Abstract
We introduce a biologically motivated, formal framework or “ontology” for dealing with many aspects of action discovery which we argue is an example of intrinsically motivated behaviour (as such, this chapter is a companion to that by Redgrave et al. in this volume). We argue that action discovery requires an interplay between separate internal forward models of prediction and inverse models mapping outcomes to actions. The process of learning actions is driven by transient changes in the animal’s policy (repetition bias) which is, in turn, a result of unpredicted, phasic sensory information (“surprise”). The notion of salience as value is introduced and broken down into contributions from novelty (or surprise), immediate reward acquisition, or general task/goal attainment. Many other aspects of biological action discovery emerge naturally in our framework which aims to guide future modelling efforts in this domain.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsNotes
- 1.
We use the term “vector”, but in the sense adopted in computer science to mean a 1D-array or n-tuple; there is no implication that these n-tuples form a true vector space. Indeed, if we use only positive-valued components (a natural choice to indicate presence of a feature), the space does not have an additive inverse.
- 2.
We use the normal convention that y denotes a function and y(t) its value at time t.
- 3.
Throughout this chapter, we use the expression “A is a subspace of B” (or A ⊂ B) to mean that A is defined over a subset of the features in B. Also, note that B is supposed to be upper case Greek β in keeping with the notation that Greek and Roman symbols refer to the external world and neural representations, respectively.
- 4.
N S ∖ M is defined over the features in N S which are not contained in N M.
- 5.
Other notions of action efficiency/optimality could have been used (e.g. minimal energy expenditure), but temporal optimality and action simplicity seem most appropriate in the context of action discovery.
- 6.
This is equivalent to xH(x) where H(x) is the Heaviside function, but the notation in the text is more expedient here.
- 7.
Some researchers call this a competence model 7.
- 8.
The striatum is the main input nucleus of the basal ganglia.
- 9.
Additionally, this may be supported by some kind of eligibility trace associated with s (a) which dopamine acts upon (Gurney et al. 2009a).
References
Allport, A., Sanders, H., Heuer, A.: Selection for action: Some behavioural and neurophysiological considerations of attention and action. In: Perspectives on Perception and Action. Lawrence Erlbaum Associates Inc., Hillsdale (1987)
Baldi, P., Itti, L.: Of bits and wows: A bayesian theory of surprise with applications to attention. Neural Netw. 23(5), 649–666 (2010)
Balleine, B.W., Dickinson, A.: Goal-directed instrumental action: Contingency and incentive learning and their cortical substrates. Neuropharmacology 37(4–5), 407–419 (1998)
Barto, A., Singh, S., Chentanez, N.: Intrinsically motivated reinforcement learning. In: 18th Annual Conference on Neural Information Processing Systems (NIPS). Vancouver (2004)
Cisek, P.: Cortical mechanisms of action selection: The affordance competition hypothesis. Philos.Trans. R. Soc. Lond B Biol. Sci. 362(1485), 1585–1599 (2007)
Cisek, P., Kalaska, J.: Neural mechanisms for interacting with a world full of action choices. Annu. Rev. Neurosci. 33, 269–298 (2010)
Comoli, E., Coizet, V., Boyes, J., Bolam, J., Canteras, N., Quirk, R., Overton, P, Redgrave, P.: A direct projection from superior colliculus to substantia nigra for detecting salient visual events. Nat. Neurosci. 6(9), 974–980 (2003)
Connor, C.E., Egeth, H.E., Yantis, S.: Visual attention: Bottom-Up versus Top-Down. Curr. Biol. 14(19), R850–R852 (2004)
Cope, A., Chambers, J., Gurney, K.: Object-based biasing for attentional control of gaze: A comparison of biologically plausible mechanisms. BMC Neurosci. 10(Suppl. 1), P19 (2009)
Dommett, E., Coizet, V., Blaha, C., Martindale, J., Lefebvre, V., Walton, N., Mayhew, J., Overton, P., Redgrave, P.: How visual stimuli activate dopaminergic neurons at short latency. Science 307(5714), 1476–1479 (2005)
Fiorillo, C.D., Tobler, P.N., Schultz, W.: Discrete coding of reward probability and uncertainty by dopamine neurons. Science 299(5614), 1898 (2003)
Friston, K.: A theory of cortical responses. Philos. Trans. R. Soc. B Biol. Sci. 360(1456), 815–836 (2005)
Gene Ontology Consortium: Creating the gene ontology resource: Design and implementation. Genome Res. 11(8), 1425–1433 (2001)
Gruber, T.: A translation approach to portable ontology specification. http://www-ksl.stanford.edu/kst/what-is-an-ontology.html (1992)
Gurney, K., Humphries, M., Redgrave, P.: Cortico-striatal plasticity for action-outcome learning using spike timing dependent eligibility. BMC Neurosci. 10(Suppl. 1), P135 (2009a)
Gurney, K., Hussain, A., Chambers, J., Abdullah, R.: Controlled and automatic processing in animals and machines with application to autonomous vehicle control. In: Controlled and Automatic Processing in Animals and Machines with Application to Autonomous Vehicle Control, Lecture Notes in Computer Science, vol. 5768, pp. 198–207. Springer, Berlin (2009b)
Ikeda, T., Hikosaka, O.: Reward-dependent gain and bias of visual responses in primate superior colliculus. Neuron 39(4), 693–700 (2003)
Körding, K.P., Wolpert, D.M.: Bayesian decision theory in sensorimotor control. Trends Cogn. Sci. 10(7), 319–326 (2006)
Marr, D., Poggio, T.: From understanding computation to understanding neural circuitry. Technical report, MIT AI Laboratory (1976)
Matsumoto, M., Hikosaka, O.: Lateral habenula as a source of negative reward signals in dopamine neurons. Nature 447(7148), 1111–1115 (2007)
Oudeyer, P., Kaplan, F.: What is intrinsic motivation? a typology of computational approaches. Front. Neurorobot. 1, 6 (2007). PMID 18958277
Poggio, T., Koch, C.: Ill-posed problems in early vision: From computational theory to analogue networks. Proc. R. Soc. Lond. B. Biol. Sci. 226(1244), 303 (1985)
Ranganath, C., Rainer, G.: Neural mechanisms for detecting and remembering novel events. Nat. Rev. Neurosci. 4(3), 193–202 (2003)
Redgrave, P., Gurney, K.: The short-latency dopamine signal: A role in discovering novel actions? Nat. Rev. Neurosci. 7(12) (2006)
Redgrave, P., Gurney, K., Reynolds, J.: What is reinforced by phasic dopamine signals? Brain Res. Rev. 58(2), 322–339 (2008)
Redgrave, P., Gurney, K., Stafford, T., Thirkettle, M., Lewis, J.: The role of the basal ganglia in discovering novel actions. In: Baldassarre, G., Mirolli, M. (eds.) Intrinsically Motivated Learning in Natural and Artificial Systems, pp. 129–149. Springer, Berlin (2012)
Redgrave, P., Prescott, T., Gurney, K.: The basal ganglia: A vertebrate solution to the selection problem? Neuroscience 89, 1009–1023 (1999)
Reynolds, J.N.J., Wickens, J.R.: Dopamine-dependent plasticity of corticostriatal synapses. Neural Netw. 15(4–6), 507–521 (2002)
Ryan, R.M., Deci, E.L.: Intrinsic and extrinsic motivations: Classic definitions and new directions. 1. Contemp. Educ. Psychol. 25(1), 54–67 (2000)
Schleidt, M., Kien, J.: Segmentation in behavior and what it can tell us about brain function. Hum. Nat. 8(1), 77–111 (1997)
Schmidhuber, J.: Driven by compression progress: A simple principle explains essential aspects of subjective beauty, novelty, surprise, interestingness, attention, curiosity, creativity, art, science, music, jokes. In: Anticipatory Behavior in Adaptive Learning Systems, pp. 48–76 (2009)
Schultz, W.: Dopamine signals for reward value and risk: Basic and recent data. Behav. Brain Funct. 6(1), 24 (2010)
Schultz, W., Dayan, P., Montague, P.: A neural substrate of prediction and reward. Science 275, 1593–1599 (1997)
Snyder, L.H., Batista, A.P., Andersen, R.A.: Coding of intention in the posterior parietal cortex. Nature 386(6621), 167–170 (1997)
Sokolov, E.N.: Higher nervous functions: The orienting reflex. Annu. Rev. Physiol. 25(1), 545–580 (1963)
Sutton, R., Barto, A.: Reinforcement Learning : An Introduction. MIT, Cambridge (1998)
Thompson, K.G., Bichot, N.P., Sato, T.R.: Frontal eye field activity before visual search errors reveals the integration of Bottom-Up and Top-Down salience. J. Neurophysiol. 93(1), 337–351 (2005)
Timberlake, W., Lucas, G.A.: The basis of superstitious behavior: Chance contingency, stimulus substitution, or appetitive behavior? J. Exp. Anal. Behav. 44(3), 279 (1985)
Tobler, P., Fiorillo, C., Schultz, W.: Adaptive coding of reward value by dopamine neurons. Science 307(5715), 1642 (2005)
Tolman, E.: Cognitive maps in rats and men. Psychol. Rev. 55(4), 189 (1948)
Wurtz, R.H., Albano, J.E.: Visual-motor function of the primate superior colliculus. Annu. Rev. Neurosci. 3(1), 189–226 (1980)
Yin, H.H., Knowlton, B.J.: The role of the basal ganglia in habit formation. Nat. Rev. Neurosci. 7(6), 464–476 (2006)
Acknowledgements
Written while the authors were in receipt of research funding from The Wellcome Trust, BBSRC and EPSRC.
This research has also received funds from the European Commission 7th Framework Programme (FP7/2007-2013), “Challenge 2 - Cognitive Systems, Interaction, Robotics”, Grant Agreement No. ICT-IP-231722, Project “IM-CLeVeR—Intrinsically Motivated Cumulative Learning Versatile Robots”. (NL was partially supported by EU Framework project EFAA (ICT-270490))
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2013 Springer-Verlag Berlin Heidelberg
About this chapter
Cite this chapter
Gurney, K., Lepora, N., Shah, A., Koene, A., Redgrave, P. (2013). Action Discovery and Intrinsic Motivation: A Biologically Constrained Formalisation. In: Baldassarre, G., Mirolli, M. (eds) Intrinsically Motivated Learning in Natural and Artificial Systems. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-32375-1_7
Download citation
DOI: https://doi.org/10.1007/978-3-642-32375-1_7
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-32374-4
Online ISBN: 978-3-642-32375-1
eBook Packages: Computer ScienceComputer Science (R0)