Unpicking influences on goal-directed behaviour across the lifespan: a computational modelling approach

Goal-directed behaviour [GDB] involves focusing on relevant information and ignoring
distractors based on specific goals. Nonetheless, goal-directed actions do not always lead to
the desired outcome. Two personality traits thought to influence GDB are impulsivity and
compulsivity, with the former initiating actions and the latter terminating them. Importantly,
impulsivity and compulsivity vary over short periods of time [1] and across the lifespan [2],
though the two are predominantly studied at trait level [3]. The link between real-life state
measures and physiological and neural markers remains unclear, and their effects on GDB is
unknown. Using an accelerated longitudinal design, the aim of this computational
developmental neuroscience project is to investigate the relationship between impulsivity,
compulsivity, and GDB across the lifespan [4].
This project will use a smartphone app to collect real-world, ecologically valid measures of
impulsivity, compulsivity and GDB during three developmental periods: late adolescence
[18-24yrs], middle adulthood [35-45yrs], and older adulthood [55-65yrs]. Longitudinal
measurements from a large sample will capture inter- and intra-individual differences and
data will be analysed using network analysis. Laboratory measures of impulsivity,
compulsivity and GDB will be collected in a subset of the sample. For example, we have
established a Pavlovian conditioning task with eye-tracking to measure goal-tracking
behaviour. Resting state fMRI data will be employed to assess how differences across the
lifespan in state and trait impulsivity and compulsivity relate to within- and between-restingstate
network architecture. Data will be analysed using computational modelling approaches.
The primary supervisor will lead on behavioural neuroscience, including laboratory measures
of impulsivity, compulsivity and GDB using eye-tracking. The secondary supervisor will lead
on computational modelling, longitudinal experience sampling and rs-fMRI. The two
supervisors provide distinct theoretical knowledge for this interdisciplinary project, with the
primary supervisor providing input relating behavioural neuroscience and the secondary
supervisor concerning computational neuroscience.