Limits...
The Elementary Operations of Human Vision Are Not Reducible to Template-Matching.

Neri P - PLoS Comput. Biol. (2015)

Bottom Line: We demonstrate that human visual processing can operate under conditions that are indistinguishable from linear-nonlinear transduction with respect to substantially different stimulus attributes of a uniquely specified target signal with associated behavioural task.Our results inform and constrain efforts at obtaining and interpreting comprehensive characterizations of the human sensory process by demonstrating its inescapably nonlinear nature, even under conditions that have been painstakingly fine-tuned to facilitate template-matching behaviour and to produce results that, at some level of inspection, do conform to linear filtering predictions.They also suggest that compliance with linear transduction may be the targeted outcome of carefully crafted nonlinear circuits, rather than default behaviour exhibited by basic components.

View Article: PubMed Central - PubMed

Affiliation: Laboratoire des Systèmes Perceptifs, CNRS UMR 8248, 29 rue d'Ulm, 75005 Paris, France.

ABSTRACT
It is generally acknowledged that biological vision presents nonlinear characteristics, yet linear filtering accounts of visual processing are ubiquitous. The template-matching operation implemented by the linear-nonlinear cascade (linear filter followed by static nonlinearity) is the most widely adopted computational tool in systems neuroscience. This simple model achieves remarkable explanatory power while retaining analytical tractability, potentially extending its reach to a wide range of systems and levels in sensory processing. The extent of its applicability to human behaviour, however, remains unclear. Because sensory stimuli possess multiple attributes (e.g. position, orientation, size), the issue of applicability may be asked by considering each attribute one at a time in relation to a family of linear-nonlinear models, or by considering all attributes collectively in relation to a specified implementation of the linear-nonlinear cascade. We demonstrate that human visual processing can operate under conditions that are indistinguishable from linear-nonlinear transduction with respect to substantially different stimulus attributes of a uniquely specified target signal with associated behavioural task. However, no specific implementation of a linear-nonlinear cascade is able to account for the entire collection of results across attributes; a satisfactory account at this level requires the introduction of a small gain-control circuit, resulting in a model that no longer belongs to the linear-nonlinear family. Our results inform and constrain efforts at obtaining and interpreting comprehensive characterizations of the human sensory process by demonstrating its inescapably nonlinear nature, even under conditions that have been painstakingly fine-tuned to facilitate template-matching behaviour and to produce results that, at some level of inspection, do conform to linear filtering predictions. They also suggest that compliance with linear transduction may be the targeted outcome of carefully crafted nonlinear circuits, rather than default behaviour exhibited by basic components.

No MeSH data available.


Perceptual filters used by human observers to detect/discriminate stimuli.A-D show aggregate (across observers) perceptual filters (PF) for detecting the vertical Gabor target (icons to the left of A) returned by reverse correlating different noise types: 2D (A), 1D (B), Θ (C) and SF (D; see Fig 1 for image samples of all four classes). Red/blue lines in A show positive/negative contours through a Gabor fit to the data. Green trace in B and green lines in C-D indicate target signal. Grey shaded regions in B-D show ±1 SEM. Panels E,G show results corresponding to A,C (2D and Θ noise probes) for discriminating vertical from horizontal Gabor targets (icons to the left of E). F plots match between 2D PF’s and target/non-target signals on y/x axes (non-target as specified in the discrimination task) for both detection (solid) and discrimination (open) across different observers (different data points). H plots filter amplitude of orientation-tuned PF’s (C,G) at 0 (peak, y axis) and ±π/2 (trough) using similar conventions. Error bars (±1 SEM) in F,H are sometimes smaller than symbols and therefore not visible. The centre value of 2D PF’s (A,E) is effectively undefined because occluded by the fixation marker (see Methods).
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4640830&req=5

pcbi.1004499.g002: Perceptual filters used by human observers to detect/discriminate stimuli.A-D show aggregate (across observers) perceptual filters (PF) for detecting the vertical Gabor target (icons to the left of A) returned by reverse correlating different noise types: 2D (A), 1D (B), Θ (C) and SF (D; see Fig 1 for image samples of all four classes). Red/blue lines in A show positive/negative contours through a Gabor fit to the data. Green trace in B and green lines in C-D indicate target signal. Grey shaded regions in B-D show ±1 SEM. Panels E,G show results corresponding to A,C (2D and Θ noise probes) for discriminating vertical from horizontal Gabor targets (icons to the left of E). F plots match between 2D PF’s and target/non-target signals on y/x axes (non-target as specified in the discrimination task) for both detection (solid) and discrimination (open) across different observers (different data points). H plots filter amplitude of orientation-tuned PF’s (C,G) at 0 (peak, y axis) and ±π/2 (trough) using similar conventions. Error bars (±1 SEM) in F,H are sometimes smaller than symbols and therefore not visible. The centre value of 2D PF’s (A,E) is effectively undefined because occluded by the fixation marker (see Methods).

Mentions: Stimuli lasted 80 ms and consisted of 3 regions (∼3×3 deg each): a central ‘probe’ region at fixation (the fixation marker consisted of a dark pixel in the centre measuring ∼3×3 arcmin and never disappeared); above and below it, two identical ‘reference’ regions containing the template (see S1 Video). The template signal consisted of a cosine-phase (peaking at centre) vertical Gabor wavelet (standard deviation (SD) of Gaussian envelope 0.5 deg, spatial frequency 1 cycles/deg), and was presented at 17% contrast within the reference regions (background luminance 30 cd/m2). On each trial, observers saw two instances of the stimulus separated by a 500-ms gap. Reference regions were identical on both instances (and across all trials), thus providing no useful information for performing the task; their purpose was to remind observers of the target signal shape, so as to facilitate a template-matching strategy [21, 26]. The probe region contained target signal plus noise mask on one instance, and non-target signal plus noise mask on the other instance. Observers were asked to select the instance (first or second) that contained the target signal by pressing one of two buttons, after which they received immediate trial-by-trial feedback (correct/incorrect). The target signal was simply the template Gabor wavelet described above (see also Fig 1A), presented at 8% (alternatively 4%) contrast in the detection (alternatively discrimination) task. The non-target signal was blank for the detection task (Fig 1F), and a horizontal variant of the target signal for the discrimination task (see icons to the left of Fig 2E). Except for taking on a different orientation, the non-target signal in the discrimination task was identical to the target-signal. Data for the two tasks were collected in separate blocks of 100 trials each. We also collected separate data for a ‘symmetric’ variant of the discrimination task. In this additional experiment, two identical reference regions containing non-target templates were presented to the left and to the right of the central probe region.


The Elementary Operations of Human Vision Are Not Reducible to Template-Matching.

Neri P - PLoS Comput. Biol. (2015)

Perceptual filters used by human observers to detect/discriminate stimuli.A-D show aggregate (across observers) perceptual filters (PF) for detecting the vertical Gabor target (icons to the left of A) returned by reverse correlating different noise types: 2D (A), 1D (B), Θ (C) and SF (D; see Fig 1 for image samples of all four classes). Red/blue lines in A show positive/negative contours through a Gabor fit to the data. Green trace in B and green lines in C-D indicate target signal. Grey shaded regions in B-D show ±1 SEM. Panels E,G show results corresponding to A,C (2D and Θ noise probes) for discriminating vertical from horizontal Gabor targets (icons to the left of E). F plots match between 2D PF’s and target/non-target signals on y/x axes (non-target as specified in the discrimination task) for both detection (solid) and discrimination (open) across different observers (different data points). H plots filter amplitude of orientation-tuned PF’s (C,G) at 0 (peak, y axis) and ±π/2 (trough) using similar conventions. Error bars (±1 SEM) in F,H are sometimes smaller than symbols and therefore not visible. The centre value of 2D PF’s (A,E) is effectively undefined because occluded by the fixation marker (see Methods).
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4640830&req=5

pcbi.1004499.g002: Perceptual filters used by human observers to detect/discriminate stimuli.A-D show aggregate (across observers) perceptual filters (PF) for detecting the vertical Gabor target (icons to the left of A) returned by reverse correlating different noise types: 2D (A), 1D (B), Θ (C) and SF (D; see Fig 1 for image samples of all four classes). Red/blue lines in A show positive/negative contours through a Gabor fit to the data. Green trace in B and green lines in C-D indicate target signal. Grey shaded regions in B-D show ±1 SEM. Panels E,G show results corresponding to A,C (2D and Θ noise probes) for discriminating vertical from horizontal Gabor targets (icons to the left of E). F plots match between 2D PF’s and target/non-target signals on y/x axes (non-target as specified in the discrimination task) for both detection (solid) and discrimination (open) across different observers (different data points). H plots filter amplitude of orientation-tuned PF’s (C,G) at 0 (peak, y axis) and ±π/2 (trough) using similar conventions. Error bars (±1 SEM) in F,H are sometimes smaller than symbols and therefore not visible. The centre value of 2D PF’s (A,E) is effectively undefined because occluded by the fixation marker (see Methods).
Mentions: Stimuli lasted 80 ms and consisted of 3 regions (∼3×3 deg each): a central ‘probe’ region at fixation (the fixation marker consisted of a dark pixel in the centre measuring ∼3×3 arcmin and never disappeared); above and below it, two identical ‘reference’ regions containing the template (see S1 Video). The template signal consisted of a cosine-phase (peaking at centre) vertical Gabor wavelet (standard deviation (SD) of Gaussian envelope 0.5 deg, spatial frequency 1 cycles/deg), and was presented at 17% contrast within the reference regions (background luminance 30 cd/m2). On each trial, observers saw two instances of the stimulus separated by a 500-ms gap. Reference regions were identical on both instances (and across all trials), thus providing no useful information for performing the task; their purpose was to remind observers of the target signal shape, so as to facilitate a template-matching strategy [21, 26]. The probe region contained target signal plus noise mask on one instance, and non-target signal plus noise mask on the other instance. Observers were asked to select the instance (first or second) that contained the target signal by pressing one of two buttons, after which they received immediate trial-by-trial feedback (correct/incorrect). The target signal was simply the template Gabor wavelet described above (see also Fig 1A), presented at 8% (alternatively 4%) contrast in the detection (alternatively discrimination) task. The non-target signal was blank for the detection task (Fig 1F), and a horizontal variant of the target signal for the discrimination task (see icons to the left of Fig 2E). Except for taking on a different orientation, the non-target signal in the discrimination task was identical to the target-signal. Data for the two tasks were collected in separate blocks of 100 trials each. We also collected separate data for a ‘symmetric’ variant of the discrimination task. In this additional experiment, two identical reference regions containing non-target templates were presented to the left and to the right of the central probe region.

Bottom Line: We demonstrate that human visual processing can operate under conditions that are indistinguishable from linear-nonlinear transduction with respect to substantially different stimulus attributes of a uniquely specified target signal with associated behavioural task.Our results inform and constrain efforts at obtaining and interpreting comprehensive characterizations of the human sensory process by demonstrating its inescapably nonlinear nature, even under conditions that have been painstakingly fine-tuned to facilitate template-matching behaviour and to produce results that, at some level of inspection, do conform to linear filtering predictions.They also suggest that compliance with linear transduction may be the targeted outcome of carefully crafted nonlinear circuits, rather than default behaviour exhibited by basic components.

View Article: PubMed Central - PubMed

Affiliation: Laboratoire des Systèmes Perceptifs, CNRS UMR 8248, 29 rue d'Ulm, 75005 Paris, France.

ABSTRACT
It is generally acknowledged that biological vision presents nonlinear characteristics, yet linear filtering accounts of visual processing are ubiquitous. The template-matching operation implemented by the linear-nonlinear cascade (linear filter followed by static nonlinearity) is the most widely adopted computational tool in systems neuroscience. This simple model achieves remarkable explanatory power while retaining analytical tractability, potentially extending its reach to a wide range of systems and levels in sensory processing. The extent of its applicability to human behaviour, however, remains unclear. Because sensory stimuli possess multiple attributes (e.g. position, orientation, size), the issue of applicability may be asked by considering each attribute one at a time in relation to a family of linear-nonlinear models, or by considering all attributes collectively in relation to a specified implementation of the linear-nonlinear cascade. We demonstrate that human visual processing can operate under conditions that are indistinguishable from linear-nonlinear transduction with respect to substantially different stimulus attributes of a uniquely specified target signal with associated behavioural task. However, no specific implementation of a linear-nonlinear cascade is able to account for the entire collection of results across attributes; a satisfactory account at this level requires the introduction of a small gain-control circuit, resulting in a model that no longer belongs to the linear-nonlinear family. Our results inform and constrain efforts at obtaining and interpreting comprehensive characterizations of the human sensory process by demonstrating its inescapably nonlinear nature, even under conditions that have been painstakingly fine-tuned to facilitate template-matching behaviour and to produce results that, at some level of inspection, do conform to linear filtering predictions. They also suggest that compliance with linear transduction may be the targeted outcome of carefully crafted nonlinear circuits, rather than default behaviour exhibited by basic components.

No MeSH data available.