Cone contrast computations: physical versus perceived background and colour constancy

The light reflected from an object depends both on the object's surface and on the illuminant. Visual systems attempt to resolve this intrinsic ambiguity by comparing the light reflected from the object with respect to the background by computing the difference between the object-background light sampled by three cones. The cone-contrasts for the sample-background stimulus under the test illumination should correspond to the cone-contrasts for samples matched in appearance under the standard background (C). The validity of this cone-contrast rule (Whittle, 2003) and its possible link with stability of perceived colour was studied here using six test illuminants. A successive asymmetric colour-matching task with 40 simulated Munsell samples (value 7, chroma 4) on a neutral background (N7) was used. The subject adjusted the sample under standard illuminant C to match the colour appearance of the test sample under one of the test illuminants. Brunswik Ratio (BR) was used as an index of stability of colour appearance. When computed with respect to the reference illuminant C, the cone contrast rule was violated (particularly for S-cone-contrast). However, if a new reference point based on the perceived colour of the neutral background under the test illumination was used, the cone contrast rule applied. That is, when cone contrasts of the matching samples are computed with respect to this perceived background, they correspond to cone contrasts of the test stimuli. This represents a form of discounting the illuminant for the purpose of determining an object's cone-contrast against the background, which does not vary with background illumination. These cone contrasts, however, do not provide any information about the colour appearance of objects under particular viewing conditions, unless calibrated against a standard by allowing subjects to learn particular colours.     

Sensory, computational and cognitive components of human colour constancy

When the illumination on a scene changes, so do the visual signals elicited by that scene. In spite of these changes, the objects within a scene tend to remain constant in their apparent colour. We start this review by discussing the psychophysical procedures that have been used to quantify colour constancy. The transformation imposed on the visual signals by a change in illumination dictates what the visual system must 'undo' to achieve constancy. The problem is mathematically underdetermined, and can be solved only by exploiting regularities of the visual world. The last decade has seen a substantial increase in our knowledge of such regularities as technical advances have made it possible to make empirical measurements of large numbers of environmental scenes and illuminants. This review provides a taxonomy of models of human colour constancy based first on the assumptions they make about how the inverse transformation might be simplified, and second, on how the parameters of the inverse transformation might be set by elements of a complex scene. Candidate algorithms for human colour constancy are represented graphically and pictorially, and the availability and utility of an accurate estimate of the illuminant is discussed. Throughout this review, we consider both the information that is, in principle, available and empirical assessments of what information the visual system actually uses. In the final section we discuss where in our visual systems these computations might be implemented.     

Dorsoventral asymmetry of colour discrimination in bees

Colour discrimination performance of honeybees was examined by training bees to a two-coloured disc presented on a vertical plane. Access to the food reward was through the centre of the disc. In one experiment, the upper half of the disc was yellow, and the lower half blue. In another experiment, this was vice versa. In either case, the learned disc was tested against each of a series of ten discs whose colour differed in either the upper or the lower half. A comparison between the results obtained in the two experiments reveals that colour discrimination is significantly better in the lower half of the frontal eye region than it is in the upper half. The results, similar to earlier results obtained in pattern discrimination tasks, cannot be explained by peripheral eye-region-specific specializations. It is proposed that the functional significance of the lower frontal visual field is based on more central neural mechanisms that might constitute an adaptation to the forager's natural needs. Accepted: 2 November 1998     

Comparative psychophysics of bumblebee and honeybee colour discrimination and object detection

Bumblebee (Bombus terrestris) discrimination of targets with broadband reflectance spectra was tested using simultaneous viewing conditions, enabling an accurate determination of the perceptual limit of colour discrimination excluding confounds from memory coding (experiment 1). The level of colour discrimination in bumblebees, and honeybees (Apis mellifera) (based upon previous observations), exceeds predictions of models considering receptor noise in the honeybee. Bumblebee and honeybee photoreceptors are similar in spectral shape and spacing, but bumblebees exhibit significantly poorer colour discrimination in behavioural tests, suggesting possible differences in spatial or temporal signal processing. Detection of stimuli in a Y-maze was evaluated for bumblebees (experiment 2) and honeybees (experiment 3). Honeybees detected stimuli containing both green-receptor-contrast and colour contrast at a visual angle of approximately 5 degrees , whilst stimuli that contained only colour contrast were only detected at a visual angle of 15 degrees . Bumblebees were able to detect these stimuli at a visual angle of 2.3 degrees and 2.7 degrees , respectively. A comparison of the experiments suggests a tradeoff between colour discrimination and colour detection in these two species, limited by the need to pool colour signals to overcome receptor noise. We discuss the colour processing differences and possible adaptations to specific ecological habitats.     

Colour constancy under simultaneous changes in surface position and illuminant

Two kinds of constancy underlie the everyday perception of surface colour: constancy under changes in illuminant and constancy under changes in surface position. Classically, these two constancies seem to place conflicting demands on the visual system: to both take into account the region surrounding a surface and also discount it. It is shown here, however, that the ability of observers to make surface-colour matches across simultaneous changes in test-surface position and illuminant in computer-generated 'Mondrian' patterns is almost as good as across changes in illuminant alone. Performance was no poorer when the surfaces surrounding the test surface were permuted, or when information from a potential comparison surface, the one with the highest luminance, was suppressed. Computer simulations of cone-photoreceptor activity showed that a reliable cue for making surface-colour matches in all experimental conditions was provided by the ratios of cone excitations between the test surfaces and a spatial average over the whole pattern.     

Local estimate of surface Laplacian derivation on a realistically shaped scalp surface and its performance on noisy data

A new implementation of the surface Laplacian derivation (SLD) method is desribed which reconstructs a realistically shaped, local scalp surface geometry using measured electrode positions, generates a local spectral-interpolated potential distribution function, and estimates the surface Laplacian values through a local planar parametric space using a stable numerical method combining Taylor expansions with the least-squares technique. The implementation is modified for efficient repeated SLD operations on a time series. Examples are shown of applications to evoked potential data. The resolving power of the SLD is examined as a function of the spatial signal-to-noise (SNR) ratio. The analysis suggests that the Laplacian is effective when the spatial SNR is greater than 3. It is shown that spatial low-pass filtering with a Gaussian filter can be used to reduce the effect of noise and recover useful signal if the noise is spatially incoherent.     

An investigation of colour discrimination with horses (Equus caballus)

The ability of four horses (Equus caballus) to discriminate coloured (three shades of blue, green, red, and yellow) from grey (neutral density) stimuli, produced by back projected lighting filters, was investigated in a two response forced-choice procedure. Pushes of the lever in front of a coloured screen were occasionally reinforced, pushes of the lever in front of a grey screen were never reinforced. Each colour shade was randomly paired with a grey that was brighter, one that was dimmer, and one that approximately matched the colour in terms of brightness. Each horse experienced the colours in a different order, a new colour was started after 85% correct responses over five consecutive sessions or if accuracy showed no trend over sessions. All horses reached the 85% correct with blue versus grey, three horses did so with both yellow and green versus grey. All were above chance with red versus grey but none reached criterion. Further analysis showed the wavelengths of the green stimuli used overlapped with the yellow. The results are consistent with histological and behavioural studies that suggest that horses are dichromatic. They differ from some earlier data in that they indicate horses can discriminate yellow and blue, but that they may have deficiencies in discriminating red and green.     

Simultaneous and successive colour discrimination in the honeybee (Apis mellifera)

The colour discrimination of individual free-flying honeybees (Apis mellifera) was tested with simultaneous and successive viewing conditions for a variety of broadband reflectance stimuli. For simultaneous viewing bees used form vision to discriminate patterned target stimuli from homogeneous coloured distractor stimuli, and for successive discrimination bees were required to discriminate between homogeneously coloured stimuli. Bees were significantly better at a simultaneous discrimination task, and we suggest this is explained by the inefficiency with which the bees brain can code and retrieve colour information from memory when viewing stimuli successively. Using simultaneous viewing conditions bees discriminated between the test stimuli at a level equivalent to 1 just-noticeable-difference for human colour vision. Discrimination of colours by bees with simultaneous viewing conditions exceeded previous estimates of what is possible considering models of photoreceptor noise measured in bees, which suggests spatial and/or temporal summation of colour signals for fine discrimination tasks. The results show that when behavioural experiments are used to collect data about the mechanisms facilitating colour discrimination in animals, it is important to consider the effects of the stimulus viewing conditions on results.     

Behavioral discrimination and time-series phenotyping of birdsong performance

Variation in the acoustic structure of vocal signals is important to communicate social information. However, relatively little is known about the features that receivers extract to decipher relevant social information. Here, we took an expansive, bottom-up approach to delineate the feature space that could be important for processing social information in zebra finch song. Using operant techniques, we discovered that female zebra finches can consistently discriminate brief song phrases (“motifs”) from different social contexts. We then applied machine learning algorithms to classify motifs based on thousands of time-series features and to uncover acoustic features for motif discrimination. In addition to highlighting classic acoustic features, the resulting algorithm revealed novel features for song discrimination, for example, measures of time irreversibility (i.e., the degree to which the statistical properties of the actual and time-reversed signal differ). Moreover, the algorithm accurately predicted female performance on individual motif exemplars. These data underscore and expand the promise of broad time-series phenotyping to acoustic analyses and social decision-making.     

Absolute and relational control of a sequential auditory discrimination by pigeons (Columba livia)

Recent evidence indicates that pigeons can readily learn visual discriminations based on both absolute and relational stimulus factors. To examine how these two types of control function in their non-dominant auditory modality, we tested four pigeons in a go/no-go sequential auditory discrimination in which both absolute and relational cues were redundantly available. In this task, sequences of different sounds created from one set of pitches were reinforced, while different sequences created from another set of pitches and any same sequences made from either set of pitches were not. Across three experiments, we independently varied the relative discriminability of the absolute and relational components. The pigeons were consistently and primarily controlled by the absolute fundamental pitch of our notes in all of the experiments, although this was influenced by the range and arrangement of the pitches used in each set. A majority of the pigeons also demonstrated relational control when this component was made more salient. The more robust control exhibited by absolute factors is consistent with the comparative hypothesis that birds in general may have a well-developed aptitude for processing absolute pitch in many auditory settings. The relational control is consistent with our recent evidence of same/different auditory learning by pigeons.     

Familiarization and reliability of multiple sprint running performance indices

The aims of this study were to evaluate the time-course of the familiarization process associated with a test of multiple sprint running performance and to determine the reliability of various performance indices once familiarization had been established. Eleven physically active men (mean age: 21 +/- 2 years) completed 4 multiple sprint running trials (12 x 30 m; repeated at 35-s intervals) with 7 days between trials. All testing was conducted indoors, and times were recorded by twin-beam photocells. Results revealed no apparent learning effects as evidenced by no significant (p > 0.05) between-trial differences in measures of fastest or mean 30-m sprint time. Within-subject test-retest reliability determined over 4 trials by coefficient of variation (CV) and intraclass correlation coefficient (ICC) showed excellent reliability for measures of fastest and mean sprint times (CV range: 1.34-2.24%; ICC range: 0.79-0.94). Pre- and posttrial blood lactate concentrations showed good reliability when judged in context with typical values (CV range: 12.08-18.21%; ICC range: 0.72-0.78). In contrast, and in line with previous research, fatigue data showed much greater variability (CV: 26.43%; ICC: 0.66). The results of this study suggest that high degrees of test-retest reliability can be obtained in many multiple sprint running indices without the need for prior familiarization.     

Variable discrimination and asymmetric preferences in laboratory tests of reproductive isolation between cichlid colour morphs

Mating behaviour affects reproductive isolation and phenotypic differentiation. In Lake Tanganyika, the cichlid fish Tropheus moorii diversified into numerous, currently allopatric colour variants. Allopatric isolation is periodically interrupted by dispersal and secondary contact during lake level fluctuations, making long‐term differentiation partly dependent on assortative mating. Laboratory experiments with two moderately distinct morphs revealed assortative female preferences in one (Nakaku), but random mate choice in the other morph (Mbita). No discrimination was apparent between two subtly differentiated morphs (Chimba and Moliro). Tested against each other in a previous study, the highly distinct Moliro and Nakaku exhibited strong assortative preferences. The correlation between colour pattern similarity and mate discrimination suggests that allopatry and philopatric behaviour are less crucial for the maintenance of differentiation between highly distinct morphs than for more similar morphs. Interestingly, the asymmetric isolation in one pair of morphs is congruent with a pattern of unidirectional mitochondrial introgression between populations.     

Using surface envelopes for discrimination of molecular models

Shape information about macromolecules is increasingly available but is difficult to use in modeling efforts. We demonstrate that shape information alone can often distinguish structural models of biological macromolecules. By using a data structure called a surface envelope (SE) to represent the shape of the molecule, we propose a method that generates a fitness score for the shape of a particular molecular model. This score correlates well with root mean squared deviation (RMSD) of the model to the known test structures and can be used to filter models in decoy sets. The scoring method requires both alignment of the model to the SE in three-dimensional space and assessment of the degree to which atoms in the model fill the SE. Alignment combines a hybrid algorithm using principal components and a previously published iterated closest point algorithm. We test our method against models generated from random atom perturbation from crystal structures, published decoy sets used in structure prediction, and models created from the trajectories of atoms in molecular modeling runs. We also test our alignment algorithm against experimental electron microscopic data from rice dwarf virus. The alignment performance is reliable, and we show a high correlation between model RMSD and score function. This correlation is stronger for molecular models with greater oblong character (as measured by the ratio of largest to smallest principal component).     

Spatial, colour and contrast response characteristics of mechanisms which mediate discrimination of pattern orientation and magnification

We have measured response times for the detection of a single target presented against a set of reference elements which are characterised by combinations of four different stimulus parameters; colour, contrast polarity, magnification and orientation. The aim of the experiments was to determine the response characteristics of visual mechanisms which mediate target detection through the discrimination of orientation and magnification. In the first experiments, we determined sensitivity to differences in colour and contrast polarity, and show that the mechanisms responsible for the discrimination of orientation and of magnification are both selective in their responses to colour and to contrast polarity. There are, nonetheless, residual interactions between patterns of different contrast polarities and between those of different colour, and in the latter case, weak interactions persist under equiluminance conditions. In a second set of experiments, we examined the interactions between orientation and magnification. We conclude that the responses of visual mechanisms which mediate target detection through discrimination of orientation are markedly dependent on stimulus magnification whereas those which mediate detection through discrimination of magnification are, in contrast, relatively insensitive to stimulus orientation.