Simple Brain Simulation Resources 
Home > Visual System > Frog Visual Targeting

DanceFloor by dynamic artist Jenny James. Copyright 2006 (used with permission)

Acquisitive Visual Targeting Behavior in the Frog and Toad

by David D. Olmsted (Copyright - 2001, 2006. Free to use for personal and educational purposes)
Last Revised November 6, 2006

The Acquisitive Targeting Field

Frogs and toads must decide among two different classes of acquisitive behaviors: approaching and orienting. The approach class of behavior can further be divided into four subclasses: snap, turn - snap, jump-snap, and jump. Most of these are shown in figure 1. A three inch frog will snap at straight ahead objects as far out as 5 inches, and objects to the side as far out as 3 inches.

The aquisitive targeting field is relative to head position. Turning a frog’s head 30 degrees also turns the aquisitive targeting field the same amount (Ingle - 1970). For a given frog the acquisitive behaviors are very consistent with the snap - jump boundaries not varying by more than a half inch (Ingle - 1970).

Figure 1
The Acquisitive Behavior Field of the Frog. (Ingle - 1970)

The various acquisitive actions can be elicited by electrical stimulation of the tectum as shown in figure 2. Numbered locations show the region of the contralateral visual field (the side opposite the side electrically stimulated) where the frog turned during electrical stimulation. Inner columns are progressively more rostal (noseward). Other sites produced snapping and gulping. This suggests that the tectum is the trigger of these actions.

Figure 2
Electrical Stimulation of the Tectum in the Toad Bufo bufoProduces Acquistion Actions (Ewert - 1970).

An HRP study by Masino and Grobstein (1990) reported that the neural outputs from the Tectum project to the Reticular Formation near the Superior Olive:

"Some of these collaterals (axon branches from the Tectum) cross the midline to terminate in the vicinity of the Olive on the opposite side of the brain. Previous reports have suggested that this collateralization reflects a bilateral input to the Olive itself. Our material shows the fibers to extend well beyond the olivary cell group and terminate as well near cells of the Medial Reticular Nucleus. In fact, most of the observed bouton-like swelling were closer to Medial Reticular neurons than to olivary neurons."

Response Characteristics to Prey Stimuli

A test to determine the which visual stimulus produced the best turning response in a toad is shown in figure 3 with each turn indicated by a vertical line. Disks of varying sizes were rotated around the toad Bufo bufo at a rate of 30 degrees per second (5 rotations per minute). The contrast ratio was 0.95 given by: (stimulus luminance - background luminance) / total luminance. The maximum orienting response with a 6 degree spot was an average of 25 turns per minute (one every 2.4 seconds. The actual distance of the spots was not given. The turning responses have a probabilistic component since they are not strictly periodic

The lower the contrast ratio between the stimulus and the background the lower is the response rate. The exact relationship depends on the shape of the stimulus and the motivation of the toad. At a maximum contrast ratio of 0.95 a dark stimulus has a stronger releasing efficiency than a white one (Ewert - 1970).

When a white optimal stimulus (2 degrees by 16 degrees having a contrast ratio of 0.95) is used the maximum orientation rates are elicited at stimulus angular velocities of between 30 and 60 degrees per second. (Ewert - 1970).

When a strobe light is used on a stimulus circling at 30 degrees per second the inferred motion assumption used by the toad can be found since the stimulus object appears to jump around the toad. At a flash rate of 5 per second the toad will still respond but the maximum releasing values are found at flash rates between 10 and 20 per second (Ewert - 1970)

Figure 3
Test Mechanism Used on the Toad Bufo bufo for Determining Orienting Response Rates. (Ewert - 1970)

Response Characteristics to Partly Occluded Stimuli

Often times prey objects will suddenly appear from behind a rock or plant or disappear behind the same. This means that the frog visual system must deal with partial stimulus information. Such a situation was tested by David Ingle using the apparatus shown in figure 4. A frog or toad is placed in a box having slits behind which is passed light or dark cards. All cards were scanned at 4 to 6 degrees per second. Frogs were usually placed 3 inches aways from the slits. Stimuli were presented after the frog was facing at least 45 degrees away from a slit for 10 seconds.

In this occlusion situation frogs (Rana pipiens) had different best responses when orienting and snapping were compared. They oriented best towards 4 degree wide slits but snapped best at 15 degree wide slits.In the frog's environment distant bugs will appear smaller while close bugs that are within range of its tongue will appear larger. The response rates are shown below:

SizeOrientationSnap
91%43%
46%---
15°25%64%
 

Figure 4
Testing for Orientation Towards Occluded Targets. (Ingle - 1968)

Yet angular size is only one factor in determining the probabilities of orientation versus snapping behaviors. Frogs are able to adjust for distance in order to maximally respond to their favorite sized prey of 3/16 inch thick worms as shown in figure 5. The first and third columns have the same angular size (1.2 degrees) as do columns two and four (3.6 degrees) despite being at distences of 3 and 9 inches respectively. Best response rates are from the 3/16 inch thick “worms” irrespective of distance (at 93%). The 1/16 inch thick “worm” produced a response of 41% while the 9/16 inch thick “worm” produced a response of 34%. Not accounted for in this experiment is the angular velocity.

Figure 5
Orientation Towards Stimuli Having the Same Angular Size but Differing Distences.

Frogs and toads also tend to snap more at disappearing “worms than at “worms” which are just appearing as shown in figure 6. Frogs even tend to snap at the last minute to a shrinking “worm” just before it disappears. This trend also holds for white “worms" which show snap rates of 88% for the disappearing ones and 38% for the appearing ones.

Figure 6
Frogs and Toads Snap More at Disappearing “Worms”. (Ingle - 1968)

Despite orienting towards the head of white worms toads will tend to snap at the middle of the worm.with the middle determined relative to the endpoints of the worm as shown in figure 7. Yet the absolute light levels seem to determine where the toad actuallys strikes the worm like object . See Responses of the Toad in Low Light Levels

Figure 7
In Normal Light Levels Toads Tend to Strike Towards the Middle of a “Worm” as Determined by its Ends. (Ingle - 1968)

Decision Preferences - Distance Effects

When given a choice between two stimuli frogs will tend to orient towards the nearest stimuli. In one experiment (Ingle - 1973) two yellow stimuli 1/2 x 1/8 inches were wiggled at distances 2-1/2 inches or 3 inches away from the frog. The frogs were motivated to orient by first establishing the appropriate environmental context by feeding them some mealworms. This produced the following orientation rates for the nearer stimulus when both targets were placed at equal angles on opposite sides.

  • Opposite side nearer: 45° from midline - 92%
  • Opposite side nearer: 90° from midline - 84%
  • Same side nearer: 45° from midline - 85%
  • Same side nearer: 135° from midline- 85%

Frogs and toads compensate for distance up to 15 cm as shown in figure 7. In this experiment the frog Rana pipiens had to choose between the variable sized test prey and a constant 3 degree sized prey. The optimum prey size for orientation within this distance has a width of 0.8 cm. Notice that doubling the distance from 7.5 cm to 15 cm quartered the optimum prey size from 6 degrees to nearly 1.5 degrees yet this prey object was physically the same size. The 22.5 cm distance had the same optimum angular size in degrees as the 15 cm distance indicating that size constancy broke down.  Yet significantly the decision on whether to approach a hole in a barrier is only based upon the visual angle (Ingle and Cook - 1977).

Figure 8
The frog Rana pipiens Compensates for Prey Size up to a Distance of 15 cm.
Triangle Line - 22.5 cm distant prey, Square Line - 15 cm distant prey, Circle Line - 7.5 cm distant prey. Plain Line - Ewert’s data (1970) for toad Bufo bufo. (Ingle and Cook - 1977)

Decision Preferences - Horizontal Location Effects

Frogs tend to orient preferentially towards stimuli located directly in front. In one experiment (Ingle - 1973) two yellow stimuli 1/2 x 1/8 inches were wiggled 3 inches away from the frog. The frogs were motivated to orient by first establishing the appropriate environmental context by feeding them some mealworms. This produced the following orientation rates for the more frontward (rostral) stimulus.

  • Percent midline target selected compared to 45° target - 84%
  • Percent 30° target selected compared to 90° target - 48%

Cutting the optic nerve from one eye abolished this frontal preference and even reversed it. The midline vs. 45° decision produced the following results for the given number of days after the operation:

  • 4 days - 16%
  • 4 weeks - 29%
  • 4 months - 34%

In the same experimental setup to that described above David Ingle (1973) measured the following average orientation latencies:

Stimuli 30° from midline:

  • One stimulus - 2.5 seconds
  • Two stimuli - 6.4 seconds

Stimuli 90° from midline:

  • One stimulus - 2.1 seconds
  • Two Stimuli - 2.1 seconds

Stimuli on same side:

  • One stimulus (front or back) - 2.3 seconds
  • Two stimuli - 2.4 seconds

From these experiments indecision delay seems to only be produced when the stimuli are close to the midline, that is near optimal. This might be somewhat analogous to the indecision delay found even in humans who tend to “freeze” when confronted with some novel and emotionally intense situation.

Compensation Effects

Frogs often tilt their head upwards as much as 90° during orientation to get a better view. Yet they still make accurate turning actions despite objects having differing locations in their visual field (on the retina). Since turning is a bodily referenced action some sort of remapping from retinal coordinates to body coordinates seems to be taking place. This may be one of the functions of the nucleus isthmus which has a topographic representation of all the space surrounding the frog.

Correct orientation seems to require input from both eyes via the tectums on each side of the brain. If the frog is forced to use only one eye it will overshoot the target as shown is figure 9. It appears as if the tectum's average the visual locations as seen by each eye to arrive at the correct stopping location. Frogs turning with only one tectum overshoot their target by about 50% of the required angle yet they show no error for a direct snap (Ingle - 1970).

Figure 9
Toads Overshoot the Target When Forced to Use One Eye. (Ingle - 1970)

Habituation of Responses

Repeated presentation of a stimulus to the same part of the visual field will cause the frog to cease snapping but it will still sometimes orient towards the object (Ingle - 1970). Yet these same frogs will still snap at objects located only 20 degrees away.

Habituated areas, like the acquisitive targeting field shown in figure 1 are relative to the head. A frog will snap at a habituated stimulus placed in the same position relative to the ground if its head is moved (Ingle - 1970).

An example of the rate of visual habituation is shown at the top of figure 10 while the recovery from that habituation is shown at the bottom.  The stimulus is the optimum 2 degree by 16 degree “worm”. The top illustration shows how the number of orientations per each minute declines over time if presented to the same part of the visual field. Notice the log scale on the left. The recovery pattern shown at the bottom is similar to that a damped oscillatory system. The recovery ratio (E) is the (number of responses of a second habituation stimulus series separated from the first by the recovery time - t) / (the number of responses of the first habituation stimulus series).

Figure 10
Orientation Habituation in the Toad Bufo bufo. (Ewert - 1970)

Approach Behavior is Divided into Segments

If prey is farther away than the snap zone the frog or toad will have to approach it. The approach is usually divided into several ballistic segments. This means that once the direction and length of the walk has been determined it is not adjusted even if the prey changes velocity, stops, or disappears (figures 11).

Figure 11
The Toad Bufo marinus Does Not Compensate for Changes in Prey Motion.
Toad position (arrows) and prey postion shown every 0.2 seconds. (Lock and Collett - 1979)

Figure 12 shows the distance of the initial walking segment vs. target distence when the target is moving (a) and when the target vanishes or stops (b ). Bottom graph (c) shows that walking distance is unaffected by the length of time between the initiation of walking and target stop (or vanish). The angle of the initial turn and the length of the walking segment are determined when the decision to approach the target is made. The length of the walk segment is proportional to the distance of the target showing a fine visual depth perception on the part of the toad.
Figure 12
The Length of the Walking Segment in the Toad Bufo marinus Varies with Target Distance. (Lock and Collett - 1979)

Despite the ballistic nature of the walk segment the toad still closes its eyes during the fastest part of the walking cycle as shown in figure 13. Perhaps this is to inhibit visual balance reflexes such as optokinetic nystagmus.

Figure 13
The Toad Bufo marinus Shuts Its Eyes At the Periods of Greatest Motion During the Walking Cycle (Lock and Collett - 1979)

References

Ewert, J. -P. (1970) Neural Mechanisms of Prey-catching and Avoidance Behavior in the Toad (Bufo bufo). Brain, Behavior and Evolution 3:36-56

Gaze, R.M. and Keating, M.J. (1967). Visual Responses from Ipsilateral Tectal Units in the Frog. Journal of Physiology (London) 192:52-53

Ingle, D. (1968). Visual Releasers of Prey-Catching Behavior in Toads. Brain, Behavior, and Evolution 1:500-518.

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

Ingle, D. (1973). Selective Choice Between Double Prey Objects by Frogs. Brain, Behavior and Evolution 7:127-144

Ingle, D & Cook, J. (1977). The Effect of Viewing Distance Upon Size Preference of Frogs for Prey. Vision Research 17:1009-1013

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

Masino, T., and Grobstein, P. (1990) Tectal Connectivity in the Frog Rana pipiens: Tectotegmental Projections and a General Analysis of Topographic Organization. J. Comp. Neurol. 291:103-127



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)