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

Distributed representations accelerate evolution of adaptive behaviours.

James V Stone1

  • 1Psychology Department, Sheffield University, Sheffield, United Kingdom. j.v.stone@shef.ac.uk

Plos Computational Biology
|August 7, 2007
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

How self-organization can guide evolution.

Royal Society open science·2016
Same author

A Self-Organising Model of Thermoregulatory Huddling.

PLoS computational biology·2015
Same author

Using reaction times and binary responses to estimate psychophysical performance: an information theoretic analysis.

Frontiers in neuroscience·2014
Same author

Marr: an appreciation.

Perception·2013
Same author

Footprints sticking out of the sand. Part 2: children's Bayesian priors for shape and lighting direction.

Perception·2011
Same author

Footprints sticking out of the sand. Part 1: Children's perception of naturalistic and embossed symbol stimuli.

Perception·2010
Same journal

Detection, communication, and individual identification with deep audio embeddings: A case study with North Atlantic right whales.

PLoS computational biology·2026
Same journal

Exploring the structural lexicon of the Proteome via Metric Geometry.

PLoS computational biology·2026
Same journal

Linking retinal sampling in neural encoding models to temporal profiles of visual processing in humans.

PLoS computational biology·2026
Same journal

CAdir: Joint clustering of cells and genes for single-cell transcriptomics with visualization-driven cluster quality assessment.

PLoS computational biology·2026
Same journal

Systematic design of auxotrophic strains and media conditions to probe metabolic functions in E. coli.

PLoS computational biology·2026
Same journal

Neuronal excitability and parameter variability in the Hodgkin-Huxley model.

PLoS computational biology·2026
See all related articles

This study shows that "free-lunch" learning (FLL) in neural networks accelerates skill acquisition and evolution. Partial skill learning automatically completes the skill, speeding up adaptive behavior development.

Area of Science:

  • Neuroscience
  • Computational Biology
  • Evolutionary Biology

Background:

  • Animals require extensive learning to develop innate abilities into complex skills.
  • Skills can be defined as interconnected sensory-motor associations.

Purpose of the Study:

  • To investigate the role of distributed representations in skill learning.
  • To demonstrate how partial skill learning can lead to automatic acquisition of the remaining skill components.
  • To analyze the impact of "free-lunch" learning (FLL) on skill evolution and adaptive behavior.

Main Methods:

  • Utilized linear neural network models to simulate skill acquisition.
  • Analyzed the effects of distributed representations on learning dynamics.
  • Compared learning outcomes in networks with and without FLL capabilities.

Related Experiment Videos

Main Results:

  • Within-lifetime learning of a partial skill can automatically induce learning of the remaining parts when skills are distributed.
  • Free-lunch learning (FLL) significantly accelerates skill evolution compared to non-FLL or non-learning networks.
  • FLL enhances the emergence of adaptive behaviors, both innate and learned, and speeds up the process of learned behaviors becoming innate.

Conclusions:

  • Distributed skill representations enable efficient "free-lunch" learning (FLL).
  • FLL is a key mechanism for accelerated skill evolution and adaptive behavior development in biological systems.
  • The findings provide insights into the evolutionary advantages of specific learning architectures.