Memory and mental time travel in humans and social robots

T J Prescott, Daniel Camillieri, Uriel Martinez Hernandez, Andreas Damianou, Neil Lawrence

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

From neuroscience, brain imaging, and the psychology of memory we are beginning to assemblean integrated theory of the brain sub-systems and pathways that allow the compression, storageand reconstruction of memories for past events and their use in contextualizing the present andreasoning about the future—mental time travel (MTT). Using computational models, embeddedin humanoid robots, we are seeking to test the sufficiency of this theoretical account and toevaluate the usefulness of brain-inspired memory systems for social robots. In this contribution,we describe the use of machine learning techniques—Gaussian process latent variable models—to build a multimodal memory system for the iCub humanoid robot and summarise results of thedeployment of this system for human-robot interaction. We also outline the further steps requiredto create a more complete robotic implementation of human-like autobiographical memory andMTT. We propose that generative memory models, such as those that form the core of our robotmemory system, can provide a solution to the symbol grounding problem in embodied artificialintelligence.
Original languageEnglish
Number of pages27
JournalPhilosophical Transactions of the Royal Society B: Biological Sciences
Volume374
Issue number1771
Early online date11 Mar 2019
DOIs
Publication statusPublished - 29 Apr 2019

Keywords

  • autobiographical memory
  • Gaussian process
  • latent variable space
  • mental time travel
  • symbol grounding

ASJC Scopus subject areas

  • Biochemistry, Genetics and Molecular Biology(all)
  • Agricultural and Biological Sciences(all)

Cite this

Memory and mental time travel in humans and social robots. / Prescott, T J; Camillieri, Daniel; Martinez Hernandez, Uriel; Damianou, Andreas; Lawrence, Neil.

In: Philosophical Transactions of the Royal Society B: Biological Sciences, Vol. 374, No. 1771, 29.04.2019.

Research output: Contribution to journalArticle

Prescott, T J ; Camillieri, Daniel ; Martinez Hernandez, Uriel ; Damianou, Andreas ; Lawrence, Neil. / Memory and mental time travel in humans and social robots. In: Philosophical Transactions of the Royal Society B: Biological Sciences. 2019 ; Vol. 374, No. 1771.
@article{6a8c27db7b2f4fcea17847c0a40c76f2,
title = "Memory and mental time travel in humans and social robots",
abstract = "From neuroscience, brain imaging, and the psychology of memory we are beginning to assemblean integrated theory of the brain sub-systems and pathways that allow the compression, storageand reconstruction of memories for past events and their use in contextualizing the present andreasoning about the future—mental time travel (MTT). Using computational models, embeddedin humanoid robots, we are seeking to test the sufficiency of this theoretical account and toevaluate the usefulness of brain-inspired memory systems for social robots. In this contribution,we describe the use of machine learning techniques—Gaussian process latent variable models—to build a multimodal memory system for the iCub humanoid robot and summarise results of thedeployment of this system for human-robot interaction. We also outline the further steps requiredto create a more complete robotic implementation of human-like autobiographical memory andMTT. We propose that generative memory models, such as those that form the core of our robotmemory system, can provide a solution to the symbol grounding problem in embodied artificialintelligence.",
keywords = "autobiographical memory, Gaussian process, latent variable space, mental time travel, symbol grounding",
author = "Prescott, {T J} and Daniel Camillieri and {Martinez Hernandez}, Uriel and Andreas Damianou and Neil Lawrence",
year = "2019",
month = "4",
day = "29",
doi = "10.1098/rstb.2018.0025",
language = "English",
volume = "374",
journal = "Philosophical Transactions of the Royal Society B: Biological Sciences",
issn = "0264-3839",
number = "1771",

}

TY - JOUR

T1 - Memory and mental time travel in humans and social robots

AU - Prescott, T J

AU - Camillieri, Daniel

AU - Martinez Hernandez, Uriel

AU - Damianou, Andreas

AU - Lawrence, Neil

PY - 2019/4/29

Y1 - 2019/4/29

N2 - From neuroscience, brain imaging, and the psychology of memory we are beginning to assemblean integrated theory of the brain sub-systems and pathways that allow the compression, storageand reconstruction of memories for past events and their use in contextualizing the present andreasoning about the future—mental time travel (MTT). Using computational models, embeddedin humanoid robots, we are seeking to test the sufficiency of this theoretical account and toevaluate the usefulness of brain-inspired memory systems for social robots. In this contribution,we describe the use of machine learning techniques—Gaussian process latent variable models—to build a multimodal memory system for the iCub humanoid robot and summarise results of thedeployment of this system for human-robot interaction. We also outline the further steps requiredto create a more complete robotic implementation of human-like autobiographical memory andMTT. We propose that generative memory models, such as those that form the core of our robotmemory system, can provide a solution to the symbol grounding problem in embodied artificialintelligence.

AB - From neuroscience, brain imaging, and the psychology of memory we are beginning to assemblean integrated theory of the brain sub-systems and pathways that allow the compression, storageand reconstruction of memories for past events and their use in contextualizing the present andreasoning about the future—mental time travel (MTT). Using computational models, embeddedin humanoid robots, we are seeking to test the sufficiency of this theoretical account and toevaluate the usefulness of brain-inspired memory systems for social robots. In this contribution,we describe the use of machine learning techniques—Gaussian process latent variable models—to build a multimodal memory system for the iCub humanoid robot and summarise results of thedeployment of this system for human-robot interaction. We also outline the further steps requiredto create a more complete robotic implementation of human-like autobiographical memory andMTT. We propose that generative memory models, such as those that form the core of our robotmemory system, can provide a solution to the symbol grounding problem in embodied artificialintelligence.

KW - autobiographical memory

KW - Gaussian process

KW - latent variable space

KW - mental time travel

KW - symbol grounding

UR - http://www.scopus.com/inward/record.url?scp=85062712655&partnerID=8YFLogxK

U2 - 10.1098/rstb.2018.0025

DO - 10.1098/rstb.2018.0025

M3 - Article

VL - 374

JO - Philosophical Transactions of the Royal Society B: Biological Sciences

JF - Philosophical Transactions of the Royal Society B: Biological Sciences

SN - 0264-3839

IS - 1771

ER -