situation. Yet even in the visual absence of prey which it expected to see most toads will continue on for another movement segment as shown in figure 14. In this experiment the lines represent opaque walls so once the toad moves away from a in-line position the prey disappears from view. The course on one detour is shown on the right with the lines showing the orientation of the toad. The lines on the left show the orientation of toads at their initial pause with prey at 10 cm (dashed lines) and with prey at 22 cm (solid lines). Notice they tend to point to the expected location of the prey.

Figure 14
Effect of “Disappearing” Prey on Toad *Bufo viridis*. (Collett - 1982)

References

Collett, T.S. (1982) Do Toads Plan Routes? A Study of the Detour Behavior of Bufo viridis. Journal of Comparative Physiology 146:261-271

Ingle, D. (1970). Visuomotor Functions of the Frog Optic Tectum. Brain, Behavior, and Evolution 3:57-71

Ingle, D. (1973). Two Visual Systems in the Frog. Science 181:1053-1055

Ingle, D. (1977). Detection of Stationary Objects by Frogs (Rana pipiens) After Ablation of Optic Tectum. Journal of Comparative and Physiological Psychology 91:1359-1364

Lock, A. & Collett, T. (1979). A Toads Devious Approach to Its Prey: A Study of Some Complex Uses of Depth Vision. Journal of Comparative Physiology 131:179-189

Web site by David D. Olmsted. He can be contacted at brainsim1-contact at yahoo dot com (this is an anti-spam tactic. Type the address as normal). Original site established August 21, 1998 by David D. Olmsted. New home page published August 25, 2006

Information compiled by David D. Olmsted © 1998 to 2006 (Free to use for personal and educational use)