Limits...
Spatially pooled contrast responses predict neural and perceptual similarity of naturalistic image categories.

Groen II, Ghebreab S, Lamme VA, Scholte HS - PLoS Comput. Biol. (2012)

Bottom Line: We compared our results with two other, well- known contrast statistics: Fourier power spectra and higher-order properties of contrast distributions (skewness and kurtosis).Interestingly, whereas these statistics allow for accurate image categorization, they do not predict ERP response patterns or behavioral categorization confusions.These converging computational, neural and behavioral results suggest that statistics of pooled contrast responses contain information that corresponds with perceived visual similarity in a rapid, low-level categorization task.

View Article: PubMed Central - PubMed

Affiliation: Cognitive Neuroscience Group, Department of Psychology, University of Amsterdam, Amsterdam, The Netherlands.

ABSTRACT
The visual world is complex and continuously changing. Yet, our brain transforms patterns of light falling on our retina into a coherent percept within a few hundred milliseconds. Possibly, low-level neural responses already carry substantial information to facilitate rapid characterization of the visual input. Here, we computationally estimated low-level contrast responses to computer-generated naturalistic images, and tested whether spatial pooling of these responses could predict image similarity at the neural and behavioral level. Using EEG, we show that statistics derived from pooled responses explain a large amount of variance between single-image evoked potentials (ERPs) in individual subjects. Dissimilarity analysis on multi-electrode ERPs demonstrated that large differences between images in pooled response statistics are predictive of more dissimilar patterns of evoked activity, whereas images with little difference in statistics give rise to highly similar evoked activity patterns. In a separate behavioral experiment, images with large differences in statistics were judged as different categories, whereas images with little differences were confused. These findings suggest that statistics derived from low-level contrast responses can be extracted in early visual processing and can be relevant for rapid judgment of visual similarity. We compared our results with two other, well- known contrast statistics: Fourier power spectra and higher-order properties of contrast distributions (skewness and kurtosis). Interestingly, whereas these statistics allow for accurate image categorization, they do not predict ERP response patterns or behavioral categorization confusions. These converging computational, neural and behavioral results suggest that statistics of pooled contrast responses contain information that corresponds with perceived visual similarity in a rapid, low-level categorization task.

Show MeSH

Related in: MedlinePlus

Example stimuli and computation of contrast statistics.(A), Example images of each of the 16 categories used in the behavioral and EEG experiment. Images contained randomly placed disks that differed in distribution, opacity, depth and size. Each category contained 16 unique images. (B), Consecutive steps in computing various contrast statistics. Weibull statistics are computed by filtering the image with a range of contrast filters with LGN-like scale- and gain properties, after which for each image location, the filter containing the minimal reliable response is selected. Responses of all selected filters are summed in a histogram to which the Weibull function is fitted, from which the beta and gamma parameters are derived using maximum likelihood estimation. (C), Power spectra parameters (top row) are extracted by taking the Fourier transform, averaging across directions, and computing the intercept and slope values of a line fitted to the average power spectrum. Higher-order properties of the contrast distribution (bottom row) are computed by filtering with a single-scale center-surround filter, after which skewness and kurtosis of the resulting contrast distribution are derived. Weibull statistics (multiscale local contrast) presumably contain information present in Fourier parameters (scale statistics) as well as local contrast distribution parameters (distribution statistics).
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC3475684&req=5

pcbi-1002726-g002: Example stimuli and computation of contrast statistics.(A), Example images of each of the 16 categories used in the behavioral and EEG experiment. Images contained randomly placed disks that differed in distribution, opacity, depth and size. Each category contained 16 unique images. (B), Consecutive steps in computing various contrast statistics. Weibull statistics are computed by filtering the image with a range of contrast filters with LGN-like scale- and gain properties, after which for each image location, the filter containing the minimal reliable response is selected. Responses of all selected filters are summed in a histogram to which the Weibull function is fitted, from which the beta and gamma parameters are derived using maximum likelihood estimation. (C), Power spectra parameters (top row) are extracted by taking the Fourier transform, averaging across directions, and computing the intercept and slope values of a line fitted to the average power spectrum. Higher-order properties of the contrast distribution (bottom row) are computed by filtering with a single-scale center-surround filter, after which skewness and kurtosis of the resulting contrast distribution are derived. Weibull statistics (multiscale local contrast) presumably contain information present in Fourier parameters (scale statistics) as well as local contrast distribution parameters (distribution statistics).

Mentions: For the second part of the question, we needed to obtain an image-specific behavioral judgment of perceived visual similarity. However, simply judging similarity of natural scenes is problematic, because these images obviously contain rich semantic content: there are many features of natural scenes that can be similar or dissimilar, which is likely to lead to different categorization strategies by different subjects. Also, it is uncertain to what extent specific semantic tags that are provided by the researcher (e.g. ‘openness’ or ‘naturalness’, [27]), can be uniformly interpreted as a relevant stimulus dimension that has a linear mapping to processing in early vision. Therefore, to explore the variance explained by contrast response statistics in a bottom-up way, we used stimuli that were simplified model images of natural scenes (‘dead leaves’, Fig. 2A), which have similar low-level structure as natural scenes (e.g. 1/f power spectra) but are devoid of semantic content. These images are created by filling a frame with objects - much like fallen leaves can fill a forest floor – and are used in computer vision to study, for example, how the appearance and the distribution of these objects influences the low-level structure of natural scenes [28]. By manipulating properties of the objects in a controlled manner, we created distinct image categories, and then tested whether differences between these categories in contrast statistics matched with behaviorally perceived similarity by letting human observers perform a same-different categorization task on all combinations of image categories.


Spatially pooled contrast responses predict neural and perceptual similarity of naturalistic image categories.

Groen II, Ghebreab S, Lamme VA, Scholte HS - PLoS Comput. Biol. (2012)

Example stimuli and computation of contrast statistics.(A), Example images of each of the 16 categories used in the behavioral and EEG experiment. Images contained randomly placed disks that differed in distribution, opacity, depth and size. Each category contained 16 unique images. (B), Consecutive steps in computing various contrast statistics. Weibull statistics are computed by filtering the image with a range of contrast filters with LGN-like scale- and gain properties, after which for each image location, the filter containing the minimal reliable response is selected. Responses of all selected filters are summed in a histogram to which the Weibull function is fitted, from which the beta and gamma parameters are derived using maximum likelihood estimation. (C), Power spectra parameters (top row) are extracted by taking the Fourier transform, averaging across directions, and computing the intercept and slope values of a line fitted to the average power spectrum. Higher-order properties of the contrast distribution (bottom row) are computed by filtering with a single-scale center-surround filter, after which skewness and kurtosis of the resulting contrast distribution are derived. Weibull statistics (multiscale local contrast) presumably contain information present in Fourier parameters (scale statistics) as well as local contrast distribution parameters (distribution statistics).
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC3475684&req=5

pcbi-1002726-g002: Example stimuli and computation of contrast statistics.(A), Example images of each of the 16 categories used in the behavioral and EEG experiment. Images contained randomly placed disks that differed in distribution, opacity, depth and size. Each category contained 16 unique images. (B), Consecutive steps in computing various contrast statistics. Weibull statistics are computed by filtering the image with a range of contrast filters with LGN-like scale- and gain properties, after which for each image location, the filter containing the minimal reliable response is selected. Responses of all selected filters are summed in a histogram to which the Weibull function is fitted, from which the beta and gamma parameters are derived using maximum likelihood estimation. (C), Power spectra parameters (top row) are extracted by taking the Fourier transform, averaging across directions, and computing the intercept and slope values of a line fitted to the average power spectrum. Higher-order properties of the contrast distribution (bottom row) are computed by filtering with a single-scale center-surround filter, after which skewness and kurtosis of the resulting contrast distribution are derived. Weibull statistics (multiscale local contrast) presumably contain information present in Fourier parameters (scale statistics) as well as local contrast distribution parameters (distribution statistics).
Mentions: For the second part of the question, we needed to obtain an image-specific behavioral judgment of perceived visual similarity. However, simply judging similarity of natural scenes is problematic, because these images obviously contain rich semantic content: there are many features of natural scenes that can be similar or dissimilar, which is likely to lead to different categorization strategies by different subjects. Also, it is uncertain to what extent specific semantic tags that are provided by the researcher (e.g. ‘openness’ or ‘naturalness’, [27]), can be uniformly interpreted as a relevant stimulus dimension that has a linear mapping to processing in early vision. Therefore, to explore the variance explained by contrast response statistics in a bottom-up way, we used stimuli that were simplified model images of natural scenes (‘dead leaves’, Fig. 2A), which have similar low-level structure as natural scenes (e.g. 1/f power spectra) but are devoid of semantic content. These images are created by filling a frame with objects - much like fallen leaves can fill a forest floor – and are used in computer vision to study, for example, how the appearance and the distribution of these objects influences the low-level structure of natural scenes [28]. By manipulating properties of the objects in a controlled manner, we created distinct image categories, and then tested whether differences between these categories in contrast statistics matched with behaviorally perceived similarity by letting human observers perform a same-different categorization task on all combinations of image categories.

Bottom Line: We compared our results with two other, well- known contrast statistics: Fourier power spectra and higher-order properties of contrast distributions (skewness and kurtosis).Interestingly, whereas these statistics allow for accurate image categorization, they do not predict ERP response patterns or behavioral categorization confusions.These converging computational, neural and behavioral results suggest that statistics of pooled contrast responses contain information that corresponds with perceived visual similarity in a rapid, low-level categorization task.

View Article: PubMed Central - PubMed

Affiliation: Cognitive Neuroscience Group, Department of Psychology, University of Amsterdam, Amsterdam, The Netherlands.

ABSTRACT
The visual world is complex and continuously changing. Yet, our brain transforms patterns of light falling on our retina into a coherent percept within a few hundred milliseconds. Possibly, low-level neural responses already carry substantial information to facilitate rapid characterization of the visual input. Here, we computationally estimated low-level contrast responses to computer-generated naturalistic images, and tested whether spatial pooling of these responses could predict image similarity at the neural and behavioral level. Using EEG, we show that statistics derived from pooled responses explain a large amount of variance between single-image evoked potentials (ERPs) in individual subjects. Dissimilarity analysis on multi-electrode ERPs demonstrated that large differences between images in pooled response statistics are predictive of more dissimilar patterns of evoked activity, whereas images with little difference in statistics give rise to highly similar evoked activity patterns. In a separate behavioral experiment, images with large differences in statistics were judged as different categories, whereas images with little differences were confused. These findings suggest that statistics derived from low-level contrast responses can be extracted in early visual processing and can be relevant for rapid judgment of visual similarity. We compared our results with two other, well- known contrast statistics: Fourier power spectra and higher-order properties of contrast distributions (skewness and kurtosis). Interestingly, whereas these statistics allow for accurate image categorization, they do not predict ERP response patterns or behavioral categorization confusions. These converging computational, neural and behavioral results suggest that statistics of pooled contrast responses contain information that corresponds with perceived visual similarity in a rapid, low-level categorization task.

Show MeSH
Related in: MedlinePlus