TY - JOUR
T1 - Uncovering the ‘state’
T2 - Tracing the hidden state representations that structure learning and decision-making
AU - Langdon, Angela J.
AU - Song, Mingyu
AU - Niv, Yael
PY - 2019/10
Y1 - 2019/10
N2 - We review the abstract concept of a ‘state’ – an internal representation posited by reinforcement learning theories to be used by an agent, whether animal, human or artificial, to summarize the features of the external and internal environment that are relevant for future behavior on a particular task. Armed with this summary representation, an agent can make decisions and perform actions to interact effectively with the world. Here, we review recent findings from the neurobiological and behavioral literature to ask: ‘what is a state?’ with respect to the internal representations that organize learning and decision making across a range of tasks. We find that state representations include information beyond a straightforward summary of the immediate cues in the environment, providing timing or contextual information from the recent or more distant past, which allows these additional factors to influence decision making and other goal-directed behaviors in complex and perhaps unexpected ways.
AB - We review the abstract concept of a ‘state’ – an internal representation posited by reinforcement learning theories to be used by an agent, whether animal, human or artificial, to summarize the features of the external and internal environment that are relevant for future behavior on a particular task. Armed with this summary representation, an agent can make decisions and perform actions to interact effectively with the world. Here, we review recent findings from the neurobiological and behavioral literature to ask: ‘what is a state?’ with respect to the internal representations that organize learning and decision making across a range of tasks. We find that state representations include information beyond a straightforward summary of the immediate cues in the environment, providing timing or contextual information from the recent or more distant past, which allows these additional factors to influence decision making and other goal-directed behaviors in complex and perhaps unexpected ways.
KW - Decision making
KW - Dopamine
KW - Learning
KW - Reward
KW - Timing
UR - http://www.scopus.com/inward/record.url?scp=85070717716&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85070717716&partnerID=8YFLogxK
U2 - 10.1016/j.beproc.2019.103891
DO - 10.1016/j.beproc.2019.103891
M3 - Review article
C2 - 31381985
AN - SCOPUS:85070717716
VL - 167
JO - Behavioural Processes
JF - Behavioural Processes
SN - 0376-6357
M1 - 103891
ER -