Effort-based learning and decision making.
Doctoral thesis, UCL (University College London).
In the literature on healthy humans, effort is poorly studied and an extension from an animal literature is just emerging. I tested an hypothesis that physical effort is a non-trivial aspect of motivated behaviour; it serves as a cost and interacts with outcomes. To do this I conducted four experimental studies and extended the range of costs to include pain. In my first experiment, I develop a functional Magnetic Resonance Imaging (fMRI) task assessing willingness to expend effort. I show that physical effort discounts value of actions and that activity in dorsal striatum is associated with effort of selected actions. In addition to influencing choice, effort may be influenced by affective outcomes. In my second experiment, I develop a behavioural instrumental learning task examining how reward and punishment influence learning about effortful response. I show that it is easier to expend effort to gain reward and to withdraw effort to avoid punishment, but not the other way around; in other words it is more difficult to expend effort to avoid punishment and to withdraw effort to gain reward. Results from reinforcement learning modelling account for this tendency in terms of a pavlovian influence on effort. On the one hand, outcome has an influence in effort while, on the other, effort may modulate neural signalling of action anticipation and outcome delivery. In my third experiment, I develop an fMRI cue-predictive instrumental task investigating brain responses for effort anticipation and outcome evaluation. I show that activity in anterior cingulate cortex and dorsal striatum is sensitive to anticipated effort and highlight an effort modulation on activity in ventromedial prefrontal cortex and ventral striatum associated with expected outcomes. Finally, I extend my investigation of costly behaviour in effort to pain by showing an influence of context effects in pain avoidance behaviour. In summary, within this thesis I demostrate that physical effort as a cost is non-trivial in that it i) discounts value, ii) is sensitive to pavlovian influences, iii) is neurally anticipated and iv) modulates outcome signalling. I show the viability of various experimental paradigms to assess costly behaviours driven by effort and extend this endeavour by studying cost-driven pain avoidance. These experiments forge new research directions for understanding action and decision making as well as show promise for testing aberrant populations that often present with pathology that may reflect under- and over-motivated actions (e.g., apathy and perseveration).
|Title:||Effort-based learning and decision making|
|Open access status:||An open access version is available from UCL Discovery|
|UCL classification:||UCL > School of Life and Medical Sciences > Faculty of Brain Sciences > Institute of Neurology|
Archive Staff Only