boPro (Bowen's projects)

Code from my PhD in the Decision Neuroscience Lab. For all enquiries please contact [email protected], but I wouldn't expect much clarification.

The directory is divided into matlab and R code, with some overlaps in terms of the specific projects. Generally, stimulus display has been coded in matlab, and analysis is coded in R. Here is a rough summary of the projects:

TIMEDEC (7/2012 - 12/2015)

• n = 145 (120 with ECG recording)
• Attempted to relate duration reproduction and temporal discounting
• Found that ECG parameters independently related to temporal discounting and time perception

TIMEGUESS (pilot, 4/2013)

• n = 19
• Attempted to replicate magnitude/time, and oddball effects using temporal classification task
• Particpants presented with an ‘anchor’ (“About 4 seconds”), then stimulus (dots, circles or lines), after which they classify the stimulus as shorter/longer than the anchor
• Stimulus duration always the same as anchor?
• Unable to replicate oddball or magnitude effects, probably due to intervals used (supra-second)
• Final version is TG24.4

WOF (unfinished, 8/2013)

• Temporal bisection task where one stimulus codes the outcome of a gamble
• Intended to investigate the effects of reward valence on time perception

EDT (5/2013 - 7/2014)

• n = 19 (7 additional participants in pilot)
• Used primary reward (fruit juice) in an experiential discounting task
• Calibrated individual discount rates to bias choices in subsequent test
• Found that participants did not reliably “discount”, and that some appeared to follow a reward maximising strategy that accounted for the ITI
• There are a number of different versions of this paradigm, only the latest of which are in this repo

Blink Bandit (BB) (7/2013)

• The auditory bandit task Dan Bennett and I ran for the neuroeconomics grad subject
• Tried to show a relationship between spontaneous blink rate (tonic DA) and exploration/exploitation
• Promising, but unfinished

Juice space (3/2014 - 4/2014)

• n = 7 (4 sessions each)
• Used staircase algorithms (QUEST and Psi) to determine juice volume discrimination (75% detection)
• Found a weber fraction of ~0.5 for two different volumes (0.5 & 3 mL)
• Poor programming led to data being too big to store anywhere really

Juice value (5/2014)

• n = 5
• Acquired thirst, pleasantness and satiety ratings for different volumes of juice
• Found power relationship between volume and pleasantness

PIP (7/2014 - 5/2015)

• n = 75
• Used half-peak interval procedure with different rewards (juice, money, water)
• Found an effect of fruit juice on response timing
• Various versions of this correspond to the juice, money and water versions of the task (v3 is juice)

FIP (unfinished, 5/2015)

• Attempted to show time/juice effect on behavioural foraging choices
• Modelling showed only a marginal, non-measureable effect on choices

QT (6/2015 - 8/2015)

• n = 50
• Used waiting/persistence task with primary rewards
• Found that water condition waited longer than glucose/aspartame condition

PPP (12/2015 - 2/2016)

• n = 50
• Used PIP task with additional physiological measures (ECG, EOG, GSR)
• Attempt to show that physiological changes due to calorie intake mediate changes in time perception
• Marginally significant effect of maltodextrin, but no effect of aspartame
• Apparent correlation between HRV and the underestimation between task phases corresponding to liquid consumption