Skip to main content

Main menu

  • HOME
  • CONTENT
    • Early Release
    • Featured
    • Current Issue
    • Issue Archive
    • Collections
    • Podcast
  • ALERTS
  • FOR AUTHORS
    • Information for Authors
    • Fees
    • Journal Clubs
    • eLetters
    • Submit
  • EDITORIAL BOARD
  • ABOUT
    • Overview
    • Advertise
    • For the Media
    • Rights and Permissions
    • Privacy Policy
    • Feedback
  • SUBSCRIBE

User menu

  • Log in
  • My Cart

Search

  • Advanced search
Journal of Neuroscience
  • Log in
  • My Cart
Journal of Neuroscience

Advanced Search

Submit a Manuscript
  • HOME
  • CONTENT
    • Early Release
    • Featured
    • Current Issue
    • Issue Archive
    • Collections
    • Podcast
  • ALERTS
  • FOR AUTHORS
    • Information for Authors
    • Fees
    • Journal Clubs
    • eLetters
    • Submit
  • EDITORIAL BOARD
  • ABOUT
    • Overview
    • Advertise
    • For the Media
    • Rights and Permissions
    • Privacy Policy
    • Feedback
  • SUBSCRIBE
PreviousNext
Articles, Behavioral/Cognitive

Feature Extraction and Integration Underlying Perceptual Decision Making during Courtship Behavior

Jan Clemens and Bernhard Ronacher
Journal of Neuroscience 17 July 2013, 33 (29) 12136-12145; DOI: https://doi.org/10.1523/JNEUROSCI.0724-13.2013
Jan Clemens
1Behavioral Physiology Group, Department of Biology, Humboldt-Universität zu Berlin and
2Bernstein Center for Computational Neuroscience, 10115 Berlin, Germany
  • Find this author on Google Scholar
  • Find this author on PubMed
  • Search for this author on this site
Bernhard Ronacher
1Behavioral Physiology Group, Department of Biology, Humboldt-Universität zu Berlin and
2Bernstein Center for Computational Neuroscience, 10115 Berlin, Germany
  • Find this author on Google Scholar
  • Find this author on PubMed
  • Search for this author on this site
  • Article
  • Figures & Data
  • Info & Metrics
  • eLetters
  • PDF
Loading

Article Figures & Data

Figures

  • Figure 1.
    • Download figure
    • Open in new tab
    • Download powerpoint
    Figure 1.

    Structure and performance of the model and structure of stimuli. a, Layout of our model of a perceptual decision-making system. b, Dependence of cross-validation performance on the number of filters and the duration of the associated filters. The filter duration was set by the number of basis components used to represent the filter shapes, i.e., the shortest filter (8 ms) consisted of a single basis component, the longest filter (64 ms) consisted of 16 basis components. c, A transition from unimodal (left: single component, 8 ms) to bimodal filters (right: two components, 11 ms) correlated with a large increase in model performance (performance 0.36 and 0.62, respectively). This suggests that the detection of transients is fundamental to song recognition in grasshoppers. d, Filter shapes were consistent across a wide range of filter durations. Shown are the average filters across cross-validation runs of the two-filter model (first filter: red, left; second filter: green, right) for a range of filter durations. e, Envelope of the calling song of a male C. biguttulus (gray) and its reduction to a block stimulus (black). Annotations indicate the signal parameters, which varied in our stimulus set. f, A great variety of naturalistic patterns was created by starting from a simple block-like stimulus and changing the pause duration, the depth of the pause, the height of the onset accentuation, and the intensity of the syllable plateau. Onset and offset were defined relative to plateau intensity.

  • Figure 2.
    • Download figure
    • Open in new tab
    • Download powerpoint
    Figure 2.

    Two-filter model for block-like stimuli. a, Filters for both feature detectors (red and green, respectively, duration 48 ms). Colored arrows mark the time at which a downstroke in the stimulus is detected, which determines the response phase of the filter. Black arrow indicates the moment of the predicted response relative to the filter. b, Example stimulus and response (top, black, pause duration 12 ms, onset 12 dB, offset 18 dB). The horizontal black line marks 0 dB. Red and green traces show the stimulus filtered by filter 1 and 2, respectively. The responses corresponding to onsets and offsets in the stimulus (downward and upward arrows, respectively) are indicated by similar arrows in the filtered stimulus. Note that the responses of the second filter (green) were delayed relative to those of the first filter as the downstrokes (colored arrows in a) occurred at different phases during each filter. For analysis, the whole stimulus set was normalized to zero mean and unit variance. c, Nonlinearity of the first (top) and second (bottom) feature detector, transforming the output of the filter. Gray areas show distribution of output values of the filters. d, The filtered stimulus after applying the nonlinearity. The step-like nonlinearity created pulse-like outputs. Numbers right of each trace indicate the feature values, obtained by averaging the filtered and nonlinearly transformed stimulus over time. e, f, Behavioral responses plotted against feature values (e) and predicted responses (f) obtained by linear combination of both feature values (regression formula shown in f, f1, feature value of filter 1; f2, feature value of filter 2). Black dots in e and f mark the values for the example stimulus in b. r2 values show the coefficient of correlation between the behavior and the feature values (e) or the predicted response (f).

  • Figure 3.
    • Download figure
    • Open in new tab
    • Download powerpoint
    Figure 3.

    Tuning for pause duration of the feature detectors, the model, and the behavior. a, The excitatory feature was a steeply rising, weak bandpass filter for pause duration (red) and the suppressive feature was a high-pass filter with a shallower slope (green). Linear combination of both features yielded a sharp bandpass tuning for pause duration (blue). Tuning curves were normalized to range between 0 and 1 to facilitate comparison. b, Pause tuning of the model and behavior (blue and black trace, respectively). c, Pause tuning in individual grasshoppers exhibited weak bandpass tuning resembling that of the first feature (top) or monotonously increasing tuning resembling that of the second feature (bottom). The model could reproduce these interindividual variations by changing the relative weights of both features from [3.84 −1.54] in the original model to [1 0] and [1 2], respectively (blue traces). d, Block stimuli with 80 ms syllable and 4, 18, or 50 ms pauses. e, Stimuli in d filtered with the excitatory and the suppressive filter (red and green, respectively; filters shown in Fig. 2a). f, Output of the excitatory and suppressive feature detector (red and green, respectively) for the stimuli in d. Feature values shown to the right of each output trace.

  • Figure 4.
    • Download figure
    • Open in new tab
    • Download powerpoint
    Figure 4.

    Tuning for gaps and influence of intensity. a, Behavioral response (black) and model prediction (blue). The model reproduced behavior very well (r2 = 0.95) even though the training set contained no stimuli with gaps in the syllable plateau. b, Tuning of both features for gap duration. c, d, Pause tuning for block stimuli without accentuated onsets and with plateau intensities of 64, 70, and 76 dB (see red, orange, and yellow insets for schematic stimuli) in the data (c) and the model (d). e, Dependence of feature values (red and green) on plateau intensity for stimuli with a pause duration of 32 ms and an offset of 99 dB. Shown are the feature's values, scaled by their absolute weight. The excitatory feature (red) increased with intensity, while the suppressive feature (green) was relatively constant. f, Predicted and measured response values (blue and black, respectively) for the stimuli used in e. Responses increased with intensity.

  • Figure 5.
    • Download figure
    • Open in new tab
    • Download powerpoint
    Figure 5.

    Impact of onset accentuation and offset on pause tuning. Response values are color coded (see color bars). Columns correspond to the tuning of both features, the model, and the experimental data, respectively. Subtracting the suppressive from the excitatory feature yielded the model response. Different rows correspond to different slices of the same, 3D dataset: a–d, Dependence of feature values and behavior on onset and offset at an optimal pause of 12 ms. Stimuli for which the sum of onset and offsets exceeds a certain value are responded to strongly (>0.5, black line in c and d). e–h, Onset and offset at a long pause of 26 ms. Large values of the suppressive feature (arrow in f) led to reduced responses for large offsets (arrow in h). i–l, Pause and offset at an onset of 9 dB. At large offsets, tuning of pause duration was narrowest (arrow in l) due to large values of the suppressive feature (arrow in j).

Back to top

In this issue

The Journal of Neuroscience: 33 (29)
Journal of Neuroscience
Vol. 33, Issue 29
17 Jul 2013
  • Table of Contents
  • Table of Contents (PDF)
  • About the Cover
  • Index by author
  • Advertising (PDF)
  • Ed Board (PDF)
Email

Thank you for sharing this Journal of Neuroscience article.

NOTE: We request your email address only to inform the recipient that it was you who recommended this article, and that it is not junk mail. We do not retain these email addresses.

Enter multiple addresses on separate lines or separate them with commas.
Feature Extraction and Integration Underlying Perceptual Decision Making during Courtship Behavior
(Your Name) has forwarded a page to you from Journal of Neuroscience
(Your Name) thought you would be interested in this article in Journal of Neuroscience.
CAPTCHA
This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.
Print
View Full Page PDF
Citation Tools
Feature Extraction and Integration Underlying Perceptual Decision Making during Courtship Behavior
Jan Clemens, Bernhard Ronacher
Journal of Neuroscience 17 July 2013, 33 (29) 12136-12145; DOI: 10.1523/JNEUROSCI.0724-13.2013

Citation Manager Formats

  • BibTeX
  • Bookends
  • EasyBib
  • EndNote (tagged)
  • EndNote 8 (xml)
  • Medlars
  • Mendeley
  • Papers
  • RefWorks Tagged
  • Ref Manager
  • RIS
  • Zotero
Respond to this article
Request Permissions
Share
Feature Extraction and Integration Underlying Perceptual Decision Making during Courtship Behavior
Jan Clemens, Bernhard Ronacher
Journal of Neuroscience 17 July 2013, 33 (29) 12136-12145; DOI: 10.1523/JNEUROSCI.0724-13.2013
del.icio.us logo Digg logo Reddit logo Twitter logo CiteULike logo Facebook logo Google logo Mendeley logo
  • Tweet Widget
  • Facebook Like
  • Google Plus One

Jump to section

  • Article
    • Abstract
    • Introduction
    • Materials and Methods
    • Results
    • Discussion
    • Footnotes
    • References
  • Figures & Data
  • Info & Metrics
  • eLetters
  • PDF

Responses to this article

Respond to this article

Jump to comment:

No eLetters have been published for this article.

Related Articles

Cited By...

More in this TOC Section

Articles

  • Choice Behavior Guided by Learned, But Not Innate, Taste Aversion Recruits the Orbitofrontal Cortex
  • Maturation of Spontaneous Firing Properties after Hearing Onset in Rat Auditory Nerve Fibers: Spontaneous Rates, Refractoriness, and Interfiber Correlations
  • Insulin Treatment Prevents Neuroinflammation and Neuronal Injury with Restored Neurobehavioral Function in Models of HIV/AIDS Neurodegeneration
Show more Articles

Behavioral/Cognitive

  • The Spatial Reach of Neuronal Coherence and Spike-field Coupling across the Human Neocortex
  • Reward enhances memory via age-varying online and offline neural mechanisms across development
  • Animal-to-Animal Variability in Partial Hippocampal Remapping in Repeated Environments
Show more Behavioral/Cognitive
  • Home
  • Alerts
  • Visit Society for Neuroscience on Facebook
  • Follow Society for Neuroscience on Twitter
  • Follow Society for Neuroscience on LinkedIn
  • Visit Society for Neuroscience on Youtube
  • Follow our RSS feeds

Content

  • Early Release
  • Current Issue
  • Issue Archive
  • Collections

Information

  • For Authors
  • For Advertisers
  • For the Media
  • For Subscribers

About

  • About the Journal
  • Editorial Board
  • Privacy Policy
  • Contact
(JNeurosci logo)
(SfN logo)

Copyright © 2022 by the Society for Neuroscience.
JNeurosci Online ISSN: 1529-2401

The ideas and opinions expressed in JNeurosci do not necessarily reflect those of SfN or the JNeurosci Editorial Board. Publication of an advertisement or other product mention in JNeurosci should not be construed as an endorsement of the manufacturer’s claims. SfN does not assume any responsibility for any injury and/or damage to persons or property arising from or related to any use of any material contained in JNeurosci.