Patricia M. Blough

Pigeons see the Ponzo illusion

Three experiments were performed to address factors contributing to the Ponzo illusion. In Experiment 1, pigeons learned to peck at the longer of two bars in varying line contexts. When these lines converged, the birds had difficulty in learning several patterns in which a Ponzo illusion would reduce the perceived difference. In Experiment 2, the subjects chose one response if a stimulus bar was longer than a predetermined length and a second response for shorter bars. The subjects were more likely to choose “long” as the stimulus bar approached the apex of converging lines. These results suggested that pigeons experience the Ponzo illusion. In Experiment 3, the effects of contexts that did and did not form a texture gradient were compared. The magnitude of the illusion did not differ according to context lines. This result failed to support a perspective theory of the Ponzo illusion.

Reaction time assessments of gender differences in visual-spatial performance

This research examined gender differences in reaction time and accuracy on four visual tasks. Twenty-nine female and 29 male first-year college students responded via joysticks to video displayed stimuli. Computer-controlled sessions consisted of trial blocks during which subjects indicated (1) a choice depending on the form of a simple stimulus, (2) which two-dimensional representation of a “folded” box was equivalent to an “unfolded” box, (3) whether test stimuli of varying rotations were the same as or different from an upright standard displayed simul taneously, and (4) whether test forms of varying dissimilarity were the same as or different from a standard. Women were more accurate but slower on the choice task; they had higher reaction times on the mental rotation and the shape-comparison tasks. These latter gender differences interacted significantly with degree of rotation and dissimilarity of the test form, suggesting the presence of gender differences in visual-spatial strategies.

Attentional priming and visual search in pigeons.

Advance information about a targets identity improved visual search efficiency in pigeons. Experiments 1 and 2 compared information supplied by visual cues with information supplied by trial sequences. Reaction times (RTs) were lower when visual cues signaled a single target rather than two. RTs were (Experiment 1) or accuracy improved (Experiment 2) when a sequence of trials presented a single target rather than a mixture of 2. Experiments 3, 4, and 5 considered the selectivity of visual priming by introducing probe trials that reversed the usual cue-target relationship. RT was higher following such miscues than following the usual 1- or 2- target cuing relationships (Experiment 3); the miscuing effect persisted over variations in the targets concealment (Experiments 4 and 5), but did not occur when the target was presented alone (Experiment 4). The findings indicate that priming modifies an attentional mechanism and suggest that this effect accounts for search images.

Wavelength generalization and discrimination in the pigeon

This three-part study describes wavelength generalization gradients around a series of training wavelengths ranging from 480 to 645 om. Luminance was controlled for the pigeon’s spectral sensitivity. The response measure was probability of keypecking during a 2-sec stimulus presentation. Both an extinction procedure, where stimulus wavelengths occurred in 15-nm steps, and a maintained discrimination procedure, where step size was 2 to 4 nm, were used to obtain gradients. During a portion of the maintained discrimination procedure, new luminances were introduced, so that the effect of luminance level on gradient slope could be examined. Comparison of the resulting functions across training wavelengths revealed: (1) consistent differences in gradient slope in different spectral regions and (2) an increase in slope with luminance increase. The findings are related to recent electroretinographic wavelength contrast data and to psychophysical measures of wavelength discriminability.

Form perception and attention in pigeons

This report selectively reviews the authors’ research on stimulus control by visual forms. Most studies employed visual search of computer-generated displays, and the main behavioral measure was search reaction time. Pigeons classify simple line forms much as do humans, and there was some evidence for feature extraction, though a search asymmetry probe failed to identify several putative features. Dimensional analysis revealed a new quantitative relation between similarity and the probability of detecting a target. It also identified candidates for integral and separable stimulus dimensions. Similarity and set-size variables influenced the extent to which attention was engaged. Further studies addressed the manner in which attention is focused, revealing relationships between priming and search images. In these cases, attention to a selected form is probably driven by a top-down assessment of predictability.

Visual search in pigeons: Effects of memory set size and display variables

A visual search procedure compared memory and perceptual effects. Pigeons pecked at a target letter in a three-key forced-choice procedure. Experiment 1 varied memory set and display size. Neither had a significant effect on accuracy, and changing memory set from 1 to 2 items did not affect reaction time. However, reaction times grew longer as display size increased from 3 to 9 items (p < .005) and memory set size increased from 1 to 4 (p < .05). Experiment 2 extended memory set size to 6 and display size to 18. Four experienced subjects showed no significant effects of memory set size on accuracy or reaction time, while reaction time increased (p < .005) with increasing display size. Experiment 3 considered effects of distractor redundancy and target-distractor similarity. Mean reaction times for three experienced birds showed persisting display size effects that depended on target-distractor similarity; search was fastest for redundant distractor sets in which the items were relatively dissimilar to the target. Except for the persistence of a display-size effect, these outcomes are consistent with human data.

Detectability and choice during visual search : joint effects of sequential priming and discriminability

Pigeons searched for symbolic targets among heterogeneous distractor items displayed on a video monitor. Phase 1 varied target identity and overall display size, thus establishing differential discriminabilities of three target symbols. Phase 2 varied the relative probability of these targets within sessions. The findings showed that reaction time was lower not only when targets were more discriminable, but also when they were relatively frequent; these effects did not depend on the discriminability of the less frequent targets. Phase 3 was similar in design but provided occasional choice trials on which two targets appeared. The birds were more likely to respond to the more frequent target on such trials only if it was also the most discriminable. The data are not consistent with certain predictions from Guilford and Dawkins’ (1987) reinterpretation of effects attributed to search images. The data indicate that detection and choice are modified jointly by priming-induced expectancies and stimulus-driven perceptual processes.

Effects of the inclination of context lines on perception of the Ponzo illusion by pigeons

Pigeons discriminated the length of a bar located between two context lines. Responses to one key were reinforced when the bar was longer than a predetermined length, and those to the other key were reinforced when the bar was shorter. The inclination of the context lines was systematically varied from 54.6° (converging upward) to 125.4° (converging downward). Five out of 6 subjects tended to report “long” when the bars were located near the apex of the context lines, regardless of whether the context lines were oriented upward or downward. The magnitude of the illusion varied almost linearly with the ratio of the length of the stimulus bar to the gap between the bar and the context lines. This relationship held equally for upward- and downward-converging context lines.

Photopic spectral sensitivity determined electroretinographically for the pigeon eye

Abstract Two techniques were successful in producing photopic ERG responses in the pigeon. One method used a conventional flicker vane to deliver flashes of monochromatic light to the eye at a rate of 35 Hz. The second used a pattern of alternating light and dim bars whose displacements with respect to each other yielded an electrical response, while the overall luminous flux and stray light remained constant. Both of these techniques have been used with the human eye to isolate the photopic component of the ERG. Spectral sensitivity data were obtained in two pigeons with both types of stimulus. The two methods produced similar spectral sensitivity curves having broad maxima through the wavelength range from 570 to 600 nm. The flicker data showed lower variability, with some evidence that sensitivity seemed to reach its maximum in a range that lay a few nanometers lower than that derived from the alternating bar data. Overall, however, the functions agree well with the behavioral data of Blough and the ganglion cell spike data of Granit. Thus, the present findings suggest that the spectral sensitivity indicated by those more complex measures is determined early enough to hold also for ERG responses from the retinas inner nuclear layer.

Electrical responses of the pigeon eye to changes in wavelength of the stimulating light

Abstract An electroretinographic technique is used to measure responses of the pigeon eye to differences in wavelength. Alternation between any two monochromatic lights of matched luminance yields response potential waves the amplitudes of which are related to the differences in wavelength. The results may be accounted for by a set of three underlying color mechanisms whose output functions when added algebraically yield a good fit to the measured amplitudes. Thus the responses are consistent with a simple, additive trichromatic theory. At the level of the electroretinogram there is an optimum region from 510 to 540 nm for color-differentiating responses. The region of 585–600 nm shows somewhat poorer differentiation, and the region around 570 nm, together with those below 510 nm and above 615 nm, are poorest of all.