Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Isotropic sequence order learning using a novel linear algorithm in a closed loop behavioural system.

B Porr1, P Wörgötter

  • 1Department of Psychology, University of Stirling, Stirling, UK.

Bio Systems
|December 3, 2002
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Classification of chemically modified red blood cells in microflow using machine learning video analysis.

Soft matter·2023
Same author

Proof of concept evaluation of the electroencephalophone as a discriminator between wakefulness and general anaesthesia.

British journal of anaesthesia·2013
Same author

Synaptic modifications depend on synapse location and activity: a biophysical model of STDP.

Bio Systems·2005
Same author

How to "hear" visual disparities: real-time stereoscopic spatial depth analysis using temporal resonance.

Biological cybernetics·1998
Same journal

Spatiotemporal bursting in simulated cultures of cortical neurons.

Bio Systems·2026
Same journal

A brief discussion on recent models shedding light on how life emerged.

Bio Systems·2026
Same journal

Memory-based strategy reputation and adaptive learning in spatial evolutionary games: A robust agent-based model for cooperation dynamics.

Bio Systems·2026
Same journal

Coherent Photonic Biofields: Revisiting Fritz-Albert Popp's Hypothesis.

Bio Systems·2026
Same journal

Ruliological Resilience: Pattern Restoration and Robustness in Wolfram Patterns. A Basis for Regeneration, Not Just in Cone Shells?

Bio Systems·2026
Same journal

The quantum-to-classical transducer: A thermodynamic and quantum mechanical framework for the emergence of bioenergetics.

Bio Systems·2026
See all related articles

This study introduces an isotropic algorithm for sequence order learning, enabling robots to anticipate events by replacing external rewards with internal reflex loops. This allows for autonomous behavior and earlier anticipatory actions.

Area of Science:

  • Robotics
  • Machine Learning
  • Computational Neuroscience

Background:

  • Standard sequence learning models like temporal difference (TD)-learning require external rewards, conflicting with autonomous behavior's internally defined goals.
  • Autonomous agents need to learn from internally defined goals, not external rewards.
  • Classical conditioning involves learning causal relations to react to the earliest signal.

Purpose of the Study:

  • To present an isotropic algorithm for sequence order learning in a closed-loop system.
  • To enable autonomous behavior in a robot by replacing external rewards with an internal reflex loop.
  • To develop a system that learns to anticipate events, overcoming the delay inherent in reflex actions.

Main Methods:

  • Implemented an isotropic algorithm in a behaving robot, creating a closed-loop system where actions influence sensor inputs.

Related Experiment Videos

  • Replaced external rewards with a reflex loop that includes the environment.
  • Utilized a correlation-based learning rule involving the derivative of a linear neuron's output to modify synaptic weights.

    • Main Results:

    • The algorithm enabled the robot to learn causal relations and anticipate events, replacing delayed reflex actions with earlier anticipatory ones.
    • Synaptic weights stabilized once the reflex loop became silent, indicating successful learning.
    • The learning rule's weight change curve showed similarity to the temporal Hebb learning rule.

    Conclusions:

    • The developed isotropic algorithm facilitates autonomous behavior by enabling internal goal definition and anticipatory actions.
    • The system successfully learns sequence order, overcoming the inherent delay of reflex loops through learned anticipation.
    • This approach offers a novel method for sequence order learning in artificial systems, inspired by biological learning principles.