Barkley-Levenson & Galvan (2014) Neural representation of expected value in the adolescent brain

• the aim was

  to investigate whether adolescents attach more value to rewards than adults do. to identify neural development of expected value (EV) in the brains of adolescents
• research method

  quasi-experiment in lab conditions, using independent measures design
• sample

  19 healthy, right-handed adults (8 male, 11 female), aged 25-30 years, mean age 27.9 years

  22 healthy, right-handed adolescents (11 male, 11 female, aged 13-17 years, mean age 15.6 years

  recruited from poster and internet advertisements in America

  all reported no prior diagnosis of psychiatric or neurologic illness or development delays
• IV

  age of participant (adult/adolescent)
• DVs

  differences in 1. neural activation and 2. behavioural responses to gambling task
• procedure

  all participants gave informed consent (parental consent of under 18s also obtained). participants initially attended the laboratory for 'an intake session' for neuro-imaging (to be acclimatised to the fMRI through mock scan). in this session researchers collected data on participants' primary source of income and amount of spending money per month: the mean for adolescents was $52.50; for adults it was $467 (done to investigate differences between the two groups was due to the novelty of the reward rather than developmental differences - ie adolescents might have less experience of spending money, and value it more highly than adults). all were given $20 to use as 'playing' money during the fMRI task, once intake was complete. they were also informed that there was an opportunity to win up to $20 more in addition to their playing money by gambling, but that there was a possibility that they would lose it during the gambling fMRI task. one week later participants returned for the fMRI session. for each trial they had to decide whether they would be willing to gamble for their money. they were told one of the trials they accepted would be selected and played at the end of the scan for real money, and the outcome would be added/subtracted from their overall payment of $20. this was done to encourage participants to risk-take as they would normally. In actuality, all participants were assigned a payment of $5 and $10 to ensure no participants had to return money.
• procedure pt 2

  participants were trained on how to use the computerised gambling programme prior to testing. in task, during scan participants were presented with a series of gambles across 144 trials (per person). on screen they saw 'spinner' with a 50% probability of gaining the amount shown on one side of a "spinner" and a 50% probability of losing the amount shown on the other side. there were a range of profit values between +$5 and +$20 and loss amounts between−$5 and −$20. within these trials there were 24 win only trials and 24 loss-only trials, the rest were 50/50 chance. the side of the spinner in which the win and loss appeared and the order of the stimuli was counterbalanced across participants.
• results - neural activation

  *the fMRI data showed more activation of the VS in adolescents as the EV increased compared to adults (ie when EV of gamble was greater, VS of adolescents became much more active than it did in adults). this was the case even when groups were matched on income and overall acceptability of gambles, suggesting that adolescents do not have a special preference for money.

  *even when gambling behaviour was similar across both groups, VS activation was still greater in adolescents - could be explained by the theory that adolescents are less skilled than adults at calculating the EV of gambles (but behavioural data didn't suggest that adolescents were any less adept at discriminating advantageous from disadvantageous gambles)
• results - behavioural differences

  *in no risk trials (eg gain-only or loss-only) both groups made similar judgements - suggests adolescents behave similarly to adults when there's no risk involved

  *in both groups, increased EV also increased likelihood of accepting a gamble, although it had a greater influence on response of adolescents than adults (this 'hyperactive' reward sensitivity was found to precede more advantageous risk-taking in adolescents, suggesting they were focused more on the higher amount they could win and less concerned with the relatively smaller amount they could lose)

  *all trials with EV of zero were accepted significantly more than trials with negative EV

  *both adults and adolescents virtually never accepted gambles with negative EV (ie more to lose, less to win) - shows that adolescents were equally capable as adults in avoiding disadvantageous choices

  *amount of disposable income didn't have an effect and there were no differences in reaction times between adolescents and adults
• conclusions

  1. adolescents place greater value on rewards than adults do

  2. neural representations of value in adolescents are linked to increased risk-taking behaviour

  3. adolescents behave similarly to adults when there's no risk involved (in gambling)
• expected value

  sum of all of the possible outcomes of a particular choice multiplied by their probabilities. a person uses this to judge whether risks involved in a certain course of action are worth taking in order to gain a reward