Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 86 of the 86 references it cites were found in a registry or an index.

Prior object-knowledge sharpens properties of early visual feature-detectors

Open access
Teufel, Christoph, Dakin, Steven C., Fletcher, Paul C. · 2018 · Scientific Reports

Abstract

Early stages of visual processing are carried out by neural circuits activated by simple and specific features, such as the orientation of an edge. A fundamental question in human vision is how the brain organises such intrinsically local information into meaningful properties of objects. Classic models of visual processing emphasise a one-directional flow of information from early feature-detectors to higher-level information-processing. By contrast to this view, and in line with predictive-coding models of perception, here, we provide evidence from human vision that high-level object representations dynamically interact with the earliest stages of cortical visual processing. In two experiments, we used ambiguous stimuli that, depending on the observer’s prior object-knowledge, can be perceived as either coherent objects or as a collection of meaningless patches. By manipulating object knowledge we were able to determine its impact on processing of low-level features while keeping sensory stimulation identical. Both studies demonstrate that perception of local features is facilitated in a manner consistent with an observer’s high-level object representation (i.e., with no effect on object-inconsistent features). Our results cannot be ascribed to attentional influences. Rather, they suggest that high-level object representations interact with and sharpen early feature-detectors, optimising their performance for the current perceptual context.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1038/s41598-018-28845-5

Cites

86 of 86 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1 Example of a two-tone (aka ‘Mooney’) image. On first viewing, this image appears as a series of meaningless black and white patches. To experience the effect of top-down knowledge on perception, the reader should look at the template in Fig. 2a for some time before returning to this two-tone image.
Figure 1 Example of a two-tone (aka ‘Mooney’) image. On first viewing, this image appears as a series of meaningless black and white patches. To experience the effect of top-down knowledge on perception, the reader should look at the template in Fig. 2a for some time before returning to this two-tone image.CC BY · PubMed Central
Figure 2 Stimulus generation, experimental procedure, and tasks in Experiments 1 and 2. (a) The left panel shows the template image from which the two-tone image in the right panel was generated by binarising greyscale values around a threshold. Red ovals in the right panel indicate two example locations that lack a contour in the two-tone image but contain a meaningful contour in the template image. (b) Small edge elements were embedded in two-tone images at such locations (right panel). Locations and orientations were identified by filtering the template image with a bank of orientation-tuned Gabor filters (left panel, orientation is colour-coded). For the purpose of illustration, the edge element is shown in high contrast. (c) Testing started with the choice of 20 two-tone images and the measurement of a preliminary threshold (Experiment 1 only). Thresholds were then estimated in two Before blocks (max. 480 trials), in which observers saw probe stimuli embedded in two-tone images. Subsequently, observers were exposed to the template images in two blocks. Finally, thresholds were measured again in two After blocks (max 480 trials), in which observers saw exactly the same stimuli as in the Before blocks. (d) The top panel shows the interval structure of a single trial. After presentation of the two-tone image and a fixation square, which indicated the location and possible orientation of the edge probe, a small grey patch appeared on the two-tone image. Every trial consisted of two such intervals: In Experiment 1, the grey patch in one interval contained a faint edge, the other did not (lower panel). Within each trial, observers were asked to indicate which of the two intervals contained the edge element. In Experiment 2, the grey patch of both intervals contained high contrast edge elements but these differed slightly in orientation (lower panel). The task for the participants was to indicate which interval contained the more clockwise oriented edge element. Contrast and orientation difference were adjusted across trials to allow estimating the observers’ threshold in detecting an edge (Experiment 1) and discriminating between orientations of two edges (Experiment 2). In both experiments, edge elements were shown along the invisible contour (Along condition) or, in a control condition, orthogonal to the contour (Orthogonal condition). (e) Illustration of the tasks used to familiarise observers with the templates. Between the Before and the After blocks, during which thresholds for edge detection and discrimination were measured, observers participated in two tasks to provide them with prior object knowledge. First, in an active search paradigm (left panel), they viewed each template image individually and were asked to respond to the appearance of a dot by clicking on it with a cursor. The dot did not appear with an abrupt onset but was very slowly ramped up in luminance, forcing observers to constantly scan the image and, thus, actively pick up information from it. In a second, passive viewing task (right panel), observers were shown a gradual blend from template image to two-tone image and back again.
Figure 2 Stimulus generation, experimental procedure, and tasks in Experiments 1 and 2. (a) The left panel shows the template image from which the two-tone image in the right panel was generated by binarising greyscale values around a threshold. Red ovals in the right panel indicate two example locations that lack a contour in the two-tone image but contain a meaningful contour in the template image. (b) Small edge elements were embedded in two-tone images at such locations (right panel). Locations and orientations were identified by filtering the template image with a bank of orientation-tuned Gabor filters (left panel, orientation is colour-coded). For the purpose of illustration, the edge element is shown in high contrast. (c) Testing started with the choice of 20 two-tone images and the measurement of a preliminary threshold (Experiment 1 only). Thresholds were then estimated in two Before blocks (max. 480 trials), in which observers saw probe stimuli embedded in two-tone images. Subsequently, observers were exposed to the template images in two blocks. Finally, thresholds were measured again in two After blocks (max 480 trials), in which observers saw exactly the same stimuli as in the Before blocks. (d) The top panel shows the interval structure of a single trial. After presentation of the two-tone image and a fixation square, which indicated the location and possible orientation of the edge probe, a small grey patch appeared on the two-tone image. Every trial consisted of two such intervals: In Experiment 1, the grey patch in one interval contained a faint edge, the other did not (lower panel). Within each trial, observers were asked to indicate which of the two intervals contained the edge element. In Experiment 2, the grey patch of both intervals contained high contrast edge elements but these differed slightly in orientation (lower panel). The task for the participants was to indicate which interval contained the more clockwise oriented edge element. Contrast and orientation difference were adjusted across trials to allow estimating the observers’ threshold in detecting an edge (Experiment 1) and discriminating between orientations of two edges (Experiment 2). In both experiments, edge elements were shown along the invisible contour (Along condition) or, in a control condition, orthogonal to the contour (Orthogonal condition). (e) Illustration of the tasks used to familiarise observers with the templates. Between the Before and the After blocks, during which thresholds for edge detection and discrimination were measured, observers participated in two tasks to provide them with prior object knowledge. First, in an active search paradigm (left panel), they viewed each template image individually and were asked to respond to the appearance of a dot by clicking on it with a cursor. The dot did not appear with an abrupt onset but was very slowly ramped up in luminance, forcing observers to constantly scan the image and, thus, actively pick up information from it. In a second, passive viewing task (right panel), observers were shown a gradual blend from template image to two-tone image and back again.CC BY · PubMed Central
Figure 3 Results of Experiment 1. (a and b) Psychometric functions from one example observer with dotted lines illustrating thresholds. Panel (a) shows results from the Along condition, panel (b) from the Orthogonal control condition. As can be seen in panel (a), the leftward shift of the psychometric function in the After session (in red) compared to the Before session (in green) for the Along condition indicates a lower absolute contrast detection threshold for this condition as a consequence of template exposure. (c) The plot shows the proportional change in absolute contrast detection threshold (mean ± SEM) from Before to After session in the test and the control condition. (d) This graph illustrates the parameter values that characterise the slope of the psychometric function (mean ± SEM).
Figure 3 Results of Experiment 1. (a and b) Psychometric functions from one example observer with dotted lines illustrating thresholds. Panel (a) shows results from the Along condition, panel (b) from the Orthogonal control condition. As can be seen in panel (a), the leftward shift of the psychometric function in the After session (in red) compared to the Before session (in green) for the Along condition indicates a lower absolute contrast detection threshold for this condition as a consequence of template exposure. (c) The plot shows the proportional change in absolute contrast detection threshold (mean ± SEM) from Before to After session in the test and the control condition. (d) This graph illustrates the parameter values that characterise the slope of the psychometric function (mean ± SEM).CC BY · PubMed Central
Figure 4 Results of Experiment 2. (a and b) Psychometric functions from one example observer with dotted lines illustrating thresholds. The dot size indicates relative number of trials per stimulus level. Panel (a) shows results from the Along condition, panel (b) from the Orthogonal control condition. As can be seen in (a), the psychometric function in the After session (red) is steeper than in the Before session (green) for the Along condition indicating that template exposure improved orientation discrimination. (c) The plot shows the proportional change in orientation difference threshold (mean ± SEM) from the Before to the After session, in the test and the control condition. (d) This graph illustrates the orientation difference between reference and comparison at which the two were perceived as equal (Point of Subjective Equality or PSE) (mean ± SEM). A PSE of 0 indicates that the two edges were perceived to have the same orientation when there was no difference between them.
Figure 4 Results of Experiment 2. (a and b) Psychometric functions from one example observer with dotted lines illustrating thresholds. The dot size indicates relative number of trials per stimulus level. Panel (a) shows results from the Along condition, panel (b) from the Orthogonal control condition. As can be seen in (a), the psychometric function in the After session (red) is steeper than in the Before session (green) for the Along condition indicating that template exposure improved orientation discrimination. (c) The plot shows the proportional change in orientation difference threshold (mean ± SEM) from the Before to the After session, in the test and the control condition. (d) This graph illustrates the orientation difference between reference and comparison at which the two were perceived as equal (Point of Subjective Equality or PSE) (mean ± SEM). A PSE of 0 indicates that the two edges were perceived to have the same orientation when there was no difference between them.CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 86 references

Research works printed but not linked: 3

  1. De Valois, R. L. & De Valois, K. K. Spatial Vision. (Oxford University Press, 1991).
  2. Pasupathy, A. in Fundamentals of Vision: Low and Mid-Level Processes in Perception (eds. Martinez-Conde, S., Macknik, S., Martinez, M., Alonso, J. M. & Tse, P.) 154, 293–313 (Elsevier, 2006).
  3. Marr, D. Vision. (The MIT Press, 1982).

Cited by

See all 129

Metadata

Authors
Teufel, C., Dakin, S. C., & Fletcher, P. C.
Affiliations
  • Cardiff University (GB)
  • University of Auckland (NZ)
  • University of Cambridge (GB)
Corresponding author
Teufel, Christoph
Venue
Scientific Reports
Published
2018-07-18
Volume
8
Issue
1
Pages
10853-10853
ISSN
2045-2322 (ISSN-L)
OpenAlex
W2809827720
Cited by
119
References
86
Citations by year
2026: 5 · 2025: 11 · 2024: 7 · 2023: 44 · 2022: 9 · 2021: 15 · 2020: 15 · 2019: 12 · 2018: 1
FWCI
6.40
Type
journal article
Language
en
Open access
Licence
Locations
  • Scientific Reports (publishedVersion, cc-by)
  • Europe PMC (PubMed Central) (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • ORCA Online Research @Cardiff (Cardiff University) (acceptedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion)
  • ResearchSpace (University of Auckland) (submittedVersion, cc-by)
  • Apollo (University of Cambridge) (submittedVersion, cc-by)
  • Apollo (cc-by)
Funders
  • Wellcome Trust (WT095692MA)
  • Bernard Wolfe Health Neuroscience Fund
Topics
  • Life Sciences › Neuroscience › Cognitive Neuroscience › Visual perception and processing mechanisms
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Visual Attention and Saliency Detection
  • Life Sciences › Neuroscience › Cognitive Neuroscience › Face Recognition and Perception
Keywords
predictive coding, object representation, ambiguous stimuli, top-down modulation, object perception, visual cortex
MeSH terms
Adolescent, Adult, Attention, Female, Humans, Male, Orientation, Photic Stimulation, Visual Cortex, Visual Perception, Young Adult

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Prior object-knowledge sharpens properties of early visual feature-detectors. https://app.cochat.ai/papers/10.1038/s41598-018-28845-5
The paper
Teufel, C., Dakin, S. C., & Fletcher, P. C. (2018). Prior object-knowledge sharpens properties of early visual feature-detectors. Scientific Reports. https://doi.org/10.1038/s41598-018-28845-5

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.