Effects of luminance, size, and angular velocity on the recognition of nonlocomotive prey models by the praying mantis

Adult females of the mantis Tenodera angustipennis were presented with the "nonlocomotive" prey model, a static rectangle with two lines oscillating regularly at its sides, generated on a computer display. The models were varied in rectangle luminance (black, gray, and light gray), rectangle height (0.72, 3.6, and 18 mm), rectangle width (0.72, 3.6, and 18 mm), and angular velocity of oscillating lines (65°, 260°, and 1040°/s) to examine their effects on prey recognition. Before striking the model, the mantis sometimes showed peering movements that involved swaying its body from side to side. The black model of medium size (both height and width) elicited higher rates of fixation, peering, and strike responses than the large, small, or gray model. The model of medium angular velocity elicited a higher strike rate than that of large or small angular velocity, but angular velocity had little effect on fixation and peering. We conclude that mantises respond to a rectangle in deciding whether to fixate, and to both rectangle and lines in deciding whether to strike after fixation. Received: September 2, 1999 / Accepted: March 21, 2000     

Responses to non-locomotive prey models by the praying mantis,Tenodera angustipennis Saussure

Adult females of the praying mantisTenodera angustipennis were presented with computer-generated images, and the attractiveness of “non-locomotive” prey models was examined. Mantises fixated and struck the “body and leg” model (consisting of an immobile black square on a white background with 2 black lines oscillating randomly at its sides) more frequently than the “leg” model (only oscillating lines) or the “body” model (static square only). This indicates that the model consisting of a static object and moving lines effectively elicits mantis strike behavior, although it is “non-locomotive.”     

Visual stimuli that elicit visual tracking, approaching and striking behavior from an unusual praying mantis, Euchomenella macrops (Insecta: Mantodea)

In comparison to other similarly sized mantis species examined in previous studies, Euchomenella macrops has a significantly smaller head, shorter foreleg tibia, but longer prothorax which have been interpreted as specializations for the capture of smaller, slower prey. We tested this conjecture by assessing the rates at which computer generated stimuli elicit visual tracking, approaching, and striking behaviors by adult females. When presented with black disks moving erratically against a white background, strike rate rose progressively as disks enlarged up to 44 deg (visual angle) if the disks moved rapidly (e.g., 143 deg/s); at slower speeds (113, 127 deg/s), smaller disks (<27 deg) were preferred. When black moved linearly from the visual periphery to visual field center (at 73 or 143 deg/s) and then stopped, E. macrops struck consistently at disks as small as 5 deg after movement ceased. E. macrops also struck at higher rates in response to 23 deg erratically moving (subjective) red (versus subjective blue or green) disks that were luminance matched to a grey background although they tracked all colors at equally high rates. Unlike some other species, E. macrops did not strike at higher rates in response to elongated rectangular stimuli moving parallel (versus perpendicular) to their long axis, although the former elicited higher rates of approaching. An analysis of tracking behavior revealed that virtually all tracking movements were a result of head (versus) prothorax rotation.     

Prey capture in mantids: A non-stereotyped component of lunge

The praying mantis Tenodora aridifolia sinensis strikes at prey with the pincer-like motion of its prothoracic legs. During strike the mantis moves its body forward toward the prey in a lunge which is propelled by its four walking legs. Using a tethered mantis preparation we have studied the lunge produced by the movement of the walking legs. We have found that lunge is correctly oriented toward prey no matter where it moves in three-dimensional space. This demonstrates that the lunge that accompanies the strike is in this species aimed and not invariant in distance and direction as suggested for other mantids.     

Visual deprivation and distance estimation in the praying mantis larva

Abstract. Young larvae of the praying mantis, Tenodera sinensis Saussure, were placed on an off-centre island surrounded by a round arena with six black bars painted on a white inner wall. In this situation, it was shown that the horizontal peering movements of the head often seen in mantids are in fact used to measure distances; motion parallax may be involved in this process. Aimed jumps that followed peering were taken to be the distinct result of an absolute distance measurement. Specific visual deprivation such as painting over of certain parts of the eye with opaque black varnish or degeneration of the fovea with sulforhodamine showed that: absolute evaluation of distance is only possible with two fully intact eyes; the peering mechanism is under visual control; and visual experience has a long-term effect on distance measurement involving peering movements.     

How Far Stationary Contrast Boundaries Can Be Away to Elicit Behavioral Responses in Praying Mantis

The aim of the present study was to investigate the distance at which vertical black and white stripes (contrast boundaries) elicit object-related behavioral responses in 6th instar and adults of the praying mantis Mantis religiosa. The mantids reacted when the contrast boundaries were not further away than 60 cm. However, with increasing distance (>20 cm), the contrast boundaries became progressively less significant for the mantids. Jumps/preparation of jumps could be observed between 10 and 30 cm. The results are supportive for distance measurement of up to 20–30 cm, which corresponds to distance accessible for the insect. It seems that image motion cues induced by peering movements play an important role.     

Visual stimuli that elicit appetitive behaviors in three morphologically distinct species of praying mantis

We assessed the differences in appetitive responses to visual stimuli by three species of praying mantis (Insecta: Mantodea), Tenodera aridifolia sinensis, Mantis religiosa, and Cilnia humeralis. Tethered, adult females watched computer generated stimuli (erratically moving disks or linearly moving rectangles) that varied along predetermined parameters. Three responses were scored: tracking, approaching, and striking. Threshold stimulus size (diameter) for tracking and striking at disks ranged from 3.5 deg (C. humeralis) to 7.8 deg (M. religiosa), and from 3.3 deg (C. humeralis) to 11.7 deg (M. religiosa), respectively. Unlike the other species which struck at disks as large as 44 deg, T. a. sinensis displayed a preference for 14 deg disks. Disks moving at 143 deg/s were preferred by all species. M. religiosa exhibited the most approaching behavior, and with T. a. sinensis distinguished between rectangular stimuli moving parallel versus perpendicular to their long axes. C. humeralis did not make this distinction. Stimulus sizes that elicited the target behaviors were not related to mantis size. However, differences in compound eye morphology may be related to species differences: C. humeralis’ eyes are farthest apart, and it has an apparently narrower binocular visual field which may affect retinal inputs to movement-sensitive visual interneurons.     

A chemical correlate of learning in a praying mantis

Bidimensional and unidimensional maps of amine-containing components extracted from brains of the praying mantis (Stagmatoptera biocellata) were obtained using high-voltage electrophoresis and chromatography, and high-voltage electrophoresis alone.Bidimensional maps from control insects, i.e. animals that did not receive training, showed four distinctive spots and one less intense spot (number 5). On the other hand, bidimensional maps from trained animals, i.e. mantids that were trained not to attack a moving star, showed the same spots 1–4, plus an intense spot (number 5) and an extra componet (number 6).Unidimensional maps from brains of mantids that were trained not to attack a moving star (‘star-group’) showed two extra components in comparison with maps from the control insects. On the other hand, when mantids received training similarto that of the star-group, but using a fly that could not be caught as a stimulus, instead of a mobile star, they did not learn and their maps were similar to those from control mantids.The techniques used in this paper to obtain the maps suggest that they are maps of peptides of low molecular weight. The possible correlation between the appearance of extra spots in the maps and a learning process is discussed.     

Investigating saccade programming in the praying mantis Tenodera aridifolia using distracter interference paradigms

To investigate the saccadic system in the mantis, I applied distracter interference paradigms. These involved presenting the mantis with a fixation target and one or several distracters supposed to affect saccades towards the target. When a single target was presented, a medium-sized target located in its lower visual field elicited higher rates of saccade response. This preference for target size and position was also observed when a target and a distracter were presented simultaneously. That is, the mantis chose and fixated the target rather than a distracter that was much smaller or larger than the target, or was located above the target. Furthermore, the mantis' preference was not affected by increasing the number of distracters. However, the presence of the distracter decreased the occurrence rate of saccade and increased the response time to saccade. I conclude that distracter interference paradigms are an effective way of investigating the visual processing underlying saccade generation in the mantis. Possible mechanisms of saccade generation in the mantis are discussed.     

Response of the Praying Mantis,Sphodromantis Viridis,to Target Change in Size and to Target Visual Occlusion

Mantises (Mantodea, Mantidae) visually detect insect prey and capture it by a ballistic strike of their specialized forelegs. We tested predatory responses of female mantis, Sphodromantis viridis, to computer generated visual stimuli, to determine the effects of (i) target size and velocity (ii) discrete changes in target size and (iii) visual occlusion. Maximal predatory responses were elicited by stimuli that (i) subtended ~20°–23° horizontally and ~16°–19° vertically, at the eye, and moved across the screen at angular velocities of ~46°–119°/s, (ii) increased in size in a stepwise manner, with step duration ≥0.8 s, while stimuli decreasing in size elicited only peering movements, (iii) Stimuli disappearing gradually behind a virtual occlusion elicited one or more head saccades but not actual interception.